diff options
Diffstat (limited to 'lib/stdlib')
56 files changed, 6440 insertions, 1405 deletions
diff --git a/lib/stdlib/doc/src/Makefile b/lib/stdlib/doc/src/Makefile index 26602764a6..af27efa6c1 100644 --- a/lib/stdlib/doc/src/Makefile +++ b/lib/stdlib/doc/src/Makefile @@ -1,7 +1,7 @@ # # %CopyrightBegin% # -# Copyright Ericsson AB 1997-2016. All Rights Reserved. +# Copyright Ericsson AB 1997-2017. All Rights Reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. @@ -98,13 +98,15 @@ XML_REF3_FILES = \ sys.xml \ timer.xml \ unicode.xml \ + uri_string.xml \ win32reg.xml \ zip.xml XML_REF6_FILES = stdlib_app.xml -XML_PART_FILES = part.xml part_notes.xml part_notes_history.xml -XML_CHAPTER_FILES = io_protocol.xml unicode_usage.xml notes.xml notes_history.xml assert_hrl.xml +XML_PART_FILES = part.xml +XML_CHAPTER_FILES = introduction.xml io_protocol.xml unicode_usage.xml \ + notes.xml assert_hrl.xml BOOK_FILES = book.xml @@ -131,9 +133,9 @@ SPECS_FILES = $(XML_REF3_FILES:%.xml=$(SPECDIR)/specs_%.xml) TOP_SPECS_FILE = specs.xml # ---------------------------------------------------- -# FLAGS +# FLAGS # ---------------------------------------------------- -XML_FLAGS += +XML_FLAGS += SPECS_FLAGS = -I../../include -I../../../kernel/include @@ -150,7 +152,7 @@ html: $(HTML_REF_MAN_FILE) man: $(MAN3_FILES) $(MAN6_FILES) -debug opt: +debug opt: clean clean_docs: rm -rf $(HTMLDIR)/* @@ -158,7 +160,7 @@ clean clean_docs: rm -f $(MAN6DIR)/* rm -f $(TOP_PDF_FILE) $(TOP_PDF_FILE:%.pdf=%.fo) rm -f $(SPECDIR)/* - rm -f errs core *~ + rm -f errs core *~ $(SPECDIR)/specs_erl_id_trans.xml: escript $(SPECS_EXTRACTOR) $(SPECS_FLAGS) \ @@ -166,7 +168,7 @@ $(SPECDIR)/specs_erl_id_trans.xml: # ---------------------------------------------------- # Release Target -# ---------------------------------------------------- +# ---------------------------------------------------- include $(ERL_TOP)/make/otp_release_targets.mk release_docs_spec: docs diff --git a/lib/stdlib/doc/src/c.xml b/lib/stdlib/doc/src/c.xml index 7666699183..697e1715e7 100644 --- a/lib/stdlib/doc/src/c.xml +++ b/lib/stdlib/doc/src/c.xml @@ -94,6 +94,15 @@ </func> <func> + <name name="erlangrc" arity="1"/> + <fsummary>Load an erlang resource file.</fsummary> + <desc> + <p>Search <c>PathList</c> and load <c>.erlang</c> resource file if + found.</p> + </desc> + </func> + + <func> <name name="flush" arity="0"/> <fsummary>Flush any messages sent to the shell.</fsummary> <desc> diff --git a/lib/stdlib/doc/src/digraph.xml b/lib/stdlib/doc/src/digraph.xml index 5332d7aba5..db96beed6c 100644 --- a/lib/stdlib/doc/src/digraph.xml +++ b/lib/stdlib/doc/src/digraph.xml @@ -170,6 +170,10 @@ <p>If the edge would create a cycle in an <seealso marker="#acyclic_digraph">acyclic digraph</seealso>, <c>{error, {bad_edge, <anno>Path</anno>}}</c> is returned. + If <c><anno>G</anno></c> already has an edge with value + <c><anno>E</anno></c> connecting a different pair of vertices, + <c>{error, {bad_edge, [<anno>V1</anno>, <anno>V2</anno>]}}</c> + is returned. If either of <c><anno>V1</anno></c> or <c><anno>V2</anno></c> is not a vertex of digraph <c><anno>G</anno></c>, <c>{error, {bad_vertex, </c><anno>V</anno><c>}}</c> is diff --git a/lib/stdlib/doc/src/ets.xml b/lib/stdlib/doc/src/ets.xml index 576959b1c8..a0ec22c515 100644 --- a/lib/stdlib/doc/src/ets.xml +++ b/lib/stdlib/doc/src/ets.xml @@ -1961,7 +1961,7 @@ true</pre> The return value is a list of the new counter values from each update operation in the same order as in the operation list. If an empty list is specified, nothing is updated and an empty list is - returned. If the function fails, no updates is done.</p> + returned. If the function fails, no updates are done.</p> <p>The specified <c><anno>Key</anno></c> is used to identify the object by either <em>matching</em> the key of an object in a <c>set</c> table, or <em>compare equal</em> to the key of an object in an diff --git a/lib/stdlib/doc/src/fascicules.xml b/lib/stdlib/doc/src/fascicules.xml deleted file mode 100644 index 0ded9007e0..0000000000 --- a/lib/stdlib/doc/src/fascicules.xml +++ /dev/null @@ -1,18 +0,0 @@ -<?xml version="1.0" encoding="utf-8" ?> -<!DOCTYPE fascicules SYSTEM "fascicules.dtd"> - -<fascicules> - <fascicule file="part" href="part_frame.html" entry="no"> - STDLIB User's Guide - </fascicule> - <fascicule file="ref_man" href="ref_man_frame.html" entry="yes"> - Reference Manual - </fascicule> - <fascicule file="part_notes" href="part_notes_frame.html" entry="no"> - Release Notes - </fascicule> - <fascicule file="" href="../../../../doc/print.html" entry="no"> - Off-Print - </fascicule> -</fascicules> - diff --git a/lib/stdlib/doc/src/filelib.xml b/lib/stdlib/doc/src/filelib.xml index 1b69e84d31..5e631aac21 100644 --- a/lib/stdlib/doc/src/filelib.xml +++ b/lib/stdlib/doc/src/filelib.xml @@ -50,16 +50,16 @@ <p> Functionality in this module generally assumes valid input and does not necessarily fail on input that does not use a valid - encoding. You can validate the encoding of a filename using - <seealso marker="stdlib:filename#validate/1">filename:validate/1</seealso>. + encoding, but may instead very likely produce invalid output. </p> <p> File operations used to accept filenames containing null characters (integer value zero). This caused - the name to be truncated at the first null character. - Filenames containing null characters inside the filename + the name to be truncated and in some cases arguments + to primitive operations to be mixed up. Filenames + containing null characters inside the filename are now <em>rejected</em> and will cause primitive - file operations fail. + file operations to fail. </p> </note> <warning><p> diff --git a/lib/stdlib/doc/src/filename.xml b/lib/stdlib/doc/src/filename.xml index b6028fc066..d2608ad542 100644 --- a/lib/stdlib/doc/src/filename.xml +++ b/lib/stdlib/doc/src/filename.xml @@ -64,16 +64,16 @@ <p> Functionality in this module generally assumes valid input and does not necessarily fail on input that does not use a valid - encoding. You can validate the encoding of a filename using - <seealso marker="#validate/1">filename:validate/1</seealso>. + encoding, but may instead very likely produce invalid output. </p> <p> File operations used to accept filenames containing null characters (integer value zero). This caused - the name to be truncated at the first null character. - Filenames containing null characters inside the filename + the name to be truncated and in some cases arguments + to primitive operations to be mixed up. Filenames + containing null characters inside the filename are now <em>rejected</em> and will cause primitive - file operations fail. + file operations to fail. </p> </note> <warning><p> @@ -583,54 +583,6 @@ unsafe</pre> </desc> </func> - <func> - <name name="validate" arity="1"/> - <fsummary>Validate encoding of filename</fsummary> - <desc> - <p> - Validates filename encoding. Returns <c>true</c> if - <c><anno>FileName</anno></c> has a valid encoding; - otherwise, returns <c>false</c>. - </p> - <taglist> - <tag>Ordinary Filename</tag> - <item> - <p> - Type: <c><anno>FileName</anno> = </c><seealso marker="kernel:file#type-name"><c>file:name()</c></seealso> - </p> - <p> - Validates encoding against the - <seealso marker="kernel:file#native_name_encoding/0">native file - name encoding</seealso>, and the - capabilities of the operating system used. - Regardless of configuration and OS, null - characters (integer value zero) will be - rejected by the validation (even when only - present at the end of the filename). - </p> - </item> - <tag><seealso marker="unicode_usage#notes-about-raw-filenames">Raw - Filename</seealso></tag> - <item> - <p> - Type: <c><anno>FileName</anno> = binary()</c> - </p> - <p> - The encoding will not be interpreted, but - null bytes (integer value zero) will be - rejected by the validation (even when only - present at the end of the filename). - </p> - </item> - </taglist> - <p> - For information on filename encoding see the documentation - of unicode filenames in - <seealso marker="stdlib:unicode_usage#unicode_file_names">STDLIB - Users Guide ➜ Using Unicode in Erlang ➜ Unicode Filenames</seealso>. - </p> - </desc> - </func> </funcs> </erlref> diff --git a/lib/stdlib/doc/src/gen_statem.xml b/lib/stdlib/doc/src/gen_statem.xml index 8de6ed754f..4a824f073e 100644 --- a/lib/stdlib/doc/src/gen_statem.xml +++ b/lib/stdlib/doc/src/gen_statem.xml @@ -1329,7 +1329,7 @@ handle_event(_, _, State, Data) -> <c><anno>T</anno></c> is the time-out time. <c>{clean_timeout,<anno>T</anno>}</c> works like just <c>T</c> described in the note above - and uses a proxy process for <c>T < infinity</c>, + and uses a proxy process while <c>{dirty_timeout,<anno>T</anno>}</c> bypasses the proxy process which is more lightweight. </p> @@ -1339,8 +1339,12 @@ handle_event(_, _, State, Data) -> with <c>{dirty_timeout,<anno>T</anno>}</c> to avoid that the calling process dies when the call times out, you will have to be prepared to handle - a late reply. - So why not just let the calling process die? + a late reply. Note that there is an odd chance + to get a late reply even with + <c>{dirty_timeout,infinity}</c> or <c>infinity</c> + for example in the event of network problems. + So why not just let the calling process die + by not catching the exception? </p> </note> <p> diff --git a/lib/stdlib/doc/src/maps.xml b/lib/stdlib/doc/src/maps.xml index 8c7270816b..987d92989d 100644 --- a/lib/stdlib/doc/src/maps.xml +++ b/lib/stdlib/doc/src/maps.xml @@ -33,16 +33,31 @@ <p>This module contains functions for maps processing.</p> </description> + <datatypes> + <datatype> + <name name="iterator"/> + <desc> + <p>An iterator representing the key value associations in a map.</p> + <p>Created using <seealso marker="#iterator-1"><c>maps:iterator/1</c></seealso>.</p> + <p>Consumed by <seealso marker="#next-1"><c>maps:next/1</c></seealso>, + <seealso marker="#filter-2"><c>maps:filter/2</c></seealso>, + <seealso marker="#fold-3"><c>maps:fold/3</c></seealso> and + <seealso marker="#map-2"><c>maps:map/2</c></seealso>.</p> + </desc> + </datatype> + </datatypes> + <funcs> <func> <name name="filter" arity="2"/> <fsummary>Select pairs that satisfy a predicate.</fsummary> <desc> - <p>Returns a map <c><anno>Map2</anno></c> for which predicate - <c><anno>Pred</anno></c> holds true in <c><anno>Map1</anno></c>.</p> + <p>Returns a map <c><anno>Map</anno></c> for which predicate + <c><anno>Pred</anno></c> holds true in <c><anno>MapOrIter</anno></c>.</p> <p>The call fails with a <c>{badmap,Map}</c> exception if - <c><anno>Map1</anno></c> is not a map, or with <c>badarg</c> if - <c><anno>Pred</anno></c> is not a function of arity 2.</p> + <c><anno>MapOrIter</anno></c> is not a map or valid iterator, + or with <c>badarg</c> if <c><anno>Pred</anno></c> is not a + function of arity 2.</p> <p><em>Example:</em></p> <code type="none"> > M = #{a => 2, b => 3, c=> 4, "a" => 1, "b" => 2, "c" => 4}, @@ -76,12 +91,16 @@ <fsummary></fsummary> <desc> <p>Calls <c>F(K, V, AccIn)</c> for every <c><anno>K</anno></c> to value - <c><anno>V</anno></c> association in <c><anno>Map</anno></c> in + <c><anno>V</anno></c> association in <c><anno>MapOrIter</anno></c> in any order. Function <c>fun F/3</c> must return a new accumulator, which is passed to the next successive call. This function returns the final value of the accumulator. The initial accumulator value <c><anno>Init</anno></c> is returned if the map is empty.</p> + <p>The call fails with a <c>{badmap,Map}</c> exception if + <c><anno>MapOrIter</anno></c> is not a map or valid iterator, + or with <c>badarg</c> if <c><anno>Fun</anno></c> is not a + function of arity 3.</p> <p><em>Example:</em></p> <code type="none"> > Fun = fun(K,V,AccIn) when is_list(K) -> AccIn + V end, @@ -169,6 +188,32 @@ false</code> </func> <func> + <name name="iterator" arity="1"/> + <fsummary>Create a map iterator.</fsummary> + <desc> + <p>Returns a map iterator <c><anno>Iterator</anno></c> that can + be used by <seealso marker="#next-1"><c>maps:next/1</c></seealso> + to traverse the key-value associations in a map. When iterating + over a map, the memory usage is guaranteed to be bounded no matter + the size of the map.</p> + <p>The call fails with a <c>{badmap,Map}</c> exception if + <c><anno>Map</anno></c> is not a map.</p> + <p><em>Example:</em></p> + <code type="none"> +> M = #{ a => 1, b => 2 }. +#{a => 1,b => 2} +> I = maps:iterator(M). +[{a,1},{b,2}] +> {K1, V1, I2} = maps:next(I). +{a,1,[{b,2}]} +> {K2, V2, I3} = maps:next(I2). +{b,2,[]} +> maps:next(I3). +none</code> + </desc> + </func> + + <func> <name name="keys" arity="1"/> <fsummary></fsummary> <desc> @@ -188,12 +233,16 @@ false</code> <name name="map" arity="2"/> <fsummary></fsummary> <desc> - <p>Produces a new map <c><anno>Map2</anno></c> by calling function + <p>Produces a new map <c><anno>Map</anno></c> by calling function <c>fun F(K, V1)</c> for every <c><anno>K</anno></c> to value - <c><anno>V1</anno></c> association in <c><anno>Map1</anno></c> in + <c><anno>V1</anno></c> association in <c><anno>MapOrIter</anno></c> in any order. Function <c>fun F/2</c> must return value <c><anno>V2</anno></c> to be associated with key <c><anno>K</anno></c> - for the new map <c><anno>Map2</anno></c>.</p> + for the new map <c><anno>Map</anno></c>.</p> + <p>The call fails with a <c>{badmap,Map}</c> exception if + <c><anno>MapOrIter</anno></c> is not a map or valid iterator, + or with <c>badarg</c> if <c><anno>Fun</anno></c> is not a + function of arity 2.</p> <p><em>Example:</em></p> <code type="none"> > Fun = fun(K,V1) when is_list(K) -> V1*2 end, @@ -234,6 +283,35 @@ false</code> </func> <func> + <name name="next" arity="1"/> + <fsummary>Get the next key and value from an iterator.</fsummary> + <desc> + <p>Returns the next key-value association in + <c><anno>Iterator</anno></c> and a new iterator for the + remaining associations in the iterator. + </p> + <p> + If there are no more associations in the iterator, + <c>none</c> is returned. + </p> + <p><em>Example:</em></p> + <code type="none"> +> Map = #{a => 1, b => 2, c => 3}. +#{a => 1,b => 2,c => 3} +> Iter = maps:iterator(Map). +[{a,1},{b,2},{c,3}] +> {_, _, Iter1} = maps:next(Iter). +{a,1,[{b,2},{c,3}]} +> {_, _, Iter2} = maps:next(Iter1). +{b,2,[{c,3}]} +> {_, _, Iter3} = maps:next(Iter2). +{c,3,[]} +> maps:next(Iter3). +none</code> + </desc> + </func> + + <func> <name name="put" arity="3"/> <fsummary></fsummary> <desc> diff --git a/lib/stdlib/doc/src/notes.xml b/lib/stdlib/doc/src/notes.xml index d396f1de8f..b61e5b9b9e 100644 --- a/lib/stdlib/doc/src/notes.xml +++ b/lib/stdlib/doc/src/notes.xml @@ -31,6 +31,49 @@ </header> <p>This document describes the changes made to the STDLIB application.</p> +<section><title>STDLIB 3.4.3</title> + + <section><title>Fixed Bugs and Malfunctions</title> + <list> + <item> + <p> Make <c>ets:i/1</c> exit cleaner when ^D is input + while browsing a table. Only the old Erlang shell is + affected (<c>erl(1)</c> flag <c>-oldshell</c>). </p> + <p> + Own Id: OTP-14663</p> + </item> + <item> + <p> + Fixed handling of windows UNC paths in module + <c>filename</c>.</p> + <p> + Own Id: OTP-14693</p> + </item> + </list> + </section> + + + <section><title>Improvements and New Features</title> + <list> + <item> + <p> + Improve performance of the new string functionality when + handling ASCII characters.</p> + <p> + Own Id: OTP-14670</p> + </item> + <item> + <p> + Added a clarification to the documentation of + <c>unicode:characters_to_list/2</c>.</p> + <p> + Own Id: OTP-14798</p> + </item> + </list> + </section> + +</section> + <section><title>STDLIB 3.4.2</title> <section><title>Fixed Bugs and Malfunctions</title> diff --git a/lib/stdlib/doc/src/part_notes.xml b/lib/stdlib/doc/src/part_notes.xml deleted file mode 100644 index 461de749dd..0000000000 --- a/lib/stdlib/doc/src/part_notes.xml +++ /dev/null @@ -1,39 +0,0 @@ -<?xml version="1.0" encoding="utf-8" ?> -<!DOCTYPE part SYSTEM "part.dtd"> - -<part xmlns:xi="http://www.w3.org/2001/XInclude"> - <header> - <copyright> - <year>2004</year><year>2016</year> - <holder>Ericsson AB. All Rights Reserved.</holder> - </copyright> - <legalnotice> - Licensed under the Apache License, Version 2.0 (the "License"); - you may not use this file except in compliance with the License. - You may obtain a copy of the License at - - http://www.apache.org/licenses/LICENSE-2.0 - - Unless required by applicable law or agreed to in writing, software - distributed under the License is distributed on an "AS IS" BASIS, - WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - See the License for the specific language governing permissions and - limitations under the License. - - </legalnotice> - - <title>STDLIB Release Notes</title> - <prepared></prepared> - <docno></docno> - <date></date> - <rev></rev> - </header> - <description> - <p>The Standard Erlang Libraries application, <em>STDLIB</em>, - contains modules for manipulating lists, strings and files etc.</p> - <p>For information about older versions, see - <url href="part_notes_history_frame.html">Release Notes History</url>.</p> - </description> - <xi:include href="notes.xml"/> -</part> - diff --git a/lib/stdlib/doc/src/part_notes_history.xml b/lib/stdlib/doc/src/part_notes_history.xml deleted file mode 100644 index 8fd048a41e..0000000000 --- a/lib/stdlib/doc/src/part_notes_history.xml +++ /dev/null @@ -1,39 +0,0 @@ -<?xml version="1.0" encoding="utf-8" ?> -<!DOCTYPE part SYSTEM "part.dtd"> - -<part> - <header> - <copyright> - <year>2006</year> - <year>2016</year> - <holder>Ericsson AB, All Rights Reserved</holder> - </copyright> - <legalnotice> - Licensed under the Apache License, Version 2.0 (the "License"); - you may not use this file except in compliance with the License. - You may obtain a copy of the License at - - http://www.apache.org/licenses/LICENSE-2.0 - - Unless required by applicable law or agreed to in writing, software - distributed under the License is distributed on an "AS IS" BASIS, - WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - See the License for the specific language governing permissions and - limitations under the License. - - The Initial Developer of the Original Code is Ericsson AB. - </legalnotice> - - <title>STDLIB Release Notes History</title> - <prepared></prepared> - <docno></docno> - <date></date> - <rev></rev> - </header> - <description> - <p>The Standard Erlang Libraries application, <em>STDLIB</em>, - contains modules for manipulating lists, strings and files etc.</p> - </description> - <include file="notes_history"></include> -</part> - diff --git a/lib/stdlib/doc/src/ref_man.xml b/lib/stdlib/doc/src/ref_man.xml index 878a3babc5..68bfddbc71 100644 --- a/lib/stdlib/doc/src/ref_man.xml +++ b/lib/stdlib/doc/src/ref_man.xml @@ -93,6 +93,7 @@ <xi:include href="sys.xml"/> <xi:include href="timer.xml"/> <xi:include href="unicode.xml"/> + <xi:include href="uri_string.xml"/> <xi:include href="win32reg.xml"/> <xi:include href="zip.xml"/> </application> diff --git a/lib/stdlib/doc/src/specs.xml b/lib/stdlib/doc/src/specs.xml index 45b207b13d..d559adf9b6 100644 --- a/lib/stdlib/doc/src/specs.xml +++ b/lib/stdlib/doc/src/specs.xml @@ -60,6 +60,7 @@ <xi:include href="../specs/specs_sys.xml"/> <xi:include href="../specs/specs_timer.xml"/> <xi:include href="../specs/specs_unicode.xml"/> + <xi:include href="../specs/specs_uri_string.xml"/> <xi:include href="../specs/specs_win32reg.xml"/> <xi:include href="../specs/specs_zip.xml"/> </specs> diff --git a/lib/stdlib/doc/src/unicode.xml b/lib/stdlib/doc/src/unicode.xml index e86f45431f..d822aca89c 100644 --- a/lib/stdlib/doc/src/unicode.xml +++ b/lib/stdlib/doc/src/unicode.xml @@ -239,8 +239,13 @@ <c><anno>InEncoding</anno></c>.</p> </item> </list> - <p>Only when <c><anno>InEncoding</anno></c> is one of the UTF - encodings, integers in the list are allowed to be > 255.</p> + <p> + Note that integers in the list always represent code points + regardless of <c><anno>InEncoding</anno></c> passed. If + <c><anno>InEncoding</anno> latin1</c> is passed, only code + points < 256 are allowed; otherwise, all valid unicode code + points are allowed. + </p> <p>If <c><anno>InEncoding</anno></c> is <c>latin1</c>, parameter <c><anno>Data</anno></c> corresponds to the <c>iodata()</c> type, but for <c>unicode</c>, parameter <c><anno>Data</anno></c> can diff --git a/lib/stdlib/doc/src/unicode_usage.xml b/lib/stdlib/doc/src/unicode_usage.xml index ff1f864e22..789e063c12 100644 --- a/lib/stdlib/doc/src/unicode_usage.xml +++ b/lib/stdlib/doc/src/unicode_usage.xml @@ -857,6 +857,10 @@ Eshell V5.10.1 (abort with ^G) <section> <marker id="notes-about-raw-filenames"/> <title>Notes About Raw Filenames</title> + <note><p> + Note that raw filenames <em>not</em> necessarily are encoded the + same way as on the OS level. + </p></note> <p>Raw filenames were introduced together with Unicode filename support in ERTS 5.8.2 (Erlang/OTP R14B01). The reason "raw filenames" were introduced in the system was diff --git a/lib/stdlib/doc/src/uri_string.xml b/lib/stdlib/doc/src/uri_string.xml new file mode 100644 index 0000000000..21f470e763 --- /dev/null +++ b/lib/stdlib/doc/src/uri_string.xml @@ -0,0 +1,326 @@ +<?xml version="1.0" encoding="utf-8" ?> +<!DOCTYPE erlref SYSTEM "erlref.dtd"> + +<erlref> + <header> + <copyright> + <year>2017</year><year>2017</year> + <holder>Ericsson AB. All Rights Reserved.</holder> + </copyright> + <legalnotice> + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. + </legalnotice> + + <title>uri_string</title> + <prepared>Péter Dimitrov</prepared> + <docno>1</docno> + <date>2017-10-24</date> + <rev>A</rev> + </header> + <module>uri_string</module> + <modulesummary>URI processing functions.</modulesummary> + <description> + <p>This module contains functions for parsing and handling URIs + (<url href="https://www.ietf.org/rfc/rfc3986.txt">RFC 3986</url>) and + form-urlencoded query strings (<url href="https://www.w3.org/TR/html5/forms.html">HTML5</url>). + </p> + <p>A URI is an identifier consisting of a sequence of characters matching the syntax + rule named <em>URI</em> in <url href="https://www.ietf.org/rfc/rfc3986.txt">RFC 3986</url>. + </p> + <p> The generic URI syntax consists of a hierarchical sequence of components referred + to as the scheme, authority, path, query, and fragment:</p> + <pre> + URI = scheme ":" hier-part [ "?" query ] [ "#" fragment ] + hier-part = "//" authority path-abempty + / path-absolute + / path-rootless + / path-empty + scheme = ALPHA *( ALPHA / DIGIT / "+" / "-" / "." ) + authority = [ userinfo "@" ] host [ ":" port ] + userinfo = *( unreserved / pct-encoded / sub-delims / ":" ) + + reserved = gen-delims / sub-delims + gen-delims = ":" / "/" / "?" / "#" / "[" / "]" / "@" + sub-delims = "!" / "$" / "&" / "'" / "(" / ")" + / "*" / "+" / "," / ";" / "=" + + unreserved = ALPHA / DIGIT / "-" / "." / "_" / "~" + </pre><br></br> + <p>The interpretation of a URI depends only on the characters used and not on how those + characters are represented in a network protocol.</p> + <p>The functions implemented by this module cover the following use cases:</p> + <list type="bulleted"> + <item>Parsing URIs into its components and returing a map<br></br> + <seealso marker="#parse/1"><c>parse/1</c></seealso> + </item> + <item>Recomposing a map of URI components into a URI string<br></br> + <seealso marker="#recompose/1"><c>recompose/1</c></seealso> + </item> + <item>Changing inbound binary and percent-encoding of URIs<br></br> + <seealso marker="#transcode/2"><c>transcode/2</c></seealso> + </item> + <item>Transforming URIs into a normalized form<br></br> + <seealso marker="#normalize/1"><c>normalize/1</c></seealso> + </item> + <item>Composing form-urlencoded query strings from a list of key-value pairs<br></br> + <seealso marker="#compose_query/1"><c>compose_query/1</c></seealso><br></br> + <seealso marker="#compose_query/2"><c>compose_query/2</c></seealso> + </item> + <item>Dissecting form-urlencoded query strings into a list of key-value pairs<br></br> + <seealso marker="#dissect_query/1"><c>dissect_query/1</c></seealso> + </item> + </list> + <p>There are four different encodings present during the handling of URIs:</p> + <list type="bulleted"> + <item>Inbound binary encoding in binaries</item> + <item>Inbound percent-encoding in lists and binaries</item> + <item>Outbound binary encoding in binaries</item> + <item>Outbound percent-encoding in lists and binaries</item> + </list> + <p>Functions with <c>uri_string()</c> argument accept lists, binaries and + mixed lists (lists with binary elements) as input type. All of the functions but + <c>transcode/2</c> expects input as lists of unicode codepoints, UTF-8 encoded binaries + and UTF-8 percent-encoded URI parts ("%C3%B6" corresponds to the unicode character "ö").</p> + <p>Unless otherwise specified the return value type and encoding are the same as the input + type and encoding. That is, binary input returns binary output, list input returns a list + output but mixed input returns list output.</p> + <p>In case of lists there is only percent-encoding. In binaries, however, both binary encoding + and percent-encoding shall be considered. <c>transcode/2</c> provides the means to convert + between the supported encodings, it takes a <c>uri_string()</c> and a list of options + specifying inbound and outbound encodings.</p> + <p><url href="https://www.ietf.org/rfc/rfc3986.txt">RFC 3986</url> does not mandate any specific + character encoding and it is usually defined by the protocol or surrounding text. This library + takes the same assumption, binary and percent-encoding are handled as one configuration unit, + they cannot be set to different values.</p> + </description> + + <datatypes> + <datatype> + <name name="error"/> + <desc> + <p>Error tuple indicating the type of error. Possible values of the second component:</p> + <list type="bulleted"> + <item><c>invalid_character</c></item> + <item><c>invalid_encoding</c></item> + <item><c>invalid_input</c></item> + <item><c>invalid_map</c></item> + <item><c>invalid_percent_encoding</c></item> + <item><c>invalid_scheme</c></item> + <item><c>invalid_uri</c></item> + <item><c>invalid_utf8</c></item> + <item><c>missing_value</c></item> + </list> + <p>The third component is a term providing additional information about the + cause of the error.</p> + </desc> + </datatype> + <datatype> + <name name="uri_map"/> + <desc> + <p>Map holding the main components of a URI.</p> + </desc> + </datatype> + <datatype> + <name name="uri_string"/> + <desc> + <p>List of unicode codepoints, a UTF-8 encoded binary, or a mix of the two, + representing an <url href="https://www.ietf.org/rfc/rfc3986.txt">RFC 3986</url> + compliant URI (<em>percent-encoded form</em>). + A URI is a sequence of characters from a very limited set: the letters of + the basic Latin alphabet, digits, and a few special characters.</p> + </desc> + </datatype> + </datatypes> + + <funcs> + + <func> + <name name="compose_query" arity="1"/> + <fsummary>Compose urlencoded query string.</fsummary> + <desc> + <p>Composes a form-urlencoded <c><anno>QueryString</anno></c> based on a + <c><anno>QueryList</anno></c>, a list of non-percent-encoded key-value pairs. + Form-urlencoding is defined in section + 4.10.22.6 of the <url href="https://www.w3.org/TR/html5/forms.html">HTML5</url> + specification. + </p> + <p>See also the opposite operation <seealso marker="#dissect_query/1"> + <c>dissect_query/1</c></seealso>. + </p> + <p><em>Example:</em></p> + <pre> +1> <input>uri_string:compose_query([{"foo bar","1"},{"city","örebro"}]).</input> +<![CDATA["foo+bar=1&city=%C3%B6rebro"]]> +2> <![CDATA[uri_string:compose_query([{<<"foo bar">>,<<"1">>}, +2> {<<"city">>,<<"örebro"/utf8>>}]).]]> +<![CDATA[<<"foo+bar=1&city=%C3%B6rebro">>]]> + </pre> + </desc> + </func> + + <func> + <name name="compose_query" arity="2"/> + <fsummary>Compose urlencoded query string.</fsummary> + <desc> + <p>Same as <c>compose_query/1</c> but with an additional + <c><anno>Options</anno></c> parameter, that controls the encoding ("charset") + used by the encoding algorithm. There are two supported encodings: <c>utf8</c> + (or <c>unicode</c>) and <c>latin1</c>. + </p> + <p>Each character in the entry's name and value that cannot be expressed using + the selected character encoding, is replaced by a string consisting of a U+0026 + AMPERSAND character (<![CDATA[&]]>), a "#" (U+0023) character, one or more ASCII + digits representing the Unicode code point of the character in base ten, and + finally a ";" (U+003B) character. + </p> + <p>Bytes that are out of the range 0x2A, 0x2D, 0x2E, 0x30 to 0x39, 0x41 to 0x5A, 0x5F, + 0x61 to 0x7A, are percent-encoded (U+0025 PERCENT SIGN character (%) followed by + uppercase ASCII hex digits representing the hexadecimal value of the byte). + </p> + <p>See also the opposite operation <seealso marker="#dissect_query/1"> + <c>dissect_query/1</c></seealso>. + </p> + <p><em>Example:</em></p> + <pre> +1> <input>uri_string:compose_query([{"foo bar","1"},{"city","örebro"}],</input> +1> [{encoding, latin1}]). +<![CDATA["foo+bar=1&city=%F6rebro" +2> uri_string:compose_query([{<<"foo bar">>,<<"1">>}, +2> {<<"city">>,<<"東京"/utf8>>}], [{encoding, latin1}]).]]> +<![CDATA[<<"foo+bar=1&city=%26%2326481%3B%26%2320140%3B">>]]> + </pre> + </desc> + </func> + + <func> + <name name="dissect_query" arity="1"/> + <fsummary>Dissect query string.</fsummary> + <desc> + <p>Dissects an urlencoded <c><anno>QueryString</anno></c> and returns a + <c><anno>QueryList</anno></c>, a list of non-percent-encoded key-value pairs. + Form-urlencoding is defined in section + 4.10.22.6 of the <url href="https://www.w3.org/TR/html5/forms.html">HTML5</url> + specification. + </p> + <p>It is not as strict for its input as the decoding algorithm defined by + <url href="https://www.w3.org/TR/html5/forms.html">HTML5</url> + and accepts all unicode characters.</p> + <p>See also the opposite operation <seealso marker="#compose_query/1"> + <c>compose_query/1</c></seealso>. + </p> + <p><em>Example:</em></p> + <pre> +1> <input><![CDATA[uri_string:dissect_query("foo+bar=1&city=%C3%B6rebro").]]></input> +[{"foo bar","1"},{"city","örebro"}] +2> <![CDATA[uri_string:dissect_query(<<"foo+bar=1&city=%26%2326481%3B%26%2320140%3B">>).]]> +<![CDATA[[{<<"foo bar">>,<<"1">>}, + {<<"city">>,<<230,157,177,228,186,172>>}] ]]> + </pre> + </desc> + </func> + + <func> + <name name="normalize" arity="1"/> + <fsummary>Syntax-based normalization.</fsummary> + <desc> + <p>Transforms <c><anno>URIString</anno></c> into a normalized form + using Syntax-Based Normalization as defined by + <url href="https://www.ietf.org/rfc/rfc3986.txt">RFC 3986</url>.</p> + <p>This function implements case normalization, percent-encoding + normalization, path segment normalization and scheme based normalization + for HTTP(S) with basic support for FTP, SSH, SFTP and TFTP.</p> + <p><em>Example:</em></p> + <pre> +1> <input>uri_string:normalize("/a/b/c/./../../g").</input> +"/a/g" +2> <![CDATA[uri_string:normalize(<<"mid/content=5/../6">>).]]> +<![CDATA[<<"mid/6">>]]> +3> uri_string:normalize("http://localhost:80"). +"https://localhost/" + </pre> + </desc> + </func> + + <func> + <name name="parse" arity="1"/> + <fsummary>Parse URI into a map.</fsummary> + <desc> + <p>Parses an <url href="https://www.ietf.org/rfc/rfc3986.txt">RFC 3986</url> + compliant <c>uri_string()</c> into a <c>uri_map()</c>, that holds the parsed + components of the <c>URI</c>. + If parsing fails, an error tuple is returned.</p> + <p>See also the opposite operation <seealso marker="#recompose/1"> + <c>recompose/1</c></seealso>.</p> + <p><em>Example:</em></p> + <pre> +1> <input>uri_string:parse("foo://[email protected]:8042/over/there?name=ferret#nose").</input> +#{fragment => "nose",host => "example.com", + path => "/over/there",port => 8042,query => "name=ferret", + scheme => foo,userinfo => "user"} +2> <![CDATA[uri_string:parse(<<"foo://[email protected]:8042/over/there?name=ferret">>).]]> +<![CDATA[#{host => <<"example.com">>,path => <<"/over/there">>, + port => 8042,query => <<"name=ferret">>,scheme => <<"foo">>, + userinfo => <<"user">>}]]> + </pre> + </desc> + </func> + + <func> + <name name="recompose" arity="1"/> + <fsummary>Recompose URI.</fsummary> + <desc> + <p>Creates an <url href="https://www.ietf.org/rfc/rfc3986.txt">RFC 3986</url> compliant + <c><anno>URIString</anno></c> (percent-encoded), based on the components of + <c><anno>URIMap</anno></c>. + If the <c><anno>URIMap</anno></c> is invalid, an error tuple is returned.</p> + <p>See also the opposite operation <seealso marker="#parse/1"> + <c>parse/1</c></seealso>.</p> + <p><em>Example:</em></p> + <pre> +1> <input>URIMap = #{fragment => "nose", host => "example.com", path => "/over/there",</input> +1> port => 8042, query => "name=ferret", scheme => "foo", userinfo => "user"}. +#{fragment => "top",host => "example.com", + path => "/over/there",port => 8042,query => "?name=ferret", + scheme => foo,userinfo => "user"} + +2> <input>uri_string:recompose(URIMap).</input> +"foo://example.com:8042/over/there?name=ferret#nose"</pre> + </desc> + </func> + + <func> + <name name="transcode" arity="2"/> + <fsummary>Transcode URI.</fsummary> + <desc> + <p>Transcodes an <url href="https://www.ietf.org/rfc/rfc3986.txt">RFC 3986</url> + compliant <c><anno>URIString</anno></c>, + where <c><anno>Options</anno></c> is a list of tagged tuples, specifying the inbound + (<c>in_encoding</c>) and outbound (<c>out_encoding</c>) encodings. <c>in_encoding</c> + and <c>out_encoding</c> specifies both binary encoding and percent-encoding for the + input and output data. Mixed encoding, where binary encoding is not the same as + percent-encoding, is not supported. + If an argument is invalid, an error tuple is returned.</p> + <p><em>Example:</em></p> + <pre> +1> <input><![CDATA[uri_string:transcode(<<"foo%00%00%00%F6bar"/utf32>>,]]></input> +1> [{in_encoding, utf32},{out_encoding, utf8}]). +<![CDATA[<<"foo%C3%B6bar"/utf8>>]]> +2> uri_string:transcode("foo%F6bar", [{in_encoding, latin1}, +2> {out_encoding, utf8}]). +"foo%C3%B6bar" + </pre> + </desc> + </func> + + </funcs> +</erlref> diff --git a/lib/stdlib/doc/src/user_guide.gif b/lib/stdlib/doc/src/user_guide.gif Binary files differdeleted file mode 100644 index e6275a803d..0000000000 --- a/lib/stdlib/doc/src/user_guide.gif +++ /dev/null diff --git a/lib/stdlib/src/Makefile b/lib/stdlib/src/Makefile index bf836203ec..8b156929d7 100644 --- a/lib/stdlib/src/Makefile +++ b/lib/stdlib/src/Makefile @@ -121,6 +121,7 @@ MODULES= \ timer \ unicode \ unicode_util \ + uri_string \ win32reg \ zip diff --git a/lib/stdlib/src/base64.erl b/lib/stdlib/src/base64.erl index 5885745fb1..6ea4147abf 100644 --- a/lib/stdlib/src/base64.erl +++ b/lib/stdlib/src/base64.erl @@ -24,22 +24,11 @@ -export([encode/1, decode/1, mime_decode/1, encode_to_string/1, decode_to_string/1, mime_decode_to_string/1]). -%%------------------------------------------------------------------------- %% The following type is a subtype of string() for return values %% of (some) functions of this module. -%%------------------------------------------------------------------------- - -type ascii_string() :: [1..255]. -type ascii_binary() :: binary(). -%%------------------------------------------------------------------------- -%% encode_to_string(ASCII) -> Base64String -%% ASCII - string() | binary() -%% Base64String - string() -%% -%% Description: Encodes a plain ASCII string (or binary) into base64. -%%------------------------------------------------------------------------- - -spec encode_to_string(Data) -> Base64String when Data :: ascii_string() | ascii_binary(), Base64String :: ascii_string(). @@ -47,107 +36,89 @@ encode_to_string(Bin) when is_binary(Bin) -> encode_to_string(binary_to_list(Bin)); encode_to_string(List) when is_list(List) -> - encode_l(List). - -%%------------------------------------------------------------------------- -%% encode(ASCII) -> Base64 -%% ASCII - string() | binary() -%% Base64 - binary() -%% -%% Description: Encodes a plain ASCII string (or binary) into base64. -%%------------------------------------------------------------------------- + encode_list_to_string(List). -spec encode(Data) -> Base64 when Data :: ascii_string() | ascii_binary(), Base64 :: ascii_binary(). encode(Bin) when is_binary(Bin) -> - encode_binary(Bin); + encode_binary(Bin, <<>>); encode(List) when is_list(List) -> - list_to_binary(encode_l(List)). + encode_list(List, <<>>). --spec encode_l(ascii_string()) -> ascii_string(). - -encode_l([]) -> +encode_list_to_string([]) -> []; -encode_l([A]) -> - [b64e(A bsr 2), - b64e((A band 3) bsl 4), $=, $=]; -encode_l([A,B]) -> - [b64e(A bsr 2), - b64e(((A band 3) bsl 4) bor (B bsr 4)), - b64e((B band 15) bsl 2), $=]; -encode_l([A,B,C|Ls]) -> - BB = (A bsl 16) bor (B bsl 8) bor C, +encode_list_to_string([B1]) -> + [b64e(B1 bsr 2), + b64e((B1 band 3) bsl 4), $=, $=]; +encode_list_to_string([B1,B2]) -> + [b64e(B1 bsr 2), + b64e(((B1 band 3) bsl 4) bor (B2 bsr 4)), + b64e((B2 band 15) bsl 2), $=]; +encode_list_to_string([B1,B2,B3|Ls]) -> + BB = (B1 bsl 16) bor (B2 bsl 8) bor B3, [b64e(BB bsr 18), b64e((BB bsr 12) band 63), b64e((BB bsr 6) band 63), - b64e(BB band 63) | encode_l(Ls)]. - -encode_binary(Bin) -> - Split = 3*(byte_size(Bin) div 3), - <<Main0:Split/binary,Rest/binary>> = Bin, - Main = << <<(b64e(C)):8>> || <<C:6>> <= Main0 >>, - case Rest of - <<A:6,B:6,C:4>> -> - <<Main/binary,(b64e(A)):8,(b64e(B)):8,(b64e(C bsl 2)):8,$=:8>>; - <<A:6,B:2>> -> - <<Main/binary,(b64e(A)):8,(b64e(B bsl 4)):8,$=:8,$=:8>>; - <<>> -> - Main - end. + b64e(BB band 63) | encode_list_to_string(Ls)]. -%%------------------------------------------------------------------------- -%% mime_decode(Base64) -> ASCII -%% decode(Base64) -> ASCII -%% Base64 - string() | binary() -%% ASCII - binary() -%% -%% Description: Decodes an base64 encoded string to plain ASCII. -%% mime_decode strips away all characters not Base64 before converting, -%% whereas decode crashes if an illegal character is found -%%------------------------------------------------------------------------- +encode_binary(<<>>, A) -> + A; +encode_binary(<<B1:8>>, A) -> + <<A/bits,(b64e(B1 bsr 2)):8,(b64e((B1 band 3) bsl 4)):8,$=:8,$=:8>>; +encode_binary(<<B1:8, B2:8>>, A) -> + <<A/bits,(b64e(B1 bsr 2)):8, + (b64e(((B1 band 3) bsl 4) bor (B2 bsr 4))):8, + (b64e((B2 band 15) bsl 2)):8, $=:8>>; +encode_binary(<<B1:8, B2:8, B3:8, Ls/bits>>, A) -> + BB = (B1 bsl 16) bor (B2 bsl 8) bor B3, + encode_binary(Ls, + <<A/bits,(b64e(BB bsr 18)):8, + (b64e((BB bsr 12) band 63)):8, + (b64e((BB bsr 6) band 63)):8, + (b64e(BB band 63)):8>>). + +encode_list([], A) -> + A; +encode_list([B1], A) -> + <<A/bits,(b64e(B1 bsr 2)):8,(b64e((B1 band 3) bsl 4)):8,$=:8,$=:8>>; +encode_list([B1,B2], A) -> + <<A/bits,(b64e(B1 bsr 2)):8, + (b64e(((B1 band 3) bsl 4) bor (B2 bsr 4))):8, + (b64e((B2 band 15) bsl 2)):8, $=:8>>; +encode_list([B1,B2,B3|Ls], A) -> + BB = (B1 bsl 16) bor (B2 bsl 8) bor B3, + encode_list(Ls, + <<A/bits,(b64e(BB bsr 18)):8, + (b64e((BB bsr 12) band 63)):8, + (b64e((BB bsr 6) band 63)):8, + (b64e(BB band 63)):8>>). + +%% mime_decode strips away all characters not Base64 before +%% converting, whereas decode crashes if an illegal character is found -spec decode(Base64) -> Data when Base64 :: ascii_string() | ascii_binary(), Data :: ascii_binary(). decode(Bin) when is_binary(Bin) -> - decode_binary(<<>>, Bin); + decode_binary(Bin, <<>>); decode(List) when is_list(List) -> - list_to_binary(decode_l(List)). + decode_list(List, <<>>). -spec mime_decode(Base64) -> Data when Base64 :: ascii_string() | ascii_binary(), Data :: ascii_binary(). mime_decode(Bin) when is_binary(Bin) -> - mime_decode_binary(<<>>, Bin); + mime_decode_binary(Bin, <<>>); mime_decode(List) when is_list(List) -> - mime_decode(list_to_binary(List)). - --spec decode_l(ascii_string()) -> ascii_string(). - -decode_l(List) -> - L = strip_spaces(List, []), - decode(L, []). - --spec mime_decode_l(ascii_string()) -> ascii_string(). + mime_decode_list(List, <<>>). -mime_decode_l(List) -> - L = strip_illegal(List, [], 0), - decode(L, []). - -%%------------------------------------------------------------------------- -%% mime_decode_to_string(Base64) -> ASCII -%% decode_to_string(Base64) -> ASCII -%% Base64 - string() | binary() -%% ASCII - binary() -%% -%% Description: Decodes an base64 encoded string to plain ASCII. -%% mime_decode strips away all characters not Base64 before converting, -%% whereas decode crashes if an illegal character is found -%%------------------------------------------------------------------------- +%% mime_decode_to_string strips away all characters not Base64 before +%% converting, whereas decode_to_string crashes if an illegal +%% character is found -spec decode_to_string(Base64) -> DataString when Base64 :: ascii_string() | ascii_binary(), @@ -156,7 +127,7 @@ mime_decode_l(List) -> decode_to_string(Bin) when is_binary(Bin) -> decode_to_string(binary_to_list(Bin)); decode_to_string(List) when is_list(List) -> - decode_l(List). + decode_list_to_string(List). -spec mime_decode_to_string(Base64) -> DataString when Base64 :: ascii_string() | ascii_binary(), @@ -165,202 +136,326 @@ decode_to_string(List) when is_list(List) -> mime_decode_to_string(Bin) when is_binary(Bin) -> mime_decode_to_string(binary_to_list(Bin)); mime_decode_to_string(List) when is_list(List) -> - mime_decode_l(List). - -%% One-based decode map. --define(DECODE_MAP, - {bad,bad,bad,bad,bad,bad,bad,bad,ws,ws,bad,bad,ws,bad,bad, %1-15 - bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad, %16-31 - ws,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,62,bad,bad,bad,63, %32-47 - 52,53,54,55,56,57,58,59,60,61,bad,bad,bad,eq,bad,bad, %48-63 - bad,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14, - 15,16,17,18,19,20,21,22,23,24,25,bad,bad,bad,bad,bad, - bad,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40, - 41,42,43,44,45,46,47,48,49,50,51,bad,bad,bad,bad,bad, - bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad, - bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad, - bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad, - bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad, - bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad, - bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad, - bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad, - bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad}). - -decode_binary(Result0, <<C:8,T0/bits>>) -> - case element(C, ?DECODE_MAP) of - bad -> - erlang:error({badarg,C}); - ws -> - decode_binary(Result0, T0); - eq -> - case strip_ws(T0) of - <<$=:8,T/binary>> -> - <<>> = strip_ws(T), - Split = byte_size(Result0) - 1, - <<Result:Split/bytes,_:4>> = Result0, - Result; - T -> - <<>> = strip_ws(T), - Split = byte_size(Result0) - 1, - <<Result:Split/bytes,_:2>> = Result0, - Result - end; - Bits -> - decode_binary(<<Result0/bits,Bits:6>>, T0) - end; -decode_binary(Result, <<>>) -> - true = is_binary(Result), - Result. + mime_decode_list_to_string(List). %% Skipping pad character if not at end of string. Also liberal about %% excess padding and skipping of other illegal (non-base64 alphabet) %% characters. See section 3.3 of RFC4648 -mime_decode_binary(Result, <<0:8,T/bits>>) -> - mime_decode_binary(Result, T); -mime_decode_binary(Result0, <<C:8,T/bits>>) -> - case element(C, ?DECODE_MAP) of - Bits when is_integer(Bits) -> - mime_decode_binary(<<Result0/bits,Bits:6>>, T); +mime_decode_list([0 | Cs], A) -> + mime_decode_list(Cs, A); +mime_decode_list([C1 | Cs], A) -> + case b64d(C1) of + B1 when is_integer(B1) -> mime_decode_list(Cs, A, B1); + _ -> mime_decode_list(Cs, A) % eq is padding + end; +mime_decode_list([], A) -> + A. + +mime_decode_list([0 | Cs], A, B1) -> + mime_decode_list(Cs, A, B1); +mime_decode_list([C2 | Cs], A, B1) -> + case b64d(C2) of + B2 when is_integer(B2) -> + mime_decode_list(Cs, A, B1, B2); + _ -> mime_decode_list(Cs, A, B1) % eq is padding + end. + +mime_decode_list([0 | Cs], A, B1, B2) -> + mime_decode_list(Cs, A, B1, B2); +mime_decode_list([C3 | Cs], A, B1, B2) -> + case b64d(C3) of + B3 when is_integer(B3) -> + mime_decode_list(Cs, A, B1, B2, B3); + eq=B3 -> + mime_decode_list_after_eq(Cs, A, B1, B2, B3); + _ -> mime_decode_list(Cs, A, B1, B2) + end. + +mime_decode_list([0 | Cs], A, B1, B2, B3) -> + mime_decode_list(Cs, A, B1, B2, B3); +mime_decode_list([C4 | Cs], A, B1, B2, B3) -> + case b64d(C4) of + B4 when is_integer(B4) -> + mime_decode_list(Cs, <<A/bits,B1:6,B2:6,B3:6,B4:6>>); eq -> - mime_decode_binary_after_eq(Result0, T, false); - _ -> - mime_decode_binary(Result0, T) + mime_decode_list_after_eq(Cs, A, B1, B2, B3); + _ -> mime_decode_list(Cs, A, B1, B2, B3) + end. + +mime_decode_list_after_eq([0 | Cs], A, B1, B2, B3) -> + mime_decode_list_after_eq(Cs, A, B1, B2, B3); +mime_decode_list_after_eq([C | Cs], A, B1, B2, B3) -> + case b64d(C) of + B when is_integer(B) -> + %% More valid data, skip the eq as invalid + case B3 of + eq -> mime_decode_list(Cs, A, B1, B2, B); + _ -> mime_decode_list(Cs, <<A/bits,B1:6,B2:6,B3:6,B:6>>) + end; + _ -> mime_decode_list_after_eq(Cs, A, B1, B2, B3) end; -mime_decode_binary(Result, _) -> - true = is_binary(Result), - Result. - -mime_decode_binary_after_eq(Result, <<0:8,T/bits>>, Eq) -> - mime_decode_binary_after_eq(Result, T, Eq); -mime_decode_binary_after_eq(Result0, <<C:8,T/bits>>, Eq) -> - case element(C, ?DECODE_MAP) of - bad -> - mime_decode_binary_after_eq(Result0, T, Eq); - ws -> - mime_decode_binary_after_eq(Result0, T, Eq); +mime_decode_list_after_eq([], A, B1, B2, eq) -> + <<A/bits,B1:6,(B2 bsr 4):2>>; +mime_decode_list_after_eq([], A, B1, B2, B3) -> + <<A/bits,B1:6,B2:6,(B3 bsr 2):4>>. + +mime_decode_binary(<<0:8, Cs/bits>>, A) -> + mime_decode_binary(Cs, A); +mime_decode_binary(<<C1:8, Cs/bits>>, A) -> + case b64d(C1) of + B1 when is_integer(B1) -> mime_decode_binary(Cs, A, B1); + _ -> mime_decode_binary(Cs, A) % eq is padding + end; +mime_decode_binary(<<>>, A) -> + A. + +mime_decode_binary(<<0:8, Cs/bits>>, A, B1) -> + mime_decode_binary(Cs, A, B1); +mime_decode_binary(<<C2:8, Cs/bits>>, A, B1) -> + case b64d(C2) of + B2 when is_integer(B2) -> + mime_decode_binary(Cs, A, B1, B2); + _ -> mime_decode_binary(Cs, A, B1) % eq is padding + end. + +mime_decode_binary(<<0:8, Cs/bits>>, A, B1, B2) -> + mime_decode_binary(Cs, A, B1, B2); +mime_decode_binary(<<C3:8, Cs/bits>>, A, B1, B2) -> + case b64d(C3) of + B3 when is_integer(B3) -> + mime_decode_binary(Cs, A, B1, B2, B3); + eq=B3 -> + mime_decode_binary_after_eq(Cs, A, B1, B2, B3); + _ -> mime_decode_binary(Cs, A, B1, B2) + end. + +mime_decode_binary(<<0:8, Cs/bits>>, A, B1, B2, B3) -> + mime_decode_binary(Cs, A, B1, B2, B3); +mime_decode_binary(<<C4:8, Cs/bits>>, A, B1, B2, B3) -> + case b64d(C4) of + B4 when is_integer(B4) -> + mime_decode_binary(Cs, <<A/bits,B1:6,B2:6,B3:6,B4:6>>); eq -> - mime_decode_binary_after_eq(Result0, T, true); - Bits when is_integer(Bits) -> + mime_decode_binary_after_eq(Cs, A, B1, B2, B3); + _ -> mime_decode_binary(Cs, A, B1, B2, B3) + end. + +mime_decode_binary_after_eq(<<0:8, Cs/bits>>, A, B1, B2, B3) -> + mime_decode_binary_after_eq(Cs, A, B1, B2, B3); +mime_decode_binary_after_eq(<<C:8, Cs/bits>>, A, B1, B2, B3) -> + case b64d(C) of + B when is_integer(B) -> %% More valid data, skip the eq as invalid - mime_decode_binary(<<Result0/bits,Bits:6>>, T) + case B3 of + eq -> mime_decode_binary(Cs, A, B1, B2, B); + _ -> mime_decode_binary(Cs, <<A/bits,B1:6,B2:6,B3:6,B:6>>) + end; + _ -> mime_decode_binary_after_eq(Cs, A, B1, B2, B3) + end; +mime_decode_binary_after_eq(<<>>, A, B1, B2, eq) -> + <<A/bits,B1:6,(B2 bsr 4):2>>; +mime_decode_binary_after_eq(<<>>, A, B1, B2, B3) -> + <<A/bits,B1:6,B2:6,(B3 bsr 2):4>>. + +mime_decode_list_to_string([0 | Cs]) -> + mime_decode_list_to_string(Cs); +mime_decode_list_to_string([C1 | Cs]) -> + case b64d(C1) of + B1 when is_integer(B1) -> mime_decode_list_to_string(Cs, B1); + _ -> mime_decode_list_to_string(Cs) % eq is padding end; -mime_decode_binary_after_eq(Result0, <<>>, Eq) -> - %% No more valid data. - case bit_size(Result0) rem 8 of - 0 -> - %% '====' is not uncommon. - Result0; - 4 when Eq -> - %% enforce at least one more '=' only ignoring illegals and spacing - Split = byte_size(Result0) - 1, - <<Result:Split/bytes,_:4>> = Result0, - Result; - 2 -> - %% remove 2 bits - Split = byte_size(Result0) - 1, - <<Result:Split/bytes,_:2>> = Result0, - Result +mime_decode_list_to_string([]) -> + []. + +mime_decode_list_to_string([0 | Cs], B1) -> + mime_decode_list_to_string(Cs, B1); +mime_decode_list_to_string([C2 | Cs], B1) -> + case b64d(C2) of + B2 when is_integer(B2) -> + mime_decode_list_to_string(Cs, B1, B2); + _ -> mime_decode_list_to_string(Cs, B1) % eq is padding end. -decode([], A) -> A; -decode([$=,$=,C2,C1|Cs], A) -> - Bits2x6 = (b64d(C1) bsl 18) bor (b64d(C2) bsl 12), - Octet1 = Bits2x6 bsr 16, - decode(Cs, [Octet1|A]); -decode([$=,C3,C2,C1|Cs], A) -> - Bits3x6 = (b64d(C1) bsl 18) bor (b64d(C2) bsl 12) - bor (b64d(C3) bsl 6), - Octet1 = Bits3x6 bsr 16, - Octet2 = (Bits3x6 bsr 8) band 16#ff, - decode(Cs, [Octet1,Octet2|A]); -decode([C4,C3,C2,C1| Cs], A) -> - Bits4x6 = (b64d(C1) bsl 18) bor (b64d(C2) bsl 12) - bor (b64d(C3) bsl 6) bor b64d(C4), - Octet1 = Bits4x6 bsr 16, - Octet2 = (Bits4x6 bsr 8) band 16#ff, - Octet3 = Bits4x6 band 16#ff, - decode(Cs, [Octet1,Octet2,Octet3|A]). +mime_decode_list_to_string([0 | Cs], B1, B2) -> + mime_decode_list_to_string(Cs, B1, B2); +mime_decode_list_to_string([C3 | Cs], B1, B2) -> + case b64d(C3) of + B3 when is_integer(B3) -> + mime_decode_list_to_string(Cs, B1, B2, B3); + eq=B3 -> mime_decode_list_to_string_after_eq(Cs, B1, B2, B3); + _ -> mime_decode_list_to_string(Cs, B1, B2) + end. -%%%======================================================================== -%%% Internal functions -%%%======================================================================== +mime_decode_list_to_string([0 | Cs], B1, B2, B3) -> + mime_decode_list_to_string(Cs, B1, B2, B3); +mime_decode_list_to_string([C4 | Cs], B1, B2, B3) -> + case b64d(C4) of + B4 when is_integer(B4) -> + Bits4x6 = (B1 bsl 18) bor (B2 bsl 12) bor (B3 bsl 6) bor B4, + Octet1 = Bits4x6 bsr 16, + Octet2 = (Bits4x6 bsr 8) band 16#ff, + Octet3 = Bits4x6 band 16#ff, + [Octet1, Octet2, Octet3 | mime_decode_list_to_string(Cs)]; + eq -> + mime_decode_list_to_string_after_eq(Cs, B1, B2, B3); + _ -> mime_decode_list_to_string(Cs, B1, B2, B3) + end. -strip_spaces([], A) -> A; -strip_spaces([$\s|Cs], A) -> strip_spaces(Cs, A); -strip_spaces([$\t|Cs], A) -> strip_spaces(Cs, A); -strip_spaces([$\r|Cs], A) -> strip_spaces(Cs, A); -strip_spaces([$\n|Cs], A) -> strip_spaces(Cs, A); -strip_spaces([C|Cs], A) -> strip_spaces(Cs, [C | A]). - -strip_ws(<<$\t,T/binary>>) -> - strip_ws(T); -strip_ws(<<$\n,T/binary>>) -> - strip_ws(T); -strip_ws(<<$\r,T/binary>>) -> - strip_ws(T); -strip_ws(<<$\s,T/binary>>) -> - strip_ws(T); -strip_ws(T) -> T. +mime_decode_list_to_string_after_eq([0 | Cs], B1, B2, B3) -> + mime_decode_list_to_string_after_eq(Cs, B1, B2, B3); +mime_decode_list_to_string_after_eq([C | Cs], B1, B2, B3) -> + case b64d(C) of + B when is_integer(B) -> + %% More valid data, skip the eq as invalid + case B3 of + eq -> mime_decode_list_to_string(Cs, B1, B2, B); + _ -> + Bits4x6 = (B1 bsl 18) bor (B2 bsl 12) bor (B3 bsl 6) bor B, + Octet1 = Bits4x6 bsr 16, + Octet2 = (Bits4x6 bsr 8) band 16#ff, + Octet3 = Bits4x6 band 16#ff, + [Octet1, Octet2, Octet3 | mime_decode_list_to_string(Cs)] + end; + _ -> mime_decode_list_to_string_after_eq(Cs, B1, B2, B3) + end; +mime_decode_list_to_string_after_eq([], B1, B2, eq) -> + binary_to_list(<<B1:6,(B2 bsr 4):2>>); +mime_decode_list_to_string_after_eq([], B1, B2, B3) -> + binary_to_list(<<B1:6,B2:6,(B3 bsr 2):4>>). + +decode_list([C1 | Cs], A) -> + case b64d(C1) of + ws -> decode_list(Cs, A); + B1 -> decode_list(Cs, A, B1) + end; +decode_list([], A) -> + A. -%% Skipping pad character if not at end of string. Also liberal about -%% excess padding and skipping of other illegal (non-base64 alphabet) -%% characters. See section 3.3 of RFC4648 -strip_illegal([], A, _Cnt) -> +decode_list([C2 | Cs], A, B1) -> + case b64d(C2) of + ws -> decode_list(Cs, A, B1); + B2 -> decode_list(Cs, A, B1, B2) + end. + +decode_list([C3 | Cs], A, B1, B2) -> + case b64d(C3) of + ws -> decode_list(Cs, A, B1, B2); + B3 -> decode_list(Cs, A, B1, B2, B3) + end. + +decode_list([C4 | Cs], A, B1, B2, B3) -> + case b64d(C4) of + ws -> decode_list(Cs, A, B1, B2, B3); + eq when B3 =:= eq -> only_ws(Cs, <<A/bits,B1:6,(B2 bsr 4):2>>); + eq -> only_ws(Cs, <<A/bits,B1:6,B2:6,(B3 bsr 2):4>>); + B4 -> decode_list(Cs, <<A/bits,B1:6,B2:6,B3:6,B4:6>>) + end. + +decode_binary(<<C1:8, Cs/bits>>, A) -> + case b64d(C1) of + ws -> decode_binary(Cs, A); + B1 -> decode_binary(Cs, A, B1) + end; +decode_binary(<<>>, A) -> + A. + +decode_binary(<<C2:8, Cs/bits>>, A, B1) -> + case b64d(C2) of + ws -> decode_binary(Cs, A, B1); + B2 -> decode_binary(Cs, A, B1, B2) + end. + +decode_binary(<<C3:8, Cs/bits>>, A, B1, B2) -> + case b64d(C3) of + ws -> decode_binary(Cs, A, B1, B2); + B3 -> decode_binary(Cs, A, B1, B2, B3) + end. + +decode_binary(<<C4:8, Cs/bits>>, A, B1, B2, B3) -> + case b64d(C4) of + ws -> decode_binary(Cs, A, B1, B2, B3); + eq when B3 =:= eq -> only_ws_binary(Cs, <<A/bits,B1:6,(B2 bsr 4):2>>); + eq -> only_ws_binary(Cs, <<A/bits,B1:6,B2:6,(B3 bsr 2):4>>); + B4 -> decode_binary(Cs, <<A/bits,B1:6,B2:6,B3:6,B4:6>>) + end. + +only_ws_binary(<<>>, A) -> A; -strip_illegal([0|Cs], A, Cnt) -> - strip_illegal(Cs, A, Cnt); -strip_illegal([C|Cs], A, Cnt) -> - case element(C, ?DECODE_MAP) of - bad -> - strip_illegal(Cs, A, Cnt); - ws -> - strip_illegal(Cs, A, Cnt); - eq -> - case {tail_contains_more(Cs, false), Cnt rem 4} of - {{[], _}, 0} -> - A; %% Ignore extra = - {{[], true}, 2} -> - [$=|[$=|A]]; %% 'XX==' - {{[], _}, 3} -> - [$=|A]; %% 'XXX=' - {{[H|T], _}, _} -> - %% more data, skip equals - strip_illegal(T, [H|A], Cnt+1) - end; - _ -> - strip_illegal(Cs, [C|A], Cnt+1) +only_ws_binary(<<C:8, Cs/bits>>, A) -> + case b64d(C) of + ws -> only_ws_binary(Cs, A) end. -%% Search the tail for more valid data and remember if we saw -%% another equals along the way. -tail_contains_more([], Eq) -> - {[], Eq}; -tail_contains_more(<<>>, Eq) -> - {<<>>, Eq}; -tail_contains_more([C|T]=More, Eq) -> - case element(C, ?DECODE_MAP) of - bad -> - tail_contains_more(T, Eq); - ws -> - tail_contains_more(T, Eq); - eq -> - tail_contains_more(T, true); - _ -> - {More, Eq} +decode_list_to_string([C1 | Cs]) -> + case b64d(C1) of + ws -> decode_list_to_string(Cs); + B1 -> decode_list_to_string(Cs, B1) end; -tail_contains_more(<<C:8,T/bits>> =More, Eq) -> - case element(C, ?DECODE_MAP) of - bad -> - tail_contains_more(T, Eq); - ws -> - tail_contains_more(T, Eq); - eq -> - tail_contains_more(T, true); - _ -> - {More, Eq} +decode_list_to_string([]) -> + []. + +decode_list_to_string([C2 | Cs], B1) -> + case b64d(C2) of + ws -> decode_list_to_string(Cs, B1); + B2 -> decode_list_to_string(Cs, B1, B2) end. - + +decode_list_to_string([C3 | Cs], B1, B2) -> + case b64d(C3) of + ws -> decode_list_to_string(Cs, B1, B2); + B3 -> decode_list_to_string(Cs, B1, B2, B3) + end. + +decode_list_to_string([C4 | Cs], B1, B2, B3) -> + case b64d(C4) of + ws -> + decode_list_to_string(Cs, B1, B2, B3); + eq when B3 =:= eq -> + only_ws(Cs, binary_to_list(<<B1:6,(B2 bsr 4):2>>)); + eq -> + only_ws(Cs, binary_to_list(<<B1:6,B2:6,(B3 bsr 2):4>>)); + B4 -> + Bits4x6 = (B1 bsl 18) bor (B2 bsl 12) bor (B3 bsl 6) bor B4, + Octet1 = Bits4x6 bsr 16, + Octet2 = (Bits4x6 bsr 8) band 16#ff, + Octet3 = Bits4x6 band 16#ff, + [Octet1, Octet2, Octet3 | decode_list_to_string(Cs)] + end. + +only_ws([], A) -> + A; +only_ws([C | Cs], A) -> + case b64d(C) of + ws -> only_ws(Cs, A) + end. + +%%%======================================================================== +%%% Internal functions +%%%======================================================================== + %% accessors +-compile({inline, [{b64d, 1}]}). +%% One-based decode map. +b64d(X) -> + element(X, + {bad,bad,bad,bad,bad,bad,bad,bad,ws,ws,bad,bad,ws,bad,bad, %1-15 + bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad, %16-31 + ws,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,62,bad,bad,bad,63, %32-47 + 52,53,54,55,56,57,58,59,60,61,bad,bad,bad,eq,bad,bad, %48-63 + bad,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14, + 15,16,17,18,19,20,21,22,23,24,25,bad,bad,bad,bad,bad, + bad,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40, + 41,42,43,44,45,46,47,48,49,50,51,bad,bad,bad,bad,bad, + bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad, + bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad, + bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad, + bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad, + bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad, + bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad, + bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad, + bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad,bad}). + +-compile({inline, [{b64e, 1}]}). b64e(X) -> element(X+1, {$A, $B, $C, $D, $E, $F, $G, $H, $I, $J, $K, $L, $M, $N, @@ -368,9 +463,3 @@ b64e(X) -> $a, $b, $c, $d, $e, $f, $g, $h, $i, $j, $k, $l, $m, $n, $o, $p, $q, $r, $s, $t, $u, $v, $w, $x, $y, $z, $0, $1, $2, $3, $4, $5, $6, $7, $8, $9, $+, $/}). - - -b64d(X) -> - b64d_ok(element(X, ?DECODE_MAP)). - -b64d_ok(I) when is_integer(I) -> I. diff --git a/lib/stdlib/src/c.erl b/lib/stdlib/src/c.erl index c04a201ce1..9a447af5b7 100644 --- a/lib/stdlib/src/c.erl +++ b/lib/stdlib/src/c.erl @@ -668,19 +668,23 @@ lm() -> [l(M) || M <- mm()]. %% erlangrc(Home) -%% Try to run a ".erlang" file, first in the current directory -%% else in home directory. +%% Try to run a ".erlang" file in home directory. + +-spec erlangrc() -> {ok, file:filename()} | {error, term()}. erlangrc() -> case init:get_argument(home) of {ok,[[Home]]} -> erlangrc([Home]); _ -> - f_p_e(["."], ".erlang") + {error, enoent} end. -erlangrc([Home]) -> - f_p_e([".",Home], ".erlang"). +-spec erlangrc(PathList) -> {ok, file:filename()} | {error, term()} + when PathList :: [Dir :: file:name()]. + +erlangrc([Home|_]=Paths) when is_list(Home) -> + f_p_e(Paths, ".erlang"). error(Fmt, Args) -> error_logger:error_msg(Fmt, Args). @@ -692,11 +696,11 @@ f_p_e(P, F) -> {error, E={Line, _Mod, _Term}} -> error("file:path_eval(~tp,~tp): error on line ~p: ~ts~n", [P, F, Line, file:format_error(E)]), - ok; + {error, E}; {error, E} -> error("file:path_eval(~tp,~tp): ~ts~n", [P, F, file:format_error(E)]), - ok; + {error, E}; Other -> Other end. diff --git a/lib/stdlib/src/erl_compile.erl b/lib/stdlib/src/erl_compile.erl index 18d7548fdc..f781312ca2 100644 --- a/lib/stdlib/src/erl_compile.erl +++ b/lib/stdlib/src/erl_compile.erl @@ -188,6 +188,8 @@ parse_dep_option("", T) -> {[makedep,{makedep_output,standard_io}],T}; parse_dep_option("D", T) -> {[makedep],T}; +parse_dep_option("MD", T) -> + {[makedep_side_effect],T}; parse_dep_option("F"++Opt, T0) -> {File,T} = get_option("MF", Opt, T0), {[makedep,{makedep_output,File}],T}; @@ -221,6 +223,7 @@ usage() -> "the dependencies"}, {"-MP","add a phony target for each dependency"}, {"-MD","same as -M -MT file (with default 'file')"}, + {"-MMD","generate dependencies as a side-effect"}, {"-o name","name output directory or file"}, {"-pa path","add path to the front of Erlang's code path"}, {"-pz path","add path to the end of Erlang's code path"}, diff --git a/lib/stdlib/src/erl_lint.erl b/lib/stdlib/src/erl_lint.erl index f58cb35cea..1930c462e8 100644 --- a/lib/stdlib/src/erl_lint.erl +++ b/lib/stdlib/src/erl_lint.erl @@ -144,6 +144,7 @@ value_option(Flag, Default, On, OnVal, Off, OffVal, Opts) -> :: dict:dict(ta(), #typeinfo{}), exp_types=gb_sets:empty() %Exported types :: gb_sets:set(ta()), + in_try_head=false :: boolean(), %In a try head. catch_scope = none %Inside/outside try or catch :: catch_scope() }). @@ -312,6 +313,10 @@ format_error({unused_var, V}) -> io_lib:format("variable ~w is unused", [V]); format_error({variable_in_record_def,V}) -> io_lib:format("variable ~w in record definition", [V]); +format_error({stacktrace_guard,V}) -> + io_lib:format("stacktrace variable ~w must not be used in a guard", [V]); +format_error({stacktrace_bound,V}) -> + io_lib:format("stacktrace variable ~w must not be previously bound", [V]); %% --- binaries --- format_error({undefined_bittype,Type}) -> io_lib:format("bit type ~tw undefined", [Type]); @@ -3218,11 +3223,11 @@ is_module_dialyzer_option(Option) -> try_clauses(Scs, Ccs, In, Vt, St0) -> {Csvt0,St1} = icrt_clauses(Scs, Vt, St0), - St2 = St1#lint{catch_scope=try_catch}, + St2 = St1#lint{catch_scope=try_catch,in_try_head=true}, {Csvt1,St3} = icrt_clauses(Ccs, Vt, St2), Csvt = Csvt0 ++ Csvt1, UpdVt = icrt_export(Csvt, Vt, In, St3), - {UpdVt,St3}. + {UpdVt,St3#lint{in_try_head=false}}. %% icrt_clauses(Clauses, In, ImportVarTable, State) -> %% {UpdVt,State}. @@ -3239,12 +3244,29 @@ icrt_clauses(Cs, Vt, St) -> mapfoldl(fun (C, St0) -> icrt_clause(C, Vt, St0) end, St, Cs). icrt_clause({clause,_Line,H,G,B}, Vt0, #lint{catch_scope=Scope}=St0) -> - {Hvt,Binvt,St1} = head(H, Vt0, St0), - Vt1 = vtupdate(Hvt, Binvt), - {Gvt,St2} = guard(G, vtupdate(Vt1, Vt0), St1), - Vt2 = vtupdate(Gvt, Vt1), - {Bvt,St3} = exprs(B, vtupdate(Vt2, Vt0), St2), - {vtupdate(Bvt, Vt2),St3#lint{catch_scope=Scope}}. + Vt1 = taint_stack_var(Vt0, H, St0), + {Hvt,Binvt,St1} = head(H, Vt1, St0), + Vt2 = vtupdate(Hvt, Binvt), + Vt3 = taint_stack_var(Vt2, H, St0), + {Gvt,St2} = guard(G, vtupdate(Vt3, Vt0), St1#lint{in_try_head=false}), + Vt4 = vtupdate(Gvt, Vt2), + {Bvt,St3} = exprs(B, vtupdate(Vt4, Vt0), St2), + {vtupdate(Bvt, Vt4),St3#lint{catch_scope=Scope}}. + +taint_stack_var(Vt, Pat, #lint{in_try_head=true}) -> + [{tuple,_,[_,_,{var,_,Stk}]}] = Pat, + case Stk of + '_' -> + Vt; + _ -> + lists:map(fun({V,{bound,Used,Lines}}) when V =:= Stk -> + {V,{stacktrace,Used,Lines}}; + (B) -> + B + end, Vt) + end; +taint_stack_var(Vt, _Pat, #lint{in_try_head=false}) -> + Vt. icrt_export(Vts, Vt, {Tag,Attrs}, St) -> {_File,Loc} = loc(Attrs, St), @@ -3484,6 +3506,9 @@ pat_var(V, Line, Vt, Bvt, St) -> {[{V,{bound,used,Ls}}],[], %% As this is matching, exported vars are risky. add_warning(Line, {exported_var,V,From}, St)}; + {ok,{stacktrace,_Usage,Ls}} -> + {[{V,{bound,used,Ls}}],[], + add_error(Line, {stacktrace_bound,V}, St)}; error when St#lint.recdef_top -> {[],[{V,{bound,unused,[Line]}}], add_error(Line, {variable_in_record_def,V}, St)}; @@ -3541,6 +3566,9 @@ expr_var(V, Line, Vt, St) -> false -> {[{V,{{export,From},used,Ls}}],St} end; + {ok,{stacktrace,_Usage,Ls}} -> + {[{V,{bound,used,Ls}}], + add_error(Line, {stacktrace_guard,V}, St)}; error -> {[{V,{bound,used,[Line]}}], add_error(Line, {unbound_var,V}, St)} diff --git a/lib/stdlib/src/erl_parse.yrl b/lib/stdlib/src/erl_parse.yrl index 6e72d64acc..14ca24362e 100644 --- a/lib/stdlib/src/erl_parse.yrl +++ b/lib/stdlib/src/erl_parse.yrl @@ -29,6 +29,10 @@ clause_args clause_guard clause_body expr expr_100 expr_150 expr_160 expr_200 expr_300 expr_400 expr_500 expr_600 expr_700 expr_800 expr_max +pat_expr pat_expr_200 pat_expr_300 pat_expr_400 pat_expr_500 +pat_expr_600 pat_expr_700 pat_expr_800 +pat_expr_max map_pat_expr record_pat_expr +pat_argument_list pat_exprs list tail list_comprehension lc_expr lc_exprs binary_comprehension @@ -37,7 +41,7 @@ record_expr record_tuple record_field record_fields map_expr map_tuple map_field map_field_assoc map_field_exact map_fields map_key if_expr if_clause if_clauses case_expr cr_clause cr_clauses receive_expr fun_expr fun_clause fun_clauses atom_or_var integer_or_var -try_expr try_catch try_clause try_clauses +try_expr try_catch try_clause try_clauses try_opt_stacktrace function_call argument_list exprs guard atomic strings @@ -66,7 +70,7 @@ char integer float atom string var 'spec' 'callback' % helper dot. -Expect 2. +Expect 0. Rootsymbol form. @@ -210,7 +214,7 @@ function_clause -> atom clause_args clause_guard clause_body : {clause,?anno('$1'),element(3, '$1'),'$2','$3','$4'}. -clause_args -> argument_list : element(1, '$1'). +clause_args -> pat_argument_list : element(1, '$1'). clause_guard -> 'when' guard : '$2'. clause_guard -> '$empty' : []. @@ -275,6 +279,53 @@ expr_max -> receive_expr : '$1'. expr_max -> fun_expr : '$1'. expr_max -> try_expr : '$1'. +pat_expr -> pat_expr_200 '=' pat_expr : {match,?anno('$2'),'$1','$3'}. +pat_expr -> pat_expr_200 : '$1'. + +pat_expr_200 -> pat_expr_300 comp_op pat_expr_300 : + ?mkop2('$1', '$2', '$3'). +pat_expr_200 -> pat_expr_300 : '$1'. + +pat_expr_300 -> pat_expr_400 list_op pat_expr_300 : + ?mkop2('$1', '$2', '$3'). +pat_expr_300 -> pat_expr_400 : '$1'. + +pat_expr_400 -> pat_expr_400 add_op pat_expr_500 : + ?mkop2('$1', '$2', '$3'). +pat_expr_400 -> pat_expr_500 : '$1'. + +pat_expr_500 -> pat_expr_500 mult_op pat_expr_600 : + ?mkop2('$1', '$2', '$3'). +pat_expr_500 -> pat_expr_600 : '$1'. + +pat_expr_600 -> prefix_op pat_expr_700 : + ?mkop1('$1', '$2'). +pat_expr_600 -> map_pat_expr : '$1'. +pat_expr_600 -> pat_expr_700 : '$1'. + +pat_expr_700 -> record_pat_expr : '$1'. +pat_expr_700 -> pat_expr_800 : '$1'. + +pat_expr_800 -> pat_expr_max : '$1'. + +pat_expr_max -> var : '$1'. +pat_expr_max -> atomic : '$1'. +pat_expr_max -> list : '$1'. +pat_expr_max -> binary : '$1'. +pat_expr_max -> tuple : '$1'. +pat_expr_max -> '(' pat_expr ')' : '$2'. + +map_pat_expr -> '#' map_tuple : + {map, ?anno('$1'),'$2'}. +map_pat_expr -> pat_expr_max '#' map_tuple : + {map, ?anno('$2'),'$1','$3'}. +map_pat_expr -> map_pat_expr '#' map_tuple : + {map, ?anno('$2'),'$1','$3'}. + +record_pat_expr -> '#' atom '.' atom : + {record_index,?anno('$1'),element(3, '$2'),'$4'}. +record_pat_expr -> '#' atom record_tuple : + {record,?anno('$1'),element(3, '$2'),'$3'}. list -> '[' ']' : {nil,?anno('$1')}. list -> '[' expr tail : {cons,?anno('$1'),'$2','$3'}. @@ -397,6 +448,10 @@ case_expr -> 'case' expr 'of' cr_clauses 'end' : cr_clauses -> cr_clause : ['$1']. cr_clauses -> cr_clause ';' cr_clauses : ['$1' | '$3']. +%% FIXME: merl in syntax_tools depends on patterns in a 'case' being +%% full expressions. Therefore, we can't use pat_expr here. There +%% should be a better way. + cr_clause -> expr clause_guard clause_body : {clause,?anno('$1'),['$1'],'$2','$3'}. @@ -424,11 +479,11 @@ integer_or_var -> var : '$1'. fun_clauses -> fun_clause : ['$1']. fun_clauses -> fun_clause ';' fun_clauses : ['$1' | '$3']. -fun_clause -> argument_list clause_guard clause_body : +fun_clause -> pat_argument_list clause_guard clause_body : {Args,Anno} = '$1', {clause,Anno,'fun',Args,'$2','$3'}. -fun_clause -> var argument_list clause_guard clause_body : +fun_clause -> var pat_argument_list clause_guard clause_body : {clause,element(2, '$1'),element(3, '$1'),element(1, '$2'),'$3','$4'}. try_expr -> 'try' exprs 'of' cr_clauses try_catch : @@ -446,24 +501,31 @@ try_catch -> 'after' exprs 'end' : try_clauses -> try_clause : ['$1']. try_clauses -> try_clause ';' try_clauses : ['$1' | '$3']. -try_clause -> expr clause_guard clause_body : +try_clause -> pat_expr clause_guard clause_body : A = ?anno('$1'), {clause,A,[{tuple,A,[{atom,A,throw},'$1',{var,A,'_'}]}],'$2','$3'}. -try_clause -> atom ':' expr clause_guard clause_body : +try_clause -> atom ':' pat_expr try_opt_stacktrace clause_guard clause_body : A = ?anno('$1'), - {clause,A,[{tuple,A,['$1','$3',{var,A,'_'}]}],'$4','$5'}. -try_clause -> var ':' expr clause_guard clause_body : + {clause,A,[{tuple,A,['$1','$3',{var,A,'$4'}]}],'$5','$6'}. +try_clause -> var ':' pat_expr try_opt_stacktrace clause_guard clause_body : A = ?anno('$1'), - {clause,A,[{tuple,A,['$1','$3',{var,A,'_'}]}],'$4','$5'}. + {clause,A,[{tuple,A,['$1','$3',{var,A,'$4'}]}],'$5','$6'}. +try_opt_stacktrace -> ':' var : element(3, '$2'). +try_opt_stacktrace -> '$empty' : '_'. argument_list -> '(' ')' : {[],?anno('$1')}. argument_list -> '(' exprs ')' : {'$2',?anno('$1')}. +pat_argument_list -> '(' ')' : {[],?anno('$1')}. +pat_argument_list -> '(' pat_exprs ')' : {'$2',?anno('$1')}. exprs -> expr : ['$1']. exprs -> expr ',' exprs : ['$1' | '$3']. +pat_exprs -> pat_expr : ['$1']. +pat_exprs -> pat_expr ',' pat_exprs : ['$1' | '$3']. + guard -> exprs : ['$1']. guard -> exprs ';' guard : ['$1'|'$3']. diff --git a/lib/stdlib/src/ets.erl b/lib/stdlib/src/ets.erl index 4858c8d13c..b6548626f3 100644 --- a/lib/stdlib/src/ets.erl +++ b/lib/stdlib/src/ets.erl @@ -1700,6 +1700,8 @@ choice(Height, Width, P, Mode, Tab, Key, Turn, Opos) -> io:format("~ts\n", [ErrorString]), choice(Height, Width, P, Mode, Tab, Key, Turn, Opos) end; + eof -> + ok; _ -> choice(Height, Width, P, Mode, Tab, Key, Turn, Opos) end. diff --git a/lib/stdlib/src/filename.erl b/lib/stdlib/src/filename.erl index 73eccb226e..a322bd002d 100644 --- a/lib/stdlib/src/filename.erl +++ b/lib/stdlib/src/filename.erl @@ -34,6 +34,38 @@ %% we flatten the arguments immediately on function entry as that makes %% it easier to ensure that the code works. +%% +%% *** Requirements on Raw Filename Format *** +%% +%% These requirements are due to the 'filename' module +%% in stdlib. This since it is documented that it +%% should be able to operate on raw filenames as well +%% as ordinary filenames. +%% +%% A raw filename *must* be a byte sequence where: +%% 1. Codepoints 0-127 (7-bit ascii) *must* be encoded +%% as a byte with the corresponding value. That is, +%% the most significant bit in the byte encoding the +%% codepoint is never set. +%% 2. Codepoints greater than 127 *must* be encoded +%% with the most significant bit set in *every* byte +%% encoding it. +%% +%% Latin1 and UTF-8 meet these requirements while +%% UTF-16 and UTF-32 don't. +%% +%% On Windows filenames are natively stored as malformed +%% UTF-16LE (lonely surrogates may appear). A more correct +%% description than UTF-16 would be an array of 16-bit +%% words... In order to meet the requirements of the +%% raw file format we convert the malformed UTF-16LE to +%% malformed UTF-8 which meet the requirements. +%% +%% Note that these requirements are today only OTP +%% internal (erts-stdlib internal) requirements that +%% could be changed. +%% + -export([absname/1, absname/2, absname_join/2, basename/1, basename/2, dirname/1, extension/1, join/1, join/2, pathtype/1, @@ -440,6 +472,10 @@ join(Name1, Name2) when is_atom(Name2) -> join1([UcLetter, $:|Rest], RelativeName, [], win32) when is_integer(UcLetter), UcLetter >= $A, UcLetter =< $Z -> join1(Rest, RelativeName, [$:, UcLetter+$a-$A], win32); +join1([$\\,$\\|Rest], RelativeName, [], win32) -> + join1([$/,$/|Rest], RelativeName, [], win32); +join1([$/,$/|Rest], RelativeName, [], win32) -> + join1(Rest, RelativeName, [$/,$/], win32); join1([$\\|Rest], RelativeName, Result, win32) -> join1([$/|Rest], RelativeName, Result, win32); join1([$/|Rest], RelativeName, [$., $/|Result], OsType) -> @@ -468,6 +504,10 @@ join1([Atom|Rest], RelativeName, Result, OsType) when is_atom(Atom) -> join1b(<<UcLetter, $:, Rest/binary>>, RelativeName, [], win32) when is_integer(UcLetter), UcLetter >= $A, UcLetter =< $Z -> join1b(Rest, RelativeName, [$:, UcLetter+$a-$A], win32); +join1b(<<$\\,$\\,Rest/binary>>, RelativeName, [], win32) -> + join1b(<<$/,$/,Rest/binary>>, RelativeName, [], win32); +join1b(<<$/,$/,Rest/binary>>, RelativeName, [], win32) -> + join1b(Rest, RelativeName, [$/,$/], win32); join1b(<<$\\,Rest/binary>>, RelativeName, Result, win32) -> join1b(<<$/,Rest/binary>>, RelativeName, Result, win32); join1b(<<$/,Rest/binary>>, RelativeName, [$., $/|Result], OsType) -> @@ -478,6 +518,8 @@ join1b(<<>>, <<>>, Result, OsType) -> list_to_binary(maybe_remove_dirsep(Result, OsType)); join1b(<<>>, RelativeName, [$:|Rest], win32) -> join1b(RelativeName, <<>>, [$:|Rest], win32); +join1b(<<>>, RelativeName, [$/,$/|Result], win32) -> + join1b(RelativeName, <<>>, [$/,$/|Result], win32); join1b(<<>>, RelativeName, [$/|Result], OsType) -> join1b(RelativeName, <<>>, [$/|Result], OsType); join1b(<<>>, RelativeName, [$., $/|Result], OsType) -> @@ -491,6 +533,8 @@ maybe_remove_dirsep([$/, $:, Letter], win32) -> [Letter, $:, $/]; maybe_remove_dirsep([$/], _) -> [$/]; +maybe_remove_dirsep([$/,$/], win32) -> + [$/,$/]; maybe_remove_dirsep([$/|Name], _) -> lists:reverse(Name); maybe_remove_dirsep(Name, _) -> @@ -680,6 +724,9 @@ win32_splitb(<<Letter0,$:,Rest/binary>>) when ?IS_DRIVELETTER(Letter0) -> Letter = fix_driveletter(Letter0), L = binary:split(Rest,[<<"/">>,<<"\\">>],[global]), [<<Letter,$:>> | [ X || X <- L, X =/= <<>> ]]; +win32_splitb(<<Slash,Slash,Rest/binary>>) when ((Slash =:= $\\) orelse (Slash =:= $/)) -> + L = binary:split(Rest,[<<"/">>,<<"\\">>],[global]), + [<<"//">> | [ X || X <- L, X =/= <<>> ]]; win32_splitb(<<Slash,Rest/binary>>) when ((Slash =:= $\\) orelse (Slash =:= $/)) -> L = binary:split(Rest,[<<"/">>,<<"\\">>],[global]), [<<$/>> | [ X || X <- L, X =/= <<>> ]]; @@ -691,6 +738,8 @@ win32_splitb(Name) -> unix_split(Name) -> split(Name, [], unix). +win32_split([Slash,Slash|Rest]) when ((Slash =:= $\\) orelse (Slash =:= $/)) -> + split(Rest, [[$/,$/]], win32); win32_split([$\\|Rest]) -> win32_split([$/|Rest]); win32_split([X, $\\|Rest]) when is_integer(X) -> diff --git a/lib/stdlib/src/gen.erl b/lib/stdlib/src/gen.erl index 33af0aed8f..4b1d448487 100644 --- a/lib/stdlib/src/gen.erl +++ b/lib/stdlib/src/gen.erl @@ -49,6 +49,7 @@ | {'logfile', string()}. -type option() :: {'timeout', timeout()} | {'debug', [debug_flag()]} + | {'hibernate_after', timeout()} | {'spawn_opt', [proc_lib:spawn_option()]}. -type options() :: [option()]. diff --git a/lib/stdlib/src/gen_statem.erl b/lib/stdlib/src/gen_statem.erl index 1110d18af6..cd6312855d 100644 --- a/lib/stdlib/src/gen_statem.erl +++ b/lib/stdlib/src/gen_statem.erl @@ -296,7 +296,7 @@ (Reason :: term()). %% Format the callback module state in some sensible that is -%% often condensed way. For StatusOption =:= 'normal' the perferred +%% often condensed way. For StatusOption =:= 'normal' the preferred %% return term is [{data,[{"State",FormattedState}]}], and for %% StatusOption =:= 'terminate' it is just FormattedState. -callback format_status( @@ -510,8 +510,6 @@ call(ServerRef, Request, Timeout) -> parse_timeout(Timeout) -> case Timeout of - {clean_timeout,infinity} -> - {dirty_timeout,infinity}; {clean_timeout,_} -> Timeout; {dirty_timeout,_} -> diff --git a/lib/stdlib/src/io_lib.erl b/lib/stdlib/src/io_lib.erl index 9d447418f8..3c8430b820 100644 --- a/lib/stdlib/src/io_lib.erl +++ b/lib/stdlib/src/io_lib.erl @@ -963,7 +963,18 @@ limit_tail(Other, D) -> %% maps:from_list() creates a map with the same internal ordering of %% the selected associations as in Map. limit_map(Map, D) -> - maps:from_list(erts_internal:maps_to_list(Map, D)). + limit_map(maps:iterator(Map), D, []). + +limit_map(_I, 0, Acc) -> + maps:from_list(Acc); +limit_map(I, D, Acc) -> + case maps:next(I) of + {K, V, NextI} -> + limit_map(NextI, D-1, [{K,V} | Acc]); + none -> + maps:from_list(Acc) + end. + %% maps:from_list(limit_map_body(erts_internal:maps_to_list(Map, D), D)). %% limit_map_body(_, 0) -> [{'...', '...'}]; diff --git a/lib/stdlib/src/io_lib_pretty.erl b/lib/stdlib/src/io_lib_pretty.erl index 505613b80e..89e1931d2d 100644 --- a/lib/stdlib/src/io_lib_pretty.erl +++ b/lib/stdlib/src/io_lib_pretty.erl @@ -478,9 +478,19 @@ print_length(Term, _D, _RF, _Enc, _Str) -> print_length_map(_Map, 1, _RF, _Enc, _Str) -> {"#{...}", 6}; print_length_map(Map, D, RF, Enc, Str) when is_map(Map) -> - Pairs = print_length_map_pairs(erts_internal:maps_to_list(Map, D), D, RF, Enc, Str), + Pairs = print_length_map_pairs(limit_map(maps:iterator(Map), D, []), D, RF, Enc, Str), {{map, Pairs}, list_length(Pairs, 3)}. +limit_map(_I, 0, Acc) -> + Acc; +limit_map(I, D, Acc) -> + case maps:next(I) of + {K, V, NextI} -> + limit_map(NextI, D-1, [{K,V} | Acc]); + none -> + Acc + end. + print_length_map_pairs([], _D, _RF, _Enc, _Str) -> []; print_length_map_pairs(_Pairs, 1, _RF, _Enc, _Str) -> diff --git a/lib/stdlib/src/maps.erl b/lib/stdlib/src/maps.erl index 5dafdb282a..a13f340709 100644 --- a/lib/stdlib/src/maps.erl +++ b/lib/stdlib/src/maps.erl @@ -23,7 +23,8 @@ -export([get/3, filter/2,fold/3, map/2, size/1, update_with/3, update_with/4, - without/2, with/2]). + without/2, with/2, + iterator/1, next/1]). %% BIFs -export([get/2, find/2, from_list/1, @@ -31,6 +32,15 @@ new/0, put/3, remove/2, take/2, to_list/1, update/3, values/1]). +-opaque iterator() :: {term(), term(), iterator()} + | none | nonempty_improper_list(integer(),map()). + +-export_type([iterator/0]). + +-dialyzer({no_improper_lists, iterator/1}). + +-define(IS_ITERATOR(I), is_tuple(I) andalso tuple_size(I) == 3; I == none; is_integer(hd(I)) andalso is_map(tl(I))). + %% Shadowed by erl_bif_types: maps:get/2 -spec get(Key,Map) -> Value when Key :: term(), @@ -39,7 +49,6 @@ get(_,_) -> erlang:nif_error(undef). - -spec find(Key,Map) -> {ok, Value} | error when Key :: term(), Map :: map(), @@ -114,14 +123,20 @@ remove(_,_) -> erlang:nif_error(undef). take(_,_) -> erlang:nif_error(undef). -%% Shadowed by erl_bif_types: maps:to_list/1 -spec to_list(Map) -> [{Key,Value}] when Map :: map(), Key :: term(), Value :: term(). -to_list(_) -> erlang:nif_error(undef). +to_list(Map) when is_map(Map) -> + to_list_internal(erts_internal:map_next(0, Map, [])); +to_list(Map) -> + erlang:error({badmap,Map},[Map]). +to_list_internal([Iter, Map | Acc]) when is_integer(Iter) -> + to_list_internal(erts_internal:map_next(Iter, Map, Acc)); +to_list_internal(Acc) -> + Acc. %% Shadowed by erl_bif_types: maps:update/3 -spec update(Key,Value,Map1) -> Map2 when @@ -192,47 +207,80 @@ get(Key,Map,Default) -> erlang:error({badmap,Map},[Key,Map,Default]). --spec filter(Pred,Map1) -> Map2 when +-spec filter(Pred,MapOrIter) -> Map when Pred :: fun((Key, Value) -> boolean()), Key :: term(), Value :: term(), - Map1 :: map(), - Map2 :: map(). + MapOrIter :: map() | iterator(), + Map :: map(). filter(Pred,Map) when is_function(Pred,2), is_map(Map) -> - maps:from_list([{K,V}||{K,V}<-maps:to_list(Map),Pred(K,V)]); + maps:from_list(filter_1(Pred, iterator(Map))); +filter(Pred,Iterator) when is_function(Pred,2), ?IS_ITERATOR(Iterator) -> + maps:from_list(filter_1(Pred, Iterator)); filter(Pred,Map) -> erlang:error(error_type(Map),[Pred,Map]). - --spec fold(Fun,Init,Map) -> Acc when +filter_1(Pred, Iter) -> + case next(Iter) of + {K, V, NextIter} -> + case Pred(K,V) of + true -> + [{K,V} | filter_1(Pred, NextIter)]; + false -> + filter_1(Pred, NextIter) + end; + none -> + [] + end. + +-spec fold(Fun,Init,MapOrIter) -> Acc when Fun :: fun((K, V, AccIn) -> AccOut), Init :: term(), Acc :: term(), AccIn :: term(), AccOut :: term(), - Map :: map(), + MapOrIter :: map() | iterator(), K :: term(), V :: term(). fold(Fun,Init,Map) when is_function(Fun,3), is_map(Map) -> - lists:foldl(fun({K,V},A) -> Fun(K,V,A) end,Init,maps:to_list(Map)); + fold_1(Fun,Init,iterator(Map)); +fold(Fun,Init,Iterator) when is_function(Fun,3), ?IS_ITERATOR(Iterator) -> + fold_1(Fun,Init,Iterator); fold(Fun,Init,Map) -> erlang:error(error_type(Map),[Fun,Init,Map]). --spec map(Fun,Map1) -> Map2 when +fold_1(Fun, Acc, Iter) -> + case next(Iter) of + {K, V, NextIter} -> + fold_1(Fun, Fun(K,V,Acc), NextIter); + none -> + Acc + end. + +-spec map(Fun,MapOrIter) -> Map when Fun :: fun((K, V1) -> V2), - Map1 :: map(), - Map2 :: map(), + MapOrIter :: map() | iterator(), + Map :: map(), K :: term(), V1 :: term(), V2 :: term(). map(Fun,Map) when is_function(Fun, 2), is_map(Map) -> - maps:from_list([{K,Fun(K,V)}||{K,V}<-maps:to_list(Map)]); + maps:from_list(map_1(Fun, iterator(Map))); +map(Fun,Iterator) when is_function(Fun, 2), ?IS_ITERATOR(Iterator) -> + maps:from_list(map_1(Fun, Iterator)); map(Fun,Map) -> erlang:error(error_type(Map),[Fun,Map]). +map_1(Fun, Iter) -> + case next(Iter) of + {K, V, NextIter} -> + [{K, Fun(K, V)} | map_1(Fun, NextIter)]; + none -> + [] + end. -spec size(Map) -> non_neg_integer() when Map :: map(). @@ -242,6 +290,26 @@ size(Map) when is_map(Map) -> size(Val) -> erlang:error({badmap,Val},[Val]). +-spec iterator(Map) -> Iterator when + Map :: map(), + Iterator :: iterator(). + +iterator(M) when is_map(M) -> [0 | M]; +iterator(M) -> erlang:error({badmap, M}, [M]). + +-spec next(Iterator) -> {Key, Value, NextIterator} | 'none' when + Iterator :: iterator(), + Key :: term(), + Value :: term(), + NextIterator :: iterator(). +next({K, V, I}) -> + {K, V, I}; +next([Path | Map]) when is_integer(Path), is_map(Map) -> + erts_internal:map_next(Path, Map, iterator); +next(none) -> + none; +next(Iter) -> + erlang:error(badarg, [Iter]). -spec without(Ks,Map1) -> Map2 when Ks :: [K], @@ -250,11 +318,10 @@ size(Val) -> K :: term(). without(Ks,M) when is_list(Ks), is_map(M) -> - lists:foldl(fun(K, M1) -> ?MODULE:remove(K, M1) end, M, Ks); + lists:foldl(fun(K, M1) -> maps:remove(K, M1) end, M, Ks); without(Ks,M) -> erlang:error(error_type(M),[Ks,M]). - -spec with(Ks, Map1) -> Map2 when Ks :: [K], Map1 :: map(), @@ -263,17 +330,17 @@ without(Ks,M) -> with(Ks,Map1) when is_list(Ks), is_map(Map1) -> Fun = fun(K, List) -> - case ?MODULE:find(K, Map1) of - {ok, V} -> - [{K, V} | List]; - error -> - List - end - end, - ?MODULE:from_list(lists:foldl(Fun, [], Ks)); + case maps:find(K, Map1) of + {ok, V} -> + [{K, V} | List]; + error -> + List + end + end, + maps:from_list(lists:foldl(Fun, [], Ks)); with(Ks,M) -> erlang:error(error_type(M),[Ks,M]). -error_type(M) when is_map(M) -> badarg; +error_type(M) when is_map(M); ?IS_ITERATOR(M) -> badarg; error_type(V) -> {badmap, V}. diff --git a/lib/stdlib/src/stdlib.app.src b/lib/stdlib/src/stdlib.app.src index 41c89270aa..5fb48acfab 100644 --- a/lib/stdlib/src/stdlib.app.src +++ b/lib/stdlib/src/stdlib.app.src @@ -101,13 +101,14 @@ timer, unicode, unicode_util, + uri_string, win32reg, zip]}, {registered,[timer_server,rsh_starter,take_over_monitor,pool_master, dets]}, {applications, [kernel]}, {env, []}, - {runtime_dependencies, ["sasl-3.0","kernel-5.4.1","erts-9.1.1","crypto-3.3", + {runtime_dependencies, ["sasl-3.0","kernel-6.0","erts-10.0","crypto-3.3", "compiler-5.0"]} ]}. diff --git a/lib/stdlib/src/string.erl b/lib/stdlib/src/string.erl index 5a4d2df2a6..e01bb7d85e 100644 --- a/lib/stdlib/src/string.erl +++ b/lib/stdlib/src/string.erl @@ -74,15 +74,16 @@ -export([to_upper/1, to_lower/1]). %% -import(lists,[member/2]). - -compile({no_auto_import,[length/1]}). +-compile({inline, [btoken/2, rev/1, append/2, stack/2, search_compile/1]}). +-define(ASCII_LIST(CP1,CP2), CP1 < 256, CP2 < 256, CP1 =/= $\r). -export_type([grapheme_cluster/0]). -type grapheme_cluster() :: char() | [char()]. -type direction() :: 'leading' | 'trailing'. --dialyzer({no_improper_lists, stack/2}). +-dialyzer({no_improper_lists, [stack/2, length_b/3]}). %%% BIFs internal (not documented) should not to be used outside of this module %%% May be removed -export([list_to_float/1, list_to_integer/1]). @@ -127,8 +128,10 @@ is_empty(_) -> false. %% Count the number of grapheme clusters in chardata -spec length(String::unicode:chardata()) -> non_neg_integer(). +length(<<CP1/utf8, Bin/binary>>) -> + length_b(Bin, CP1, 0); length(CD) -> - length_1(unicode_util:gc(CD), 0). + length_1(CD, 0). %% Convert a string to a list of grapheme clusters -spec to_graphemes(String::unicode:chardata()) -> [grapheme_cluster()]. @@ -176,6 +179,8 @@ equal(A, B, true, Norm) -> %% Reverse grapheme clusters -spec reverse(String::unicode:chardata()) -> [grapheme_cluster()]. +reverse(<<CP1/utf8, Rest/binary>>) -> + reverse_b(Rest, CP1, []); reverse(CD) -> reverse_1(CD, []). @@ -186,7 +191,10 @@ reverse(CD) -> Start :: non_neg_integer(), Slice :: unicode:chardata(). slice(CD, N) when is_integer(N), N >= 0 -> - slice_l(CD, N, is_binary(CD)). + case slice_l0(CD, N) of + [] when is_binary(CD) -> <<>>; + Res -> Res + end. -spec slice(String, Start, Length) -> Slice when String::unicode:chardata(), @@ -195,9 +203,15 @@ slice(CD, N) when is_integer(N), N >= 0 -> Slice :: unicode:chardata(). slice(CD, N, Length) when is_integer(N), N >= 0, is_integer(Length), Length > 0 -> - slice_trail(slice_l(CD, N, is_binary(CD)), Length); + case slice_l0(CD, N) of + [] when is_binary(CD) -> <<>>; + L -> slice_trail(L, Length) + end; slice(CD, N, infinity) -> - slice_l(CD, N, is_binary(CD)); + case slice_l0(CD, N) of + [] when is_binary(CD) -> <<>>; + Res -> Res + end; slice(CD, _, 0) -> case is_binary(CD) of true -> <<>>; @@ -256,18 +270,22 @@ trim(Str, Dir) -> Dir :: direction() | 'both', Characters :: [grapheme_cluster()]. trim(Str, _, []) -> Str; +trim(Str, leading, [Sep]) when is_list(Str), Sep < 256 -> + trim_ls(Str, Sep); trim(Str, leading, Sep) when is_list(Sep) -> - trim_l(Str, search_pattern(Sep)); -trim(Str, trailing, Sep) when is_list(Sep) -> - trim_t(Str, 0, search_pattern(Sep)); -trim(Str, both, Sep0) when is_list(Sep0) -> - Sep = search_pattern(Sep0), - trim_t(trim_l(Str,Sep), 0, Sep). + trim_l(Str, Sep); +trim(Str, trailing, [Sep]) when is_list(Str), Sep < 256 -> + trim_ts(Str, Sep); +trim(Str, trailing, Seps0) when is_list(Seps0) -> + Seps = search_pattern(Seps0), + trim_t(Str, 0, Seps); +trim(Str, both, Sep) when is_list(Sep) -> + trim(trim(Str,leading,Sep), trailing, Sep). %% Delete trailing newlines or \r\n -spec chomp(String::unicode:chardata()) -> unicode:chardata(). chomp(Str) -> - trim_t(Str,0, {[[$\r,$\n],$\n], [$\r,$\n], [<<$\r>>,<<$\n>>]}). + trim(Str, trailing, [[$\r,$\n],$\n]). %% Split String into two parts where the leading part consists of Characters -spec take(String, Characters) -> {Leading, Trailing} when @@ -300,8 +318,7 @@ take(Str, [], Complement, Dir) -> {true, leading} -> {Str, Empty}; {true, trailing} -> {Empty, Str} end; -take(Str, Sep0, false, leading) -> - Sep = search_pattern(Sep0), +take(Str, Sep, false, leading) -> take_l(Str, Sep, []); take(Str, Sep0, true, leading) -> Sep = search_pattern(Sep0), @@ -461,6 +478,7 @@ replace(String, SearchPattern, Replacement, Where) -> SeparatorList::[grapheme_cluster()]) -> [unicode:chardata()]. lexemes([], _) -> []; +lexemes(Str, []) -> [Str]; lexemes(Str, Seps0) when is_list(Seps0) -> Seps = search_pattern(Seps0), lexemes_m(Str, Seps, []). @@ -494,13 +512,13 @@ find(String, SearchPattern, leading) -> find(String, SearchPattern, trailing) -> find_r(String, unicode:characters_to_list(SearchPattern), nomatch). -%% Fetch first codepoint and return rest in tail +%% Fetch first grapheme cluster and return rest in tail -spec next_grapheme(String::unicode:chardata()) -> maybe_improper_list(grapheme_cluster(),unicode:chardata()) | {error,unicode:chardata()}. next_grapheme(CD) -> unicode_util:gc(CD). -%% Fetch first grapheme cluster and return rest in tail +%% Fetch first codepoint and return rest in tail -spec next_codepoint(String::unicode:chardata()) -> maybe_improper_list(char(),unicode:chardata()) | {error,unicode:chardata()}. @@ -508,10 +526,23 @@ next_codepoint(CD) -> unicode_util:cp(CD). %% Internals -length_1([_|Rest], N) -> - length_1(unicode_util:gc(Rest), N+1); -length_1([], N) -> - N. +length_1([CP1|[CP2|_]=Cont], N) when ?ASCII_LIST(CP1,CP2) -> + length_1(Cont, N+1); +length_1(Str, N) -> + case unicode_util:gc(Str) of + [] -> N; + [_|Rest] -> length_1(Rest, N+1) + end. + +length_b(<<CP2/utf8, Rest/binary>>, CP1, N) + when ?ASCII_LIST(CP1,CP2) -> + length_b(Rest, CP2, N+1); +length_b(Bin0, CP1, N) -> + [_|Bin1] = unicode_util:gc([CP1|Bin0]), + case unicode_util:cp(Bin1) of + [] -> N+1; + [CP3|Bin] -> length_b(Bin, CP3, N+1) + end. equal_1([A|AR], [B|BR]) when is_integer(A), is_integer(B) -> A =:= B andalso equal_1(AR, BR); @@ -550,29 +581,66 @@ equal_norm_nocase(A0, B0, Norm) -> {L1,L2} when is_list(L1), is_list(L2) -> false end. +reverse_1([CP1|[CP2|_]=Cont], Acc) when ?ASCII_LIST(CP1,CP2) -> + reverse_1(Cont, [CP1|Acc]); reverse_1(CD, Acc) -> case unicode_util:gc(CD) of [GC|Rest] -> reverse_1(Rest, [GC|Acc]); [] -> Acc end. -slice_l(CD, N, Binary) when N > 0 -> +reverse_b(<<CP2/utf8, Rest/binary>>, CP1, Acc) + when ?ASCII_LIST(CP1,CP2) -> + reverse_b(Rest, CP2, [CP1|Acc]); +reverse_b(Bin0, CP1, Acc) -> + [GC|Bin1] = unicode_util:gc([CP1|Bin0]), + case unicode_util:cp(Bin1) of + [] -> [GC|Acc]; + [CP3|Bin] -> reverse_b(Bin, CP3, [GC|Acc]) + end. + +slice_l0(<<CP1/utf8, Bin/binary>>, N) when N > 0 -> + slice_lb(Bin, CP1, N); +slice_l0(L, N) -> + slice_l(L, N). + +slice_l([CP1|[CP2|_]=Cont], N) when ?ASCII_LIST(CP1,CP2),N > 0 -> + slice_l(Cont, N-1); +slice_l(CD, N) when N > 0 -> case unicode_util:gc(CD) of - [_|Cont] -> slice_l(Cont, N-1, Binary); - [] when Binary -> <<>>; + [_|Cont] -> slice_l(Cont, N-1); [] -> [] end; -slice_l(Cont, 0, Binary) -> - case is_empty(Cont) of - true when Binary -> <<>>; - _ -> Cont +slice_l(Cont, 0) -> + Cont. + +slice_lb(<<CP2/utf8, Bin/binary>>, CP1, N) when ?ASCII_LIST(CP1,CP2), N > 1 -> + slice_lb(Bin, CP2, N-1); +slice_lb(Bin, CP1, N) -> + [_|Rest] = unicode_util:gc([CP1|Bin]), + if N > 1 -> + case unicode_util:cp(Rest) of + [CP2|Cont] -> slice_lb(Cont, CP2, N-1); + [] -> <<>> + end; + N =:= 1 -> + Rest end. +slice_trail(Orig, N) when is_binary(Orig) -> + case Orig of + <<CP1/utf8, Bin/binary>> when N > 0 -> + Length = slice_bin(Bin, CP1, N), + Sz = byte_size(Orig) - Length, + <<Keep:Sz/binary, _/binary>> = Orig, + Keep; + _ -> <<>> + end; slice_trail(CD, N) when is_list(CD) -> - slice_list(CD, N); -slice_trail(CD, N) when is_binary(CD) -> - slice_bin(CD, N, CD). + slice_list(CD, N). +slice_list([CP1|[CP2|_]=Cont], N) when ?ASCII_LIST(CP1,CP2),N > 0 -> + [CP1|slice_list(Cont, N-1)]; slice_list(CD, N) when N > 0 -> case unicode_util:gc(CD) of [GC|Cont] -> append(GC, slice_list(Cont, N-1)); @@ -581,17 +649,16 @@ slice_list(CD, N) when N > 0 -> slice_list(_, 0) -> []. -slice_bin(CD, N, Orig) when N > 0 -> - case unicode_util:gc(CD) of - [_|Cont] -> slice_bin(Cont, N-1, Orig); - [] -> Orig +slice_bin(<<CP2/utf8, Bin/binary>>, CP1, N) when ?ASCII_LIST(CP1,CP2), N > 0 -> + slice_bin(Bin, CP2, N-1); +slice_bin(CD, CP1, N) when N > 0 -> + [_|Bin] = unicode_util:gc([CP1|CD]), + case unicode_util:cp(Bin) of + [CP2|Cont] -> slice_bin(Cont, CP2, N-1); + [] -> 0 end; -slice_bin([], 0, Orig) -> - Orig; -slice_bin(CD, 0, Orig) -> - Sz = byte_size(Orig) - byte_size(CD), - <<Keep:Sz/binary, _/binary>> = Orig, - Keep. +slice_bin(CD, CP1, 0) -> + byte_size(CD)+byte_size(<<CP1/utf8>>). uppercase_list(CPs0) -> case unicode_util:uppercase(CPs0) of @@ -641,16 +708,31 @@ casefold_bin(CPs0, Acc) -> [] -> Acc end. - +%% Fast path for ascii searching for one character in lists +trim_ls([CP1|[CP2|_]=Cont]=Str, Sep) + when ?ASCII_LIST(CP1,CP2) -> + case Sep of + CP1 -> trim_ls(Cont, Sep); + _ -> Str + end; +trim_ls(Str, Sep) -> + trim_l(Str, [Sep]). + +trim_l([CP1|[CP2|_]=Cont]=Str, Sep) + when ?ASCII_LIST(CP1,CP2) -> + case lists:member(CP1, Sep) of + true -> trim_l(Cont, Sep); + false -> Str + end; trim_l([Bin|Cont0], Sep) when is_binary(Bin) -> case bin_search_inv(Bin, Cont0, Sep) of {nomatch, Cont} -> trim_l(Cont, Sep); Keep -> Keep end; -trim_l(Str, {GCs, _, _}=Sep) when is_list(Str) -> +trim_l(Str, Sep) when is_list(Str) -> case unicode_util:gc(Str) of [C|Cs] -> - case lists:member(C, GCs) of + case lists:member(C, Sep) of true -> trim_l(Cs, Sep); false -> Str end; @@ -662,15 +744,51 @@ trim_l(Bin, Sep) when is_binary(Bin) -> [Keep] -> Keep end. -trim_t([Bin|Cont0], N, Sep) when is_binary(Bin) -> +%% Fast path for ascii searching for one character in lists +trim_ts([Sep|Cs1]=Str, Sep) -> + case Cs1 of + [] -> []; + [CP2|_] when ?ASCII_LIST(Sep,CP2) -> + Tail = trim_ts(Cs1, Sep), + case is_empty(Tail) of + true -> []; + false -> [Sep|Tail] + end; + _ -> + trim_t(Str, 0, search_pattern([Sep])) + end; +trim_ts([CP|Cont],Sep) when is_integer(CP) -> + [CP|trim_ts(Cont, Sep)]; +trim_ts(Str, Sep) -> + trim_t(Str, 0, search_pattern([Sep])). + +trim_t([CP1|Cont]=Cs0, _, {GCs,CPs,_}=Seps) when is_integer(CP1) -> + case lists:member(CP1, CPs) of + true -> + [GC|Cs1] = unicode_util:gc(Cs0), + case lists:member(GC, GCs) of + true -> + Tail = trim_t(Cs1, 0, Seps), + case is_empty(Tail) of + true -> []; + false -> append(GC,Tail) + end; + false -> + append(GC,trim_t(Cs1, 0, Seps)) + end; + false -> + [CP1|trim_t(Cont, 0, Seps)] + end; +trim_t([Bin|Cont0], N, {GCs,_,_}=Seps0) when is_binary(Bin) -> <<_:N/binary, Rest/binary>> = Bin, - case bin_search(Rest, Cont0, Sep) of + Seps = search_compile(Seps0), + case bin_search(Rest, Cont0, Seps) of {nomatch,_} -> - stack(Bin, trim_t(Cont0, 0, Sep)); + stack(Bin, trim_t(Cont0, 0, Seps)); [SepStart|Cont1] -> - case bin_search_inv(SepStart, Cont1, Sep) of + case bin_search_inv(SepStart, Cont1, GCs) of {nomatch, Cont} -> - Tail = trim_t(Cont, 0, Sep), + Tail = trim_t(Cont, 0, Seps), case is_empty(Tail) of true -> KeepSz = byte_size(Bin) - byte_size(SepStart), @@ -682,67 +800,69 @@ trim_t([Bin|Cont0], N, Sep) when is_binary(Bin) -> end; [NonSep|Cont] when is_binary(NonSep) -> KeepSz = byte_size(Bin) - byte_size(NonSep), - trim_t([Bin|Cont], KeepSz, Sep) + trim_t([Bin|Cont], KeepSz, Seps) end end; -trim_t(Str, 0, {GCs,CPs,_}=Sep) when is_list(Str) -> - case unicode_util:cp(Str) of - [CP|Cs] -> - case lists:member(CP, CPs) of +trim_t(Str, 0, {GCs,_,_}=Seps) when is_list(Str) -> + case unicode_util:gc(Str) of + [GC|Cs1] -> + case lists:member(GC, GCs) of true -> - [GC|Cs1] = unicode_util:gc(Str), - case lists:member(GC, GCs) of - true -> - Tail = trim_t(Cs1, 0, Sep), - case is_empty(Tail) of - true -> []; - false -> append(GC,Tail) - end; - false -> - append(GC,trim_t(Cs1, 0, Sep)) + Tail = trim_t(Cs1, 0, Seps), + case is_empty(Tail) of + true -> []; + false -> append(GC,Tail) end; false -> - append(CP,trim_t(Cs, 0, Sep)) + append(GC,trim_t(Cs1, 0, Seps)) end; [] -> [] end; -trim_t(Bin, N, Sep) when is_binary(Bin) -> +trim_t(Bin, N, {GCs,_,_}=Seps0) when is_binary(Bin) -> <<_:N/binary, Rest/binary>> = Bin, - case bin_search(Rest, Sep) of + Seps = search_compile(Seps0), + case bin_search(Rest, [], Seps) of {nomatch,_} -> Bin; [SepStart] -> - case bin_search_inv(SepStart, [], Sep) of + case bin_search_inv(SepStart, [], GCs) of {nomatch,_} -> KeepSz = byte_size(Bin) - byte_size(SepStart), <<Keep:KeepSz/binary, _/binary>> = Bin, Keep; [NonSep] -> KeepSz = byte_size(Bin) - byte_size(NonSep), - trim_t(Bin, KeepSz, Sep) + trim_t(Bin, KeepSz, Seps) end end. -take_l([Bin|Cont0], Sep, Acc) when is_binary(Bin) -> - case bin_search_inv(Bin, Cont0, Sep) of + +take_l([CP1|[CP2|_]=Cont]=Str, Seps, Acc) + when ?ASCII_LIST(CP1,CP2) -> + case lists:member(CP1, Seps) of + true -> take_l(Cont, Seps, [CP1|Acc]); + false -> {rev(Acc), Str} + end; +take_l([Bin|Cont0], Seps, Acc) when is_binary(Bin) -> + case bin_search_inv(Bin, Cont0, Seps) of {nomatch, Cont} -> Used = cp_prefix(Cont0, Cont), - take_l(Cont, Sep, [unicode:characters_to_binary([Bin|Used])|Acc]); + take_l(Cont, Seps, [unicode:characters_to_binary([Bin|Used])|Acc]); [Bin1|_]=After when is_binary(Bin1) -> First = byte_size(Bin) - byte_size(Bin1), <<Keep:First/binary, _/binary>> = Bin, {btoken(Keep,Acc), After} end; -take_l(Str, {GCs, _, _}=Sep, Acc) when is_list(Str) -> +take_l(Str, Seps, Acc) when is_list(Str) -> case unicode_util:gc(Str) of [C|Cs] -> - case lists:member(C, GCs) of - true -> take_l(Cs, Sep, append(rev(C),Acc)); + case lists:member(C, Seps) of + true -> take_l(Cs, Seps, append(rev(C),Acc)); false -> {rev(Acc), Str} end; [] -> {rev(Acc), []} end; -take_l(Bin, Sep, Acc) when is_binary(Bin) -> - case bin_search_inv(Bin, [], Sep) of +take_l(Bin, Seps, Acc) when is_binary(Bin) -> + case bin_search_inv(Bin, [], Seps) of {nomatch,_} -> {btoken(Bin, Acc), <<>>}; [After] -> @@ -751,27 +871,41 @@ take_l(Bin, Sep, Acc) when is_binary(Bin) -> {btoken(Keep, Acc), After} end. -take_lc([Bin|Cont0], Sep, Acc) when is_binary(Bin) -> - case bin_search(Bin, Cont0, Sep) of + +take_lc([CP1|Cont]=Str0, {GCs,CPs,_}=Seps, Acc) when is_integer(CP1) -> + case lists:member(CP1, CPs) of + true -> + [GC|Str] = unicode_util:gc(Str0), + case lists:member(GC, GCs) of + false -> take_lc(Str, Seps, append(rev(GC),Acc)); + true -> {rev(Acc), Str0} + end; + false -> + take_lc(Cont, Seps, append(CP1,Acc)) + end; +take_lc([Bin|Cont0], Seps0, Acc) when is_binary(Bin) -> + Seps = search_compile(Seps0), + case bin_search(Bin, Cont0, Seps) of {nomatch, Cont} -> Used = cp_prefix(Cont0, Cont), - take_lc(Cont, Sep, [unicode:characters_to_binary([Bin|Used])|Acc]); + take_lc(Cont, Seps, [unicode:characters_to_binary([Bin|Used])|Acc]); [Bin1|_]=After when is_binary(Bin1) -> First = byte_size(Bin) - byte_size(Bin1), <<Keep:First/binary, _/binary>> = Bin, {btoken(Keep,Acc), After} end; -take_lc(Str, {GCs, _, _}=Sep, Acc) when is_list(Str) -> +take_lc(Str, {GCs,_,_}=Seps, Acc) when is_list(Str) -> case unicode_util:gc(Str) of [C|Cs] -> case lists:member(C, GCs) of - false -> take_lc(Cs, Sep, append(rev(C),Acc)); + false -> take_lc(Cs, Seps, append(rev(C),Acc)); true -> {rev(Acc), Str} end; [] -> {rev(Acc), []} end; -take_lc(Bin, Sep, Acc) when is_binary(Bin) -> - case bin_search(Bin, [], Sep) of +take_lc(Bin, Seps0, Acc) when is_binary(Bin) -> + Seps = search_compile(Seps0), + case bin_search(Bin, [], Seps) of {nomatch,_} -> {btoken(Bin, Acc), <<>>}; [After] -> @@ -780,148 +914,192 @@ take_lc(Bin, Sep, Acc) when is_binary(Bin) -> {btoken(Keep, Acc), After} end. -take_t([Bin|Cont0], N, Sep) when is_binary(Bin) -> + +take_t([CP1|Cont]=Str0, _, {GCs,CPs,_}=Seps) when is_integer(CP1) -> + case lists:member(CP1, CPs) of + true -> + [GC|Str] = unicode_util:gc(Str0), + case lists:member(GC, GCs) of + true -> + {Head, Tail} = take_t(Str, 0, Seps), + case is_empty(Head) of + true -> {Head, append(GC,Tail)}; + false -> {append(GC,Head), Tail} + end; + false -> + {Head, Tail} = take_t(Str, 0, Seps), + {append(GC,Head), Tail} + end; + false -> + {Head, Tail} = take_t(Cont, 0, Seps), + {[CP1|Head], Tail} + end; +take_t([Bin|Cont0], N, {GCs,_,_}=Seps0) when is_binary(Bin) -> <<_:N/binary, Rest/binary>> = Bin, - case bin_search(Rest, Cont0, Sep) of + Seps = search_compile(Seps0), + case bin_search(Rest, Cont0, Seps) of {nomatch,Cont} -> Used = cp_prefix(Cont0, Cont), - {Head, Tail} = take_t(Cont, 0, Sep), + {Head, Tail} = take_t(Cont, 0, Seps), {stack(unicode:characters_to_binary([Bin|Used]), Head), Tail}; [SepStart|Cont1] -> - case bin_search_inv(SepStart, Cont1, Sep) of + case bin_search_inv(SepStart, Cont1, GCs) of {nomatch, Cont} -> - {Head, Tail} = take_t(Cont, 0, Sep), + {Head, Tail} = take_t(Cont, 0, Seps), Used = cp_prefix(Cont0, Cont), - case equal(Tail, Cont) of + case is_empty(Head) of true -> KeepSz = byte_size(Bin) - byte_size(SepStart), <<Keep:KeepSz/binary, End/binary>> = Bin, - {stack(Keep,Head), stack(stack(End,Used),Tail)}; + {Keep, stack(stack(End,Used),Tail)}; false -> {stack(unicode:characters_to_binary([Bin|Used]),Head), Tail} end; [NonSep|Cont] when is_binary(NonSep) -> KeepSz = byte_size(Bin) - byte_size(NonSep), - take_t([Bin|Cont], KeepSz, Sep) + take_t([Bin|Cont], KeepSz, Seps) end end; -take_t(Str, 0, {GCs,CPs,_}=Sep) when is_list(Str) -> - case unicode_util:cp(Str) of - [CP|Cs] -> - case lists:member(CP, CPs) of +take_t(Str, 0, {GCs,_,_}=Seps) when is_list(Str) -> + case unicode_util:gc(Str) of + [GC|Cs1] -> + case lists:member(GC, GCs) of true -> - [GC|Cs1] = unicode_util:gc(Str), - case lists:member(GC, GCs) of - true -> - {Head, Tail} = take_t(Cs1, 0, Sep), - case equal(Tail, Cs1) of - true -> {Head, append(GC,Tail)}; - false -> {append(GC,Head), Tail} - end; - false -> - {Head, Tail} = take_t(Cs, 0, Sep), - {append(CP,Head), Tail} + {Head, Tail} = take_t(Cs1, 0, Seps), + case is_empty(Head) of + true -> {Head, append(GC,Tail)}; + false -> {append(GC,Head), Tail} end; false -> - {Head, Tail} = take_t(Cs, 0, Sep), - {append(CP,Head), Tail} + {Head, Tail} = take_t(Cs1, 0, Seps), + {append(GC,Head), Tail} end; [] -> {[],[]} end; -take_t(Bin, N, Sep) when is_binary(Bin) -> +take_t(Bin, N, {GCs,_,_}=Seps0) when is_binary(Bin) -> <<_:N/binary, Rest/binary>> = Bin, - case bin_search(Rest, Sep) of + Seps = search_compile(Seps0), + case bin_search(Rest, [], Seps) of {nomatch,_} -> {Bin, <<>>}; [SepStart] -> - case bin_search_inv(SepStart, [], Sep) of + case bin_search_inv(SepStart, [], GCs) of {nomatch,_} -> KeepSz = byte_size(Bin) - byte_size(SepStart), <<Before:KeepSz/binary, End/binary>> = Bin, {Before, End}; [NonSep] -> KeepSz = byte_size(Bin) - byte_size(NonSep), - take_t(Bin, KeepSz, Sep) + take_t(Bin, KeepSz, Seps) end end. -take_tc([Bin|Cont0], N, Sep) when is_binary(Bin) -> +take_tc([CP1|[CP2|_]=Cont], _, {GCs,_,_}=Seps) when ?ASCII_LIST(CP1,CP2) -> + case lists:member(CP1, GCs) of + false -> + {Head, Tail} = take_tc(Cont, 0, Seps), + case is_empty(Head) of + true -> {Head, append(CP1,Tail)}; + false -> {append(CP1,Head), Tail} + end; + true -> + {Head, Tail} = take_tc(Cont, 0, Seps), + {append(CP1,Head), Tail} + end; +take_tc([Bin|Cont0], N, {GCs,_,_}=Seps0) when is_binary(Bin) -> <<_:N/binary, Rest/binary>> = Bin, - case bin_search_inv(Rest, Cont0, Sep) of + case bin_search_inv(Rest, Cont0, GCs) of {nomatch,Cont} -> Used = cp_prefix(Cont0, Cont), - {Head, Tail} = take_tc(Cont, 0, Sep), + {Head, Tail} = take_tc(Cont, 0, Seps0), {stack(unicode:characters_to_binary([Bin|Used]), Head), Tail}; [SepStart|Cont1] -> - case bin_search(SepStart, Cont1, Sep) of + Seps = search_compile(Seps0), + case bin_search(SepStart, Cont1, Seps) of {nomatch, Cont} -> - {Head, Tail} = take_tc(Cont, 0, Sep), + {Head, Tail} = take_tc(Cont, 0, Seps), Used = cp_prefix(Cont0, Cont), - case equal(Tail, Cont) of + case is_empty(Head) of true -> KeepSz = byte_size(Bin) - byte_size(SepStart), <<Keep:KeepSz/binary, End/binary>> = Bin, - {stack(Keep,Head), stack(stack(End,Used),Tail)}; + {Keep, stack(stack(End,Used),Tail)}; false -> {stack(unicode:characters_to_binary([Bin|Used]),Head), Tail} end; [NonSep|Cont] when is_binary(NonSep) -> KeepSz = byte_size(Bin) - byte_size(NonSep), - take_tc([Bin|Cont], KeepSz, Sep) + take_tc([Bin|Cont], KeepSz, Seps) end end; -take_tc(Str, 0, {GCs,CPs,_}=Sep) when is_list(Str) -> - case unicode_util:cp(Str) of - [CP|Cs] -> - case lists:member(CP, CPs) of - true -> - [GC|Cs1] = unicode_util:gc(Str), - case lists:member(GC, GCs) of - false -> - {Head, Tail} = take_tc(Cs1, 0, Sep), - case equal(Tail, Cs1) of - true -> {Head, append(GC,Tail)}; - false -> {append(GC,Head), Tail} - end; - true -> - {Head, Tail} = take_tc(Cs1, 0, Sep), - {append(GC,Head), Tail} - end; +take_tc(Str, 0, {GCs,_,_}=Seps) when is_list(Str) -> + case unicode_util:gc(Str) of + [GC|Cs1] -> + case lists:member(GC, GCs) of false -> - {Head, Tail} = take_tc(Cs, 0, Sep), - case equal(Tail, Cs) of - true -> {Head, append(CP,Tail)}; - false -> {append(CP,Head), Tail} - end + {Head, Tail} = take_tc(Cs1, 0, Seps), + case is_empty(Head) of + true -> {Head, append(GC,Tail)}; + false -> {append(GC,Head), Tail} + end; + true -> + {Head, Tail} = take_tc(Cs1, 0, Seps), + {append(GC,Head), Tail} end; [] -> {[],[]} end; -take_tc(Bin, N, Sep) when is_binary(Bin) -> +take_tc(Bin, N, {GCs,_,_}=Seps0) when is_binary(Bin) -> <<_:N/binary, Rest/binary>> = Bin, - case bin_search_inv(Rest, [], Sep) of + case bin_search_inv(Rest, [], GCs) of {nomatch,_} -> {Bin, <<>>}; [SepStart] -> - case bin_search(SepStart, [], Sep) of + Seps = search_compile(Seps0), + case bin_search(SepStart, [], Seps) of {nomatch,_} -> KeepSz = byte_size(Bin) - byte_size(SepStart), <<Before:KeepSz/binary, End/binary>> = Bin, {Before, End}; [NonSep] -> KeepSz = byte_size(Bin) - byte_size(NonSep), - take_tc(Bin, KeepSz, Sep) + take_tc(Bin, KeepSz, Seps) end end. -prefix_1(Cs, []) -> Cs; -prefix_1(Cs, [_]=Pre) -> - prefix_2(unicode_util:gc(Cs), Pre); -prefix_1(Cs, Pre) -> - prefix_2(unicode_util:cp(Cs), Pre). - -prefix_2([C|Cs], [C|Pre]) -> - prefix_1(Cs, Pre); -prefix_2(_, _) -> - nomatch. +prefix_1(Cs0, [GC]) -> + case unicode_util:gc(Cs0) of + [GC|Cs] -> Cs; + _ -> nomatch + end; +prefix_1([CP|Cs], [Pre|PreR]) when is_integer(CP) -> + case CP =:= Pre of + true -> prefix_1(Cs,PreR); + false -> nomatch + end; +prefix_1(<<CP/utf8, Cs/binary>>, [Pre|PreR]) -> + case CP =:= Pre of + true -> prefix_1(Cs,PreR); + false -> nomatch + end; +prefix_1(Cs0, [Pre|PreR]) -> + case unicode_util:cp(Cs0) of + [Pre|Cs] -> prefix_1(Cs,PreR); + _ -> nomatch + end. +split_1([CP1|Cs]=Cs0, [C|_]=Needle, _, Where, Curr, Acc) when is_integer(CP1) -> + case CP1=:=C of + true -> + case prefix_1(Cs0, Needle) of + nomatch -> split_1(Cs, Needle, 0, Where, append(C,Curr), Acc); + Rest when Where =:= leading -> + [rev(Curr), Rest]; + Rest when Where =:= trailing -> + split_1(Cs, Needle, 0, Where, [C|Curr], [rev(Curr), Rest]); + Rest when Where =:= all -> + split_1(Rest, Needle, 0, Where, [], [rev(Curr)|Acc]) + end; + false -> + split_1(Cs, Needle, 0, Where, append(CP1,Curr), Acc) + end; split_1([Bin|Cont0], Needle, Start, Where, Curr0, Acc) when is_binary(Bin) -> case bin_search_str(Bin, Start, Cont0, Needle) of @@ -981,32 +1159,50 @@ split_1(Bin, [_C|_]=Needle, Start, Where, Curr0, Acc) -> end end. -lexemes_m([Bin|Cont0], Seps, Ts) when is_binary(Bin) -> - case bin_search_inv(Bin, Cont0, Seps) of +lexemes_m([CP|_]=Cs0, {GCs,CPs,_}=Seps, Ts) when is_integer(CP) -> + case lists:member(CP, CPs) of + true -> + [GC|Cs2] = unicode_util:gc(Cs0), + case lists:member(GC, GCs) of + true -> + lexemes_m(Cs2, Seps, Ts); + false -> + {Lexeme,Rest} = lexeme_pick(Cs0, Seps, []), + lexemes_m(Rest, Seps, [Lexeme|Ts]) + end; + false -> + {Lexeme,Rest} = lexeme_pick(Cs0, Seps, []), + lexemes_m(Rest, Seps, [Lexeme|Ts]) + end; +lexemes_m([Bin|Cont0], {GCs,_,_}=Seps0, Ts) when is_binary(Bin) -> + case bin_search_inv(Bin, Cont0, GCs) of {nomatch,Cont} -> - lexemes_m(Cont, Seps, Ts); + lexemes_m(Cont, Seps0, Ts); Cs -> + Seps = search_compile(Seps0), {Lexeme,Rest} = lexeme_pick(Cs, Seps, []), lexemes_m(Rest, Seps, [Lexeme|Ts]) end; -lexemes_m(Cs0, {GCs, _, _}=Seps, Ts) when is_list(Cs0) -> +lexemes_m(Cs0, {GCs, _, _}=Seps0, Ts) when is_list(Cs0) -> case unicode_util:gc(Cs0) of [C|Cs] -> case lists:member(C, GCs) of true -> - lexemes_m(Cs, Seps, Ts); + lexemes_m(Cs, Seps0, Ts); false -> + Seps = search_compile(Seps0), {Lexeme,Rest} = lexeme_pick(Cs0, Seps, []), lexemes_m(Rest, Seps, [Lexeme|Ts]) end; [] -> lists:reverse(Ts) end; -lexemes_m(Bin, Seps, Ts) when is_binary(Bin) -> - case bin_search_inv(Bin, [], Seps) of +lexemes_m(Bin, {GCs,_,_}=Seps0, Ts) when is_binary(Bin) -> + case bin_search_inv(Bin, [], GCs) of {nomatch,_} -> lists:reverse(Ts); [Cs] -> + Seps = search_compile(Seps0), {Lexeme,Rest} = lexeme_pick(Cs, Seps, []), lexemes_m(Rest, Seps, add_non_empty(Lexeme,Ts)) end. @@ -1037,7 +1233,7 @@ lexeme_pick(Cs0, {GCs, CPs, _} = Seps, Tkn) when is_list(Cs0) -> true -> [GC|Cs2] = unicode_util:gc(Cs0), case lists:member(GC, GCs) of - true -> {rev(Tkn), Cs0}; + true -> {rev(Tkn), Cs2}; false -> lexeme_pick(Cs2, Seps, append(rev(GC),Tkn)) end; false -> @@ -1047,7 +1243,7 @@ lexeme_pick(Cs0, {GCs, CPs, _} = Seps, Tkn) when is_list(Cs0) -> {rev(Tkn), []} end; lexeme_pick(Bin, Seps, Tkn) when is_binary(Bin) -> - case bin_search(Bin, Seps) of + case bin_search(Bin, [], Seps) of {nomatch,_} -> {btoken(Bin,Tkn), []}; [Left] -> @@ -1056,35 +1252,38 @@ lexeme_pick(Bin, Seps, Tkn) when is_binary(Bin) -> {btoken(Lexeme, Tkn), Left} end. -nth_lexeme_m([Bin|Cont0], Seps, N) when is_binary(Bin) -> - case bin_search_inv(Bin, Cont0, Seps) of +nth_lexeme_m([Bin|Cont0], {GCs,_,_}=Seps0, N) when is_binary(Bin) -> + case bin_search_inv(Bin, Cont0, GCs) of {nomatch,Cont} -> - nth_lexeme_m(Cont, Seps, N); + nth_lexeme_m(Cont, Seps0, N); Cs when N > 1 -> - Rest = lexeme_skip(Cs, Seps), - nth_lexeme_m(Rest, Seps, N-1); + Rest = lexeme_skip(Cs, Seps0), + nth_lexeme_m(Rest, Seps0, N-1); Cs -> + Seps = search_compile(Seps0), {Lexeme,_} = lexeme_pick(Cs, Seps, []), Lexeme end; -nth_lexeme_m(Cs0, {GCs, _, _}=Seps, N) when is_list(Cs0) -> +nth_lexeme_m(Cs0, {GCs, _, _}=Seps0, N) when is_list(Cs0) -> case unicode_util:gc(Cs0) of [C|Cs] -> case lists:member(C, GCs) of true -> - nth_lexeme_m(Cs, Seps, N); + nth_lexeme_m(Cs, Seps0, N); false when N > 1 -> - Cs1 = lexeme_skip(Cs, Seps), - nth_lexeme_m(Cs1, Seps, N-1); + Cs1 = lexeme_skip(Cs, Seps0), + nth_lexeme_m(Cs1, Seps0, N-1); false -> + Seps = search_compile(Seps0), {Lexeme,_} = lexeme_pick(Cs0, Seps, []), Lexeme end; [] -> [] end; -nth_lexeme_m(Bin, Seps, N) when is_binary(Bin) -> - case bin_search_inv(Bin, [], Seps) of +nth_lexeme_m(Bin, {GCs,_,_}=Seps0, N) when is_binary(Bin) -> + Seps = search_compile(Seps0), + case bin_search_inv(Bin, [], GCs) of [Cs] when N > 1 -> Cs1 = lexeme_skip(Cs, Seps), nth_lexeme_m(Cs1, Seps, N-1); @@ -1100,16 +1299,17 @@ lexeme_skip([CP|Cs1]=Cs0, {GCs,CPs,_}=Seps) when is_integer(CP) -> true -> [GC|Cs2] = unicode_util:gc(Cs0), case lists:member(GC, GCs) of - true -> Cs0; + true -> Cs2; false -> lexeme_skip(Cs2, Seps) end; false -> lexeme_skip(Cs1, Seps) end; -lexeme_skip([Bin|Cont0], Seps) when is_binary(Bin) -> +lexeme_skip([Bin|Cont0], Seps0) when is_binary(Bin) -> + Seps = search_compile(Seps0), case bin_search(Bin, Cont0, Seps) of {nomatch,_} -> lexeme_skip(Cont0, Seps); - Cs -> Cs + Cs -> tl(unicode_util:gc(Cs)) end; lexeme_skip(Cs0, {GCs, CPs, _} = Seps) when is_list(Cs0) -> case unicode_util:cp(Cs0) of @@ -1118,7 +1318,7 @@ lexeme_skip(Cs0, {GCs, CPs, _} = Seps) when is_list(Cs0) -> true -> [GC|Cs2] = unicode_util:gc(Cs0), case lists:member(GC, GCs) of - true -> Cs0; + true -> Cs2; false -> lexeme_skip(Cs2, Seps) end; false -> @@ -1127,12 +1327,23 @@ lexeme_skip(Cs0, {GCs, CPs, _} = Seps) when is_list(Cs0) -> [] -> [] end; -lexeme_skip(Bin, Seps) when is_binary(Bin) -> - case bin_search(Bin, Seps) of +lexeme_skip(Bin, Seps0) when is_binary(Bin) -> + Seps = search_compile(Seps0), + case bin_search(Bin, [], Seps) of {nomatch,_} -> <<>>; - [Left] -> Left + [Left] -> tl(unicode_util:gc(Left)) end. +find_l([C1|Cs]=Cs0, [C|_]=Needle) when is_integer(C1) -> + case C1 of + C -> + case prefix_1(Cs0, Needle) of + nomatch -> find_l(Cs, Needle); + _ -> Cs0 + end; + _ -> + find_l(Cs, Needle) + end; find_l([Bin|Cont0], Needle) when is_binary(Bin) -> case bin_search_str(Bin, 0, Cont0, Needle) of {nomatch, _, Cont} -> @@ -1157,6 +1368,16 @@ find_l(Bin, Needle) -> {_Before, [Cs], _After} -> Cs end. +find_r([Cp|Cs]=Cs0, [C|_]=Needle, Res) when is_integer(Cp) -> + case Cp of + C -> + case prefix_1(Cs0, Needle) of + nomatch -> find_r(Cs, Needle, Res); + _ -> find_r(Cs, Needle, Cs0) + end; + _ -> + find_r(Cs, Needle, Res) + end; find_r([Bin|Cont0], Needle, Res) when is_binary(Bin) -> case bin_search_str(Bin, 0, Cont0, Needle) of {nomatch,_,Cont} -> @@ -1227,11 +1448,6 @@ cp_prefix_1(Orig, Until, Cont) -> %% Binary special -bin_search(Bin, Seps) -> - bin_search(Bin, [], Seps). - -bin_search(_Bin, Cont, {[],_,_}) -> - {nomatch, Cont}; bin_search(Bin, Cont, {Seps,_,BP}) -> bin_search_loop(Bin, 0, BP, Cont, Seps). @@ -1239,10 +1455,14 @@ bin_search(Bin, Cont, {Seps,_,BP}) -> %% i.e. å in nfd form $a "COMBINING RING ABOVE" %% and PREPEND characters like "ARABIC NUMBER SIGN" 1536 <<216,128>> %% combined with other characters are currently ignored. +search_pattern({_,_,_}=P) -> P; search_pattern(Seps) -> CPs = search_cp(Seps), - Bin = bin_pattern(CPs), - {Seps, CPs, Bin}. + {Seps, CPs, undefined}. + +search_compile({Sep, CPs, undefined}) -> + {Sep, CPs, binary:compile_pattern(bin_pattern(CPs))}; +search_compile({_,_,_}=Compiled) -> Compiled. search_cp([CP|Seps]) when is_integer(CP) -> [CP|search_cp(Seps)]; @@ -1263,9 +1483,21 @@ bin_search_loop(Bin0, Start, BinSeps, Cont, Seps) -> case binary:match(Bin, BinSeps) of nomatch -> {nomatch,Cont}; + {Where, _CL} when Cont =:= [] -> + <<_:Where/binary, Cont1/binary>> = Bin, + [GC|Cont2] = unicode_util:gc(Cont1), + case lists:member(GC, Seps) of + false when Cont2 =:= [] -> + {nomatch, []}; + false -> + Next = byte_size(Bin0) - byte_size(Cont2), + bin_search_loop(Bin0, Next, BinSeps, Cont, Seps); + true -> + [Cont1] + end; {Where, _CL} -> <<_:Where/binary, Cont0/binary>> = Bin, - Cont1 = stack(Cont0, Cont), + Cont1 = [Cont0|Cont], [GC|Cont2] = unicode_util:gc(Cont1), case lists:member(GC, Seps) of false -> @@ -1273,55 +1505,108 @@ bin_search_loop(Bin0, Start, BinSeps, Cont, Seps) -> [BinR|Cont] when is_binary(BinR) -> Next = byte_size(Bin0) - byte_size(BinR), bin_search_loop(Bin0, Next, BinSeps, Cont, Seps); - BinR when is_binary(BinR), Cont =:= [] -> - Next = byte_size(Bin0) - byte_size(BinR), - bin_search_loop(Bin0, Next, BinSeps, Cont, Seps); _ -> {nomatch, Cont2} end; - true when is_list(Cont1) -> - Cont1; true -> - [Cont1] + Cont1 end end. -bin_search_inv(Bin, Cont, {[], _, _}) -> - [Bin|Cont]; -bin_search_inv(Bin, Cont, {[Sep], _, _}) -> - bin_search_inv_1([Bin|Cont], Sep); -bin_search_inv(Bin, Cont, {Seps, _, _}) -> - bin_search_inv_n([Bin|Cont], Seps). - -bin_search_inv_1([<<>>|CPs], _) -> - {nomatch, CPs}; -bin_search_inv_1(CPs = [Bin0|Cont], Sep) when is_binary(Bin0) -> - case unicode_util:gc(CPs) of - [Sep|Bin] when is_binary(Bin), Cont =:= [] -> - bin_search_inv_1([Bin], Sep); - [Sep|[Bin|Cont]=Cs] when is_binary(Bin) -> - bin_search_inv_1(Cs, Sep); - [Sep|Cs] -> - {nomatch, Cs}; - _ -> CPs - end. +bin_search_inv(<<>>, Cont, _) -> + {nomatch, Cont}; +bin_search_inv(Bin, Cont, [Sep]) -> + bin_search_inv_1(Bin, Cont, Sep); +bin_search_inv(Bin, Cont, Seps) -> + bin_search_inv_n(Bin, Cont, Seps). + +bin_search_inv_1(<<CP1/utf8, BinRest/binary>>=Bin0, Cont, Sep) -> + case BinRest of + <<CP2/utf8, _/binary>> when ?ASCII_LIST(CP1, CP2) -> + case CP1 of + Sep -> bin_search_inv_1(BinRest, Cont, Sep); + _ -> [Bin0|Cont] + end; + _ when Cont =:= [] -> + case unicode_util:gc(Bin0) of + [Sep|Bin] -> bin_search_inv_1(Bin, Cont, Sep); + _ -> [Bin0|Cont] + end; + _ -> + case unicode_util:gc([Bin0|Cont]) of + [Sep|[Bin|Cont]] when is_binary(Bin) -> + bin_search_inv_1(Bin, Cont, Sep); + [Sep|Cs] -> + {nomatch, Cs}; + _ -> [Bin0|Cont] + end + end; +bin_search_inv_1(<<>>, Cont, _Sep) -> + {nomatch, Cont}; +bin_search_inv_1([], Cont, _Sep) -> + {nomatch, Cont}. -bin_search_inv_n([<<>>|CPs], _) -> - {nomatch, CPs}; -bin_search_inv_n([Bin0|Cont]=CPs, Seps) when is_binary(Bin0) -> - [C|Cs0] = unicode_util:gc(CPs), - case {lists:member(C, Seps), Cs0} of - {true, Cs} when is_binary(Cs), Cont =:= [] -> - bin_search_inv_n([Cs], Seps); - {true, [Bin|Cont]=Cs} when is_binary(Bin) -> - bin_search_inv_n(Cs, Seps); - {true, Cs} -> {nomatch, Cs}; - {false, _} -> CPs - end. +bin_search_inv_n(<<CP1/utf8, BinRest/binary>>=Bin0, Cont, Seps) -> + case BinRest of + <<CP2/utf8, _/binary>> when ?ASCII_LIST(CP1, CP2) -> + case lists:member(CP1,Seps) of + true -> bin_search_inv_n(BinRest, Cont, Seps); + false -> [Bin0|Cont] + end; + _ when Cont =:= [] -> + [GC|Bin] = unicode_util:gc(Bin0), + case lists:member(GC, Seps) of + true -> bin_search_inv_n(Bin, Cont, Seps); + false -> [Bin0|Cont] + end; + _ -> + [GC|Cs0] = unicode_util:gc([Bin0|Cont]), + case lists:member(GC, Seps) of + false -> [Bin0|Cont]; + true -> + case Cs0 of + [Bin|Cont] when is_binary(Bin) -> + bin_search_inv_n(Bin, Cont, Seps); + _ -> + {nomatch, Cs0} + end + end + end; +bin_search_inv_n(<<>>, Cont, _Sep) -> + {nomatch, Cont}; +bin_search_inv_n([], Cont, _Sep) -> + {nomatch, Cont}. + +bin_search_str(Bin0, Start, [], SearchCPs) -> + Compiled = binary:compile_pattern(unicode:characters_to_binary(SearchCPs)), + bin_search_str_1(Bin0, Start, Compiled, SearchCPs); bin_search_str(Bin0, Start, Cont, [CP|_]=SearchCPs) -> + First = binary:compile_pattern(<<CP/utf8>>), + bin_search_str_2(Bin0, Start, Cont, First, SearchCPs). + +bin_search_str_1(Bin0, Start, First, SearchCPs) -> + <<_:Start/binary, Bin/binary>> = Bin0, + case binary:match(Bin, First) of + nomatch -> {nomatch, byte_size(Bin0), []}; + {Where0, _} -> + Where = Start+Where0, + <<Keep:Where/binary, Cs0/binary>> = Bin0, + case prefix_1(Cs0, SearchCPs) of + nomatch -> + <<_/utf8, Cs/binary>> = Cs0, + KeepSz = byte_size(Bin0) - byte_size(Cs), + bin_search_str_1(Bin0, KeepSz, First, SearchCPs); + [] -> + {Keep, [Cs0], <<>>}; + Rest -> + {Keep, [Cs0], Rest} + end + end. + +bin_search_str_2(Bin0, Start, Cont, First, SearchCPs) -> <<_:Start/binary, Bin/binary>> = Bin0, - case binary:match(Bin, <<CP/utf8>>) of + case binary:match(Bin, First) of nomatch -> {nomatch, byte_size(Bin0), Cont}; {Where0, _} -> Where = Start+Where0, @@ -1330,7 +1615,7 @@ bin_search_str(Bin0, Start, Cont, [CP|_]=SearchCPs) -> case prefix_1(stack(Cs0,Cont), SearchCPs) of nomatch when is_binary(Cs) -> KeepSz = byte_size(Bin0) - byte_size(Cs), - bin_search_str(Bin0, KeepSz, Cont, SearchCPs); + bin_search_str_2(Bin0, KeepSz, Cont, First, SearchCPs); nomatch -> {nomatch, Where, stack([GC|Cs],Cont)}; [] -> diff --git a/lib/stdlib/src/supervisor.erl b/lib/stdlib/src/supervisor.erl index 7920e55930..e56415650f 100644 --- a/lib/stdlib/src/supervisor.erl +++ b/lib/stdlib/src/supervisor.erl @@ -31,7 +31,6 @@ %% Internal exports -export([init/1, handle_call/3, handle_cast/2, handle_info/2, terminate/2, code_change/3, format_status/2]). --export([try_again_restart/2]). %% For release_handler only -export([get_callback_module/1]). @@ -79,6 +78,7 @@ | {RestartStrategy :: strategy(), Intensity :: non_neg_integer(), Period :: pos_integer()}. +-type children() :: {Ids :: [child_id()], Db :: #{child_id() => child_rec()}}. %%-------------------------------------------------------------------------- %% Defaults @@ -96,7 +96,7 @@ pid = undefined :: child() | {restarting, pid() | undefined} | [pid()], - name :: child_id(), + id :: child_id(), mfargs :: mfargs(), restart_type :: restart(), shutdown :: shutdown(), @@ -104,16 +104,11 @@ modules = [] :: modules()}). -type child_rec() :: #child{}. --define(DICTS, dict). --define(DICT, dict:dict). --define(SETS, sets). --define(SET, sets:set). - -record(state, {name, strategy :: strategy() | 'undefined', - children = [] :: [child_rec()], - dynamics :: {'dict', ?DICT(pid(), list())} - | {'set', ?SET(pid())} + children = {[],#{}} :: children(), % Ids in start order + dynamics :: {'maps', #{pid() => list()}} + | {'sets', sets:set(pid())} | 'undefined', intensity :: non_neg_integer() | 'undefined', period :: pos_integer() | 'undefined', @@ -124,6 +119,9 @@ -type state() :: #state{}. -define(is_simple(State), State#state.strategy =:= simple_one_for_one). +-define(is_temporary(_Child_), _Child_#child.restart_type=:=temporary). +-define(is_transient(_Child_), _Child_#child.restart_type=:=transient). +-define(is_permanent(_Child_), _Child_#child.restart_type=:=permanent). -callback init(Args :: term()) -> {ok, {SupFlags :: sup_flags(), [ChildSpec :: child_spec()]}} @@ -179,16 +177,16 @@ start_child(Supervisor, ChildSpec) -> | {'error', Error}, Error :: 'running' | 'restarting' | 'not_found' | 'simple_one_for_one' | term(). -restart_child(Supervisor, Name) -> - call(Supervisor, {restart_child, Name}). +restart_child(Supervisor, Id) -> + call(Supervisor, {restart_child, Id}). -spec delete_child(SupRef, Id) -> Result when SupRef :: sup_ref(), Id :: child_id(), Result :: 'ok' | {'error', Error}, Error :: 'running' | 'restarting' | 'not_found' | 'simple_one_for_one'. -delete_child(Supervisor, Name) -> - call(Supervisor, {delete_child, Name}). +delete_child(Supervisor, Id) -> + call(Supervisor, {delete_child, Id}). %%----------------------------------------------------------------- %% Func: terminate_child/2 @@ -202,16 +200,16 @@ delete_child(Supervisor, Name) -> Id :: pid() | child_id(), Result :: 'ok' | {'error', Error}, Error :: 'not_found' | 'simple_one_for_one'. -terminate_child(Supervisor, Name) -> - call(Supervisor, {terminate_child, Name}). +terminate_child(Supervisor, Id) -> + call(Supervisor, {terminate_child, Id}). -spec get_childspec(SupRef, Id) -> Result when SupRef :: sup_ref(), Id :: pid() | child_id(), Result :: {'ok', child_spec()} | {'error', Error}, Error :: 'not_found'. -get_childspec(Supervisor, Name) -> - call(Supervisor, {get_childspec, Name}). +get_childspec(Supervisor, Id) -> + call(Supervisor, {get_childspec, Id}). -spec which_children(SupRef) -> [{Id,Child,Type,Modules}] when SupRef :: sup_ref(), @@ -246,17 +244,6 @@ check_childspecs(ChildSpecs) when is_list(ChildSpecs) -> check_childspecs(X) -> {error, {badarg, X}}. %%%----------------------------------------------------------------- -%%% Called by restart/2 --spec try_again_restart(SupRef, Child) -> ok when - SupRef :: sup_ref(), - Child :: child_id() | pid(). -try_again_restart(Supervisor, Child) -> - cast(Supervisor, {try_again_restart, Child}). - -cast(Supervisor, Req) -> - gen_server:cast(Supervisor, Req). - -%%%----------------------------------------------------------------- %%% Called by release_handler during upgrade -spec get_callback_module(Pid) -> Module when Pid :: pid(), @@ -325,7 +312,7 @@ init_children(State, StartSpec) -> init_dynamic(State, [StartSpec]) -> case check_startspec([StartSpec]) of {ok, Children} -> - {ok, State#state{children = Children}}; + {ok, dyn_init(State#state{children = Children})}; Error -> {stop, {start_spec, Error}} end; @@ -334,35 +321,34 @@ init_dynamic(_State, StartSpec) -> %%----------------------------------------------------------------- %% Func: start_children/2 -%% Args: Children = [child_rec()] in start order +%% Args: Children = children() % Ids in start order %% SupName = {local, atom()} | {global, atom()} | {pid(), Mod} -%% Purpose: Start all children. The new list contains #child's +%% Purpose: Start all children. The new map contains #child's %% with pids. %% Returns: {ok, NChildren} | {error, NChildren, Reason} -%% NChildren = [child_rec()] in termination order (reversed -%% start order) +%% NChildren = children() % Ids in termination order +%% (reversed start order) %%----------------------------------------------------------------- -start_children(Children, SupName) -> start_children(Children, [], SupName). - -start_children([Child|Chs], NChildren, SupName) -> - case do_start_child(SupName, Child) of - {ok, undefined} when Child#child.restart_type =:= temporary -> - start_children(Chs, NChildren, SupName); - {ok, Pid} -> - start_children(Chs, [Child#child{pid = Pid}|NChildren], SupName); - {ok, Pid, _Extra} -> - start_children(Chs, [Child#child{pid = Pid}|NChildren], SupName); - {error, Reason} -> - report_error(start_error, Reason, Child, SupName), - {error, lists:reverse(Chs) ++ [Child | NChildren], - {failed_to_start_child,Child#child.name,Reason}} - end; -start_children([], NChildren, _SupName) -> - {ok, NChildren}. +start_children(Children, SupName) -> + Start = + fun(Id,Child) -> + case do_start_child(SupName, Child) of + {ok, undefined} when ?is_temporary(Child) -> + remove; + {ok, Pid} -> + {update,Child#child{pid = Pid}}; + {ok, Pid, _Extra} -> + {update,Child#child{pid = Pid}}; + {error, Reason} -> + report_error(start_error, Reason, Child, SupName), + {abort,{failed_to_start_child,Id,Reason}} + end + end, + children_map(Start,Children). do_start_child(SupName, Child) -> #child{mfargs = {M, F, Args}} = Child, - case catch apply(M, F, Args) of + case do_start_child_i(M, F, Args) of {ok, Pid} when is_pid(Pid) -> NChild = Child#child{pid = Pid}, report_progress(NChild, SupName), @@ -371,10 +357,8 @@ do_start_child(SupName, Child) -> NChild = Child#child{pid = Pid}, report_progress(NChild, SupName), {ok, Pid, Extra}; - ignore -> - {ok, undefined}; - {error, What} -> {error, What}; - What -> {error, What} + Other -> + Other end. do_start_child_i(M, F, A) -> @@ -400,17 +384,17 @@ do_start_child_i(M, F, A) -> -spec handle_call(call(), term(), state()) -> {'reply', term(), state()}. handle_call({start_child, EArgs}, _From, State) when ?is_simple(State) -> - Child = hd(State#state.children), + Child = get_dynamic_child(State), #child{mfargs = {M, F, A}} = Child, Args = A ++ EArgs, case do_start_child_i(M, F, Args) of {ok, undefined} -> {reply, {ok, undefined}, State}; {ok, Pid} -> - NState = save_dynamic_child(Child#child.restart_type, Pid, Args, State), + NState = dyn_store(Pid, Args, State), {reply, {ok, Pid}, NState}; {ok, Pid, Extra} -> - NState = save_dynamic_child(Child#child.restart_type, Pid, Args, State), + NState = dyn_store(Pid, Args, State), {reply, {ok, Pid, Extra}, NState}; What -> {reply, What, State} @@ -426,121 +410,94 @@ handle_call({start_child, ChildSpec}, _From, State) -> end; %% terminate_child for simple_one_for_one can only be done with pid -handle_call({terminate_child, Name}, _From, State) when not is_pid(Name), - ?is_simple(State) -> +handle_call({terminate_child, Id}, _From, State) when not is_pid(Id), + ?is_simple(State) -> {reply, {error, simple_one_for_one}, State}; -handle_call({terminate_child, Name}, _From, State) -> - case get_child(Name, State, ?is_simple(State)) of - {value, Child} -> - case do_terminate(Child, State#state.name) of - #child{restart_type=RT} when RT=:=temporary; ?is_simple(State) -> - {reply, ok, state_del_child(Child, State)}; - NChild -> - {reply, ok, replace_child(NChild, State)} - end; - false -> +handle_call({terminate_child, Id}, _From, State) -> + case find_child(Id, State) of + {ok, Child} -> + do_terminate(Child, State#state.name), + {reply, ok, del_child(Child, State)}; + error -> {reply, {error, not_found}, State} end; %% restart_child request is invalid for simple_one_for_one supervisors -handle_call({restart_child, _Name}, _From, State) when ?is_simple(State) -> +handle_call({restart_child, _Id}, _From, State) when ?is_simple(State) -> {reply, {error, simple_one_for_one}, State}; -handle_call({restart_child, Name}, _From, State) -> - case get_child(Name, State) of - {value, Child} when Child#child.pid =:= undefined -> +handle_call({restart_child, Id}, _From, State) -> + case find_child(Id, State) of + {ok, Child} when Child#child.pid =:= undefined -> case do_start_child(State#state.name, Child) of {ok, Pid} -> - NState = replace_child(Child#child{pid = Pid}, State), + NState = set_pid(Pid, Id, State), {reply, {ok, Pid}, NState}; {ok, Pid, Extra} -> - NState = replace_child(Child#child{pid = Pid}, State), + NState = set_pid(Pid, Id, State), {reply, {ok, Pid, Extra}, NState}; Error -> {reply, Error, State} end; - {value, #child{pid=?restarting(_)}} -> + {ok, #child{pid=?restarting(_)}} -> {reply, {error, restarting}, State}; - {value, _} -> + {ok, _} -> {reply, {error, running}, State}; _ -> {reply, {error, not_found}, State} end; %% delete_child request is invalid for simple_one_for_one supervisors -handle_call({delete_child, _Name}, _From, State) when ?is_simple(State) -> +handle_call({delete_child, _Id}, _From, State) when ?is_simple(State) -> {reply, {error, simple_one_for_one}, State}; -handle_call({delete_child, Name}, _From, State) -> - case get_child(Name, State) of - {value, Child} when Child#child.pid =:= undefined -> - NState = remove_child(Child, State), +handle_call({delete_child, Id}, _From, State) -> + case find_child(Id, State) of + {ok, Child} when Child#child.pid =:= undefined -> + NState = remove_child(Id, State), {reply, ok, NState}; - {value, #child{pid=?restarting(_)}} -> + {ok, #child{pid=?restarting(_)}} -> {reply, {error, restarting}, State}; - {value, _} -> + {ok, _} -> {reply, {error, running}, State}; _ -> {reply, {error, not_found}, State} end; -handle_call({get_childspec, Name}, _From, State) -> - case get_child(Name, State, ?is_simple(State)) of - {value, Child} -> +handle_call({get_childspec, Id}, _From, State) -> + case find_child(Id, State) of + {ok, Child} -> {reply, {ok, child_to_spec(Child)}, State}; - false -> + error -> {reply, {error, not_found}, State} end; -handle_call(which_children, _From, #state{children = [#child{restart_type = temporary, - child_type = CT, - modules = Mods}]} = - State) when ?is_simple(State) -> - Reply = lists:map(fun(Pid) -> {undefined, Pid, CT, Mods} end, - ?SETS:to_list(dynamics_db(temporary, State#state.dynamics))), - {reply, Reply, State}; - -handle_call(which_children, _From, #state{children = [#child{restart_type = RType, - child_type = CT, - modules = Mods}]} = - State) when ?is_simple(State) -> - Reply = lists:map(fun({?restarting(_),_}) -> {undefined,restarting,CT,Mods}; - ({Pid, _}) -> {undefined, Pid, CT, Mods} end, - ?DICTS:to_list(dynamics_db(RType, State#state.dynamics))), +handle_call(which_children, _From, State) when ?is_simple(State) -> + #child{child_type = CT,modules = Mods} = get_dynamic_child(State), + Reply = dyn_map(fun(?restarting(_)) -> {undefined, restarting, CT, Mods}; + (Pid) -> {undefined, Pid, CT, Mods} + end, State), {reply, Reply, State}; handle_call(which_children, _From, State) -> Resp = - lists:map(fun(#child{pid = ?restarting(_), name = Name, - child_type = ChildType, modules = Mods}) -> - {Name, restarting, ChildType, Mods}; - (#child{pid = Pid, name = Name, - child_type = ChildType, modules = Mods}) -> - {Name, Pid, ChildType, Mods} - end, - State#state.children), + children_to_list( + fun(Id,#child{pid = ?restarting(_), + child_type = ChildType, modules = Mods}) -> + {Id, restarting, ChildType, Mods}; + (Id,#child{pid = Pid, + child_type = ChildType, modules = Mods}) -> + {Id, Pid, ChildType, Mods} + end, + State#state.children), {reply, Resp, State}; - -handle_call(count_children, _From, #state{children = [#child{restart_type = temporary, - child_type = CT}]} = State) - when ?is_simple(State) -> - Sz = ?SETS:size(dynamics_db(temporary, State#state.dynamics)), - Reply = case CT of - supervisor -> [{specs, 1}, {active, Sz}, - {supervisors, Sz}, {workers, 0}]; - worker -> [{specs, 1}, {active, Sz}, - {supervisors, 0}, {workers, Sz}] - end, - {reply, Reply, State}; - -handle_call(count_children, _From, #state{dynamic_restarts = Restarts, - children = [#child{restart_type = RType, - child_type = CT}]} = State) +handle_call(count_children, _From, #state{dynamic_restarts = Restarts} = State) when ?is_simple(State) -> - Sz = ?DICTS:size(dynamics_db(RType, State#state.dynamics)), - Active = Sz - Restarts, + #child{child_type = CT} = get_dynamic_child(State), + Sz = dyn_size(State), + Active = Sz - Restarts, % Restarts is always 0 for temporary children Reply = case CT of supervisor -> [{specs, 1}, {active, Active}, {supervisors, Sz}, {workers, 0}]; @@ -552,16 +509,15 @@ handle_call(count_children, _From, #state{dynamic_restarts = Restarts, handle_call(count_children, _From, State) -> %% Specs and children are together on the children list... {Specs, Active, Supers, Workers} = - lists:foldl(fun(Child, Counts) -> - count_child(Child, Counts) - end, {0,0,0,0}, State#state.children), + children_fold(fun(_Id, Child, Counts) -> + count_child(Child, Counts) + end, {0,0,0,0}, State#state.children), %% Reformat counts to a property list. Reply = [{specs, Specs}, {active, Active}, {supervisors, Supers}, {workers, Workers}], {reply, Reply, State}. - count_child(#child{pid = Pid, child_type = worker}, {Specs, Active, Supers, Workers}) -> case is_pid(Pid) andalso is_process_alive(Pid) of @@ -575,34 +531,15 @@ count_child(#child{pid = Pid, child_type = supervisor}, false -> {Specs+1, Active, Supers+1, Workers} end. - %%% If a restart attempt failed, this message is cast %%% from restart/2 in order to give gen_server the chance to %%% check it's inbox before trying again. --spec handle_cast({try_again_restart, child_id() | pid()}, state()) -> +-spec handle_cast({try_again_restart, child_id() | {'restarting',pid()}}, state()) -> {'noreply', state()} | {stop, shutdown, state()}. -handle_cast({try_again_restart,Pid}, #state{children=[Child]}=State) - when ?is_simple(State) -> - RT = Child#child.restart_type, - RPid = restarting(Pid), - case dynamic_child_args(RPid, RT, State#state.dynamics) of - {ok, Args} -> - {M, F, _} = Child#child.mfargs, - NChild = Child#child{pid = RPid, mfargs = {M, F, Args}}, - case restart(NChild,State) of - {ok, State1} -> - {noreply, State1}; - {shutdown, State1} -> - {stop, shutdown, State1} - end; - error -> - {noreply, State} - end; - -handle_cast({try_again_restart,Name}, State) -> - case lists:keyfind(Name,#child.name,State#state.children) of - Child = #child{pid=?restarting(_)} -> +handle_cast({try_again_restart,TryAgainId}, State) -> + case find_child_and_args(TryAgainId, State) of + {ok, Child = #child{pid=?restarting(_)}} -> case restart(Child,State) of {ok, State1} -> {noreply, State1}; @@ -637,10 +574,8 @@ handle_info(Msg, State) -> %% -spec terminate(term(), state()) -> 'ok'. -terminate(_Reason, #state{children=[Child]} = State) when ?is_simple(State) -> - terminate_dynamic_children(Child, dynamics_db(Child#child.restart_type, - State#state.dynamics), - State#state.name); +terminate(_Reason, State) when ?is_simple(State) -> + terminate_dynamic_children(State); terminate(_Reason, State) -> terminate_children(State#state.children, State#state.name). @@ -675,8 +610,8 @@ code_change(_, State, _) -> update_childspec(State, StartSpec) when ?is_simple(State) -> case check_startspec(StartSpec) of - {ok, [Child]} -> - {ok, State#state{children = [Child]}}; + {ok, {[_],_}=Children} -> + {ok, State#state{children = Children}}; Error -> {error, Error} end; @@ -690,39 +625,36 @@ update_childspec(State, StartSpec) -> {error, Error} end. -update_childspec1([Child|OldC], Children, KeepOld) -> - case update_chsp(Child, Children) of - {ok,NewChildren} -> - update_childspec1(OldC, NewChildren, KeepOld); +update_childspec1({[Id|OldIds], OldDb}, {Ids,Db}, KeepOld) -> + case update_chsp(maps:get(Id,OldDb), Db) of + {ok,NewDb} -> + update_childspec1({OldIds,OldDb}, {Ids,NewDb}, KeepOld); false -> - update_childspec1(OldC, Children, [Child|KeepOld]) + update_childspec1({OldIds,OldDb}, {Ids,Db}, [Id|KeepOld]) end; -update_childspec1([], Children, KeepOld) -> +update_childspec1({[],OldDb}, {Ids,Db}, KeepOld) -> + KeepOldDb = maps:with(KeepOld,OldDb), %% Return them in (kept) reverse start order. - lists:reverse(Children ++ KeepOld). - -update_chsp(OldCh, Children) -> - case lists:map(fun(Ch) when OldCh#child.name =:= Ch#child.name -> - Ch#child{pid = OldCh#child.pid}; - (Ch) -> - Ch - end, - Children) of - Children -> - false; % OldCh not found in new spec. - NewC -> - {ok, NewC} + {lists:reverse(Ids ++ KeepOld),maps:merge(KeepOldDb,Db)}. + +update_chsp(#child{id=Id}=OldChild, NewDb) -> + case maps:find(Id, NewDb) of + {ok,Child} -> + {ok,NewDb#{Id => Child#child{pid = OldChild#child.pid}}}; + error -> % Id not found in new spec. + false end. + %%% --------------------------------------------------- %%% Start a new child. %%% --------------------------------------------------- handle_start_child(Child, State) -> - case get_child(Child#child.name, State) of - false -> + case find_child(Child#child.id, State) of + error -> case do_start_child(State#state.name, Child) of - {ok, undefined} when Child#child.restart_type =:= temporary -> + {ok, undefined} when ?is_temporary(Child) -> {{ok, undefined}, State}; {ok, Pid} -> {{ok, Pid}, save_child(Child#child{pid = Pid}, State)}; @@ -731,9 +663,9 @@ handle_start_child(Child, State) -> {error, What} -> {{error, {What, Child}}, State} end; - {value, OldChild} when is_pid(OldChild#child.pid) -> + {ok, OldChild} when is_pid(OldChild#child.pid) -> {{error, {already_started, OldChild#child.pid}}, State}; - {value, _OldChild} -> + {ok, _OldChild} -> {{error, already_present}, State} end. @@ -742,63 +674,45 @@ handle_start_child(Child, State) -> %%% Returns: {ok, state()} | {shutdown, state()} %%% --------------------------------------------------- -restart_child(Pid, Reason, #state{children = [Child]} = State) when ?is_simple(State) -> - RestartType = Child#child.restart_type, - case dynamic_child_args(Pid, RestartType, State#state.dynamics) of - {ok, Args} -> - {M, F, _} = Child#child.mfargs, - NChild = Child#child{pid = Pid, mfargs = {M, F, Args}}, - do_restart(RestartType, Reason, NChild, State); - error -> - {ok, State} - end; - restart_child(Pid, Reason, State) -> - Children = State#state.children, - case lists:keyfind(Pid, #child.pid, Children) of - #child{restart_type = RestartType} = Child -> - do_restart(RestartType, Reason, Child, State); - false -> + case find_child_and_args(Pid, State) of + {ok, Child} -> + do_restart(Reason, Child, State); + error -> {ok, State} end. -do_restart(permanent, Reason, Child, State) -> +do_restart(Reason, Child, State) when ?is_permanent(Child) -> report_error(child_terminated, Reason, Child, State#state.name), restart(Child, State); -do_restart(_, normal, Child, State) -> - NState = state_del_child(Child, State), +do_restart(normal, Child, State) -> + NState = del_child(Child, State), {ok, NState}; -do_restart(_, shutdown, Child, State) -> - NState = state_del_child(Child, State), +do_restart(shutdown, Child, State) -> + NState = del_child(Child, State), {ok, NState}; -do_restart(_, {shutdown, _Term}, Child, State) -> - NState = state_del_child(Child, State), +do_restart({shutdown, _Term}, Child, State) -> + NState = del_child(Child, State), {ok, NState}; -do_restart(transient, Reason, Child, State) -> +do_restart(Reason, Child, State) when ?is_transient(Child) -> report_error(child_terminated, Reason, Child, State#state.name), restart(Child, State); -do_restart(temporary, Reason, Child, State) -> +do_restart(Reason, Child, State) when ?is_temporary(Child) -> report_error(child_terminated, Reason, Child, State#state.name), - NState = state_del_child(Child, State), + NState = del_child(Child, State), {ok, NState}. restart(Child, State) -> case add_restart(State) of {ok, NState} -> case restart(NState#state.strategy, Child, NState) of - {try_again,NState2} -> + {{try_again, TryAgainId}, NState2} -> %% Leaving control back to gen_server before %% trying again. This way other incoming requsts %% for the supervisor can be handled - e.g. a %% shutdown request for the supervisor or the %% child. - Id = if ?is_simple(State) -> Child#child.pid; - true -> Child#child.name - end, - ok = try_again_restart(self(), Id), - {ok,NState2}; - {try_again, NState2, #child{name=ChName}} -> - ok = try_again_restart(self(), ChName), + try_again_restart(TryAgainId), {ok,NState2}; Other -> Other @@ -806,124 +720,111 @@ restart(Child, State) -> {terminate, NState} -> report_error(shutdown, reached_max_restart_intensity, Child, State#state.name), - {shutdown, remove_child(Child, NState)} + {shutdown, del_child(Child, NState)} end. restart(simple_one_for_one, Child, State0) -> #child{pid = OldPid, mfargs = {M, F, A}} = Child, - State = case OldPid of + State1 = case OldPid of ?restarting(_) -> NRes = State0#state.dynamic_restarts - 1, State0#state{dynamic_restarts = NRes}; _ -> State0 end, - Dynamics = ?DICTS:erase(OldPid, dynamics_db(Child#child.restart_type, - State#state.dynamics)), + State2 = dyn_erase(OldPid, State1), case do_start_child_i(M, F, A) of {ok, Pid} -> - DynamicsDb = {dict, ?DICTS:store(Pid, A, Dynamics)}, - NState = State#state{dynamics = DynamicsDb}, + NState = dyn_store(Pid, A, State2), {ok, NState}; {ok, Pid, _Extra} -> - DynamicsDb = {dict, ?DICTS:store(Pid, A, Dynamics)}, - NState = State#state{dynamics = DynamicsDb}, + NState = dyn_store(Pid, A, State2), {ok, NState}; {error, Error} -> - NRestarts = State#state.dynamic_restarts + 1, - DynamicsDb = {dict, ?DICTS:store(restarting(OldPid), A, Dynamics)}, - NState = State#state{dynamic_restarts = NRestarts, - dynamics = DynamicsDb}, - report_error(start_error, Error, Child, State#state.name), - {try_again, NState} + ROldPid = restarting(OldPid), + NRestarts = State2#state.dynamic_restarts + 1, + State3 = State2#state{dynamic_restarts = NRestarts}, + NState = dyn_store(ROldPid, A, State3), + report_error(start_error, Error, Child, NState#state.name), + {{try_again, ROldPid}, NState} end; -restart(one_for_one, Child, State) -> +restart(one_for_one, #child{id=Id} = Child, State) -> OldPid = Child#child.pid, case do_start_child(State#state.name, Child) of {ok, Pid} -> - NState = replace_child(Child#child{pid = Pid}, State), + NState = set_pid(Pid, Id, State), {ok, NState}; {ok, Pid, _Extra} -> - NState = replace_child(Child#child{pid = Pid}, State), + NState = set_pid(Pid, Id, State), {ok, NState}; {error, Reason} -> - NState = replace_child(Child#child{pid = restarting(OldPid)}, State), + NState = set_pid(restarting(OldPid), Id, State), report_error(start_error, Reason, Child, State#state.name), - {try_again, NState} - end; -restart(rest_for_one, Child, State) -> - {ChAfter, ChBefore} = split_child(Child#child.pid, State#state.children), - ChAfter2 = terminate_children(ChAfter, State#state.name), - case start_children(ChAfter2, State#state.name) of - {ok, ChAfter3} -> - {ok, State#state{children = ChAfter3 ++ ChBefore}}; - {error, ChAfter3, {failed_to_start_child, ChName, _Reason}} - when ChName =:= Child#child.name -> - NChild = Child#child{pid=restarting(Child#child.pid)}, - NState = State#state{children = ChAfter3 ++ ChBefore}, - {try_again, replace_child(NChild,NState)}; - {error, ChAfter3, {failed_to_start_child, ChName, _Reason}} -> - NChild = lists:keyfind(ChName, #child.name, ChAfter3), - NChild2 = NChild#child{pid=?restarting(undefined)}, - NState = State#state{children = ChAfter3 ++ ChBefore}, - {try_again, replace_child(NChild2,NState), NChild2} + {{try_again,Id}, NState} end; -restart(one_for_all, Child, State) -> - Children1 = del_child(Child#child.pid, State#state.children), - Children2 = terminate_children(Children1, State#state.name), - case start_children(Children2, State#state.name) of - {ok, NChs} -> - {ok, State#state{children = NChs}}; - {error, NChs, {failed_to_start_child, ChName, _Reason}} - when ChName =:= Child#child.name -> - NChild = Child#child{pid=restarting(Child#child.pid)}, - NState = State#state{children = NChs}, - {try_again, replace_child(NChild,NState)}; - {error, NChs, {failed_to_start_child, ChName, _Reason}} -> - NChild = lists:keyfind(ChName, #child.name, NChs), - NChild2 = NChild#child{pid=?restarting(undefined)}, - NState = State#state{children = NChs}, - {try_again, replace_child(NChild2,NState), NChild2} +restart(rest_for_one, #child{id=Id} = Child, #state{name=SupName} = State) -> + {ChAfter, ChBefore} = split_child(Id, State#state.children), + {Return, ChAfter2} = restart_multiple_children(Child, ChAfter, SupName), + {Return, State#state{children = append(ChAfter2,ChBefore)}}; +restart(one_for_all, Child, #state{name=SupName} = State) -> + Children1 = del_child(Child#child.id, State#state.children), + {Return, NChildren} = restart_multiple_children(Child, Children1, SupName), + {Return, State#state{children = NChildren}}. + +restart_multiple_children(Child, Children, SupName) -> + Children1 = terminate_children(Children, SupName), + case start_children(Children1, SupName) of + {ok, NChildren} -> + {ok, NChildren}; + {error, NChildren, {failed_to_start_child, FailedId, _Reason}} -> + NewPid = if FailedId =:= Child#child.id -> + restarting(Child#child.pid); + true -> + ?restarting(undefined) + end, + {{try_again, FailedId}, set_pid(NewPid,FailedId,NChildren)} end. restarting(Pid) when is_pid(Pid) -> ?restarting(Pid); restarting(RPid) -> RPid. +-spec try_again_restart(child_id() | {'restarting',pid()}) -> 'ok'. +try_again_restart(TryAgainId) -> + gen_server:cast(self(), {try_again_restart, TryAgainId}). + %%----------------------------------------------------------------- %% Func: terminate_children/2 -%% Args: Children = [child_rec()] in termination order +%% Args: Children = children() % Ids in termination order %% SupName = {local, atom()} | {global, atom()} | {pid(),Mod} -%% Returns: NChildren = [child_rec()] in -%% startup order (reversed termination order) +%% Returns: NChildren = children() % Ids in startup order +%% % (reversed termination order) %%----------------------------------------------------------------- terminate_children(Children, SupName) -> - terminate_children(Children, SupName, []). - -%% Temporary children should not be restarted and thus should -%% be skipped when building the list of terminated children, although -%% we do want them to be shut down as many functions from this module -%% use this function to just clear everything. -terminate_children([Child = #child{restart_type=temporary} | Children], SupName, Res) -> - _ = do_terminate(Child, SupName), - terminate_children(Children, SupName, Res); -terminate_children([Child | Children], SupName, Res) -> - NChild = do_terminate(Child, SupName), - terminate_children(Children, SupName, [NChild | Res]); -terminate_children([], _SupName, Res) -> - Res. + Terminate = + fun(_Id,Child) when ?is_temporary(Child) -> + %% Temporary children should not be restarted and thus should + %% be skipped when building the list of terminated children. + do_terminate(Child, SupName), + remove; + (_Id,Child) -> + do_terminate(Child, SupName), + {update,Child#child{pid=undefined}} + end, + {ok,NChildren} = children_map(Terminate, Children), + NChildren. do_terminate(Child, SupName) when is_pid(Child#child.pid) -> case shutdown(Child#child.pid, Child#child.shutdown) of ok -> ok; - {error, normal} when Child#child.restart_type =/= permanent -> + {error, normal} when not (?is_permanent(Child)) -> ok; {error, OtherReason} -> report_error(shutdown_error, OtherReason, Child, SupName) end, - Child#child{pid = undefined}; -do_terminate(Child, _SupName) -> - Child#child{pid = undefined}. + ok; +do_terminate(_Child, _SupName) -> + ok. %%----------------------------------------------------------------- %% Shutdowns a child. We must check the EXIT value @@ -996,66 +897,50 @@ monitor_child(Pid) -> ok end. - %%----------------------------------------------------------------- -%% Func: terminate_dynamic_children/3 -%% Args: Child = child_rec() -%% Dynamics = ?DICT() | ?SET() -%% SupName = {local, atom()} | {global, atom()} | {pid(),Mod} +%% Func: terminate_dynamic_children/1 +%% Args: State %% Returns: ok %% -%% %% Shutdown all dynamic children. This happens when the supervisor is %% stopped. Because the supervisor can have millions of dynamic children, we -%% can have an significative overhead here. +%% can have a significative overhead here. %%----------------------------------------------------------------- -terminate_dynamic_children(Child, Dynamics, SupName) -> - {Pids, EStack0} = monitor_dynamic_children(Child, Dynamics), - Sz = ?SETS:size(Pids), +terminate_dynamic_children(State) -> + Child = get_dynamic_child(State), + {Pids, EStack0} = monitor_dynamic_children(Child,State), + Sz = sets:size(Pids), EStack = case Child#child.shutdown of brutal_kill -> - ?SETS:fold(fun(P, _) -> exit(P, kill) end, ok, Pids), + sets:fold(fun(P, _) -> exit(P, kill) end, ok, Pids), wait_dynamic_children(Child, Pids, Sz, undefined, EStack0); infinity -> - ?SETS:fold(fun(P, _) -> exit(P, shutdown) end, ok, Pids), + sets:fold(fun(P, _) -> exit(P, shutdown) end, ok, Pids), wait_dynamic_children(Child, Pids, Sz, undefined, EStack0); Time -> - ?SETS:fold(fun(P, _) -> exit(P, shutdown) end, ok, Pids), + sets:fold(fun(P, _) -> exit(P, shutdown) end, ok, Pids), TRef = erlang:start_timer(Time, self(), kill), wait_dynamic_children(Child, Pids, Sz, TRef, EStack0) end, %% Unroll stacked errors and report them - ?DICTS:fold(fun(Reason, Ls, _) -> - report_error(shutdown_error, Reason, - Child#child{pid=Ls}, SupName) - end, ok, EStack). - - -monitor_dynamic_children(#child{restart_type=temporary}, Dynamics) -> - ?SETS:fold(fun(P, {Pids, EStack}) -> - case monitor_child(P) of - ok -> - {?SETS:add_element(P, Pids), EStack}; - {error, normal} -> - {Pids, EStack}; - {error, Reason} -> - {Pids, ?DICTS:append(Reason, P, EStack)} - end - end, {?SETS:new(), ?DICTS:new()}, Dynamics); -monitor_dynamic_children(#child{restart_type=RType}, Dynamics) -> - ?DICTS:fold(fun(P, _, {Pids, EStack}) when is_pid(P) -> - case monitor_child(P) of - ok -> - {?SETS:add_element(P, Pids), EStack}; - {error, normal} when RType =/= permanent -> - {Pids, EStack}; - {error, Reason} -> - {Pids, ?DICTS:append(Reason, P, EStack)} - end; - (?restarting(_), _, {Pids, EStack}) -> - {Pids, EStack} - end, {?SETS:new(), ?DICTS:new()}, Dynamics). - + dict:fold(fun(Reason, Ls, _) -> + report_error(shutdown_error, Reason, + Child#child{pid=Ls}, State#state.name) + end, ok, EStack). + +monitor_dynamic_children(Child,State) -> + dyn_fold(fun(P,{Pids, EStack}) when is_pid(P) -> + case monitor_child(P) of + ok -> + {sets:add_element(P, Pids), EStack}; + {error, normal} when not (?is_permanent(Child)) -> + {Pids, EStack}; + {error, Reason} -> + {Pids, dict:append(Reason, P, EStack)} + end; + (?restarting(_), {Pids, EStack}) -> + {Pids, EStack} + end, {sets:new(), dict:new()}, State). wait_dynamic_children(_Child, _Pids, 0, undefined, EStack) -> EStack; @@ -1073,39 +958,38 @@ wait_dynamic_children(#child{shutdown=brutal_kill} = Child, Pids, Sz, TRef, EStack) -> receive {'DOWN', _MRef, process, Pid, killed} -> - wait_dynamic_children(Child, ?SETS:del_element(Pid, Pids), Sz-1, + wait_dynamic_children(Child, sets:del_element(Pid, Pids), Sz-1, TRef, EStack); {'DOWN', _MRef, process, Pid, Reason} -> - wait_dynamic_children(Child, ?SETS:del_element(Pid, Pids), Sz-1, - TRef, ?DICTS:append(Reason, Pid, EStack)) + wait_dynamic_children(Child, sets:del_element(Pid, Pids), Sz-1, + TRef, dict:append(Reason, Pid, EStack)) end; -wait_dynamic_children(#child{restart_type=RType} = Child, Pids, Sz, - TRef, EStack) -> +wait_dynamic_children(Child, Pids, Sz, TRef, EStack) -> receive {'DOWN', _MRef, process, Pid, shutdown} -> - wait_dynamic_children(Child, ?SETS:del_element(Pid, Pids), Sz-1, + wait_dynamic_children(Child, sets:del_element(Pid, Pids), Sz-1, TRef, EStack); {'DOWN', _MRef, process, Pid, {shutdown, _}} -> - wait_dynamic_children(Child, ?SETS:del_element(Pid, Pids), Sz-1, + wait_dynamic_children(Child, sets:del_element(Pid, Pids), Sz-1, TRef, EStack); - {'DOWN', _MRef, process, Pid, normal} when RType =/= permanent -> - wait_dynamic_children(Child, ?SETS:del_element(Pid, Pids), Sz-1, + {'DOWN', _MRef, process, Pid, normal} when not (?is_permanent(Child)) -> + wait_dynamic_children(Child, sets:del_element(Pid, Pids), Sz-1, TRef, EStack); {'DOWN', _MRef, process, Pid, Reason} -> - wait_dynamic_children(Child, ?SETS:del_element(Pid, Pids), Sz-1, - TRef, ?DICTS:append(Reason, Pid, EStack)); + wait_dynamic_children(Child, sets:del_element(Pid, Pids), Sz-1, + TRef, dict:append(Reason, Pid, EStack)); {timeout, TRef, kill} -> - ?SETS:fold(fun(P, _) -> exit(P, kill) end, ok, Pids), + sets:fold(fun(P, _) -> exit(P, kill) end, ok, Pids), wait_dynamic_children(Child, Pids, Sz, undefined, EStack) end. %%----------------------------------------------------------------- -%% Child/State manipulating functions. +%% Access #state.children %%----------------------------------------------------------------- %% Note we do not want to save the parameter list for temporary processes as @@ -1113,114 +997,184 @@ wait_dynamic_children(#child{restart_type=RType} = Child, Pids, Sz, %% Especially for dynamic children to simple_one_for_one supervisors %% it could become very costly as it is not uncommon to spawn %% very many such processes. -save_child(#child{restart_type = temporary, - mfargs = {M, F, _}} = Child, #state{children = Children} = State) -> - State#state{children = [Child#child{mfargs = {M, F, undefined}} |Children]}; -save_child(Child, #state{children = Children} = State) -> - State#state{children = [Child |Children]}. - -save_dynamic_child(temporary, Pid, _, #state{dynamics = Dynamics} = State) -> - DynamicsDb = dynamics_db(temporary, Dynamics), - State#state{dynamics = {set, ?SETS:add_element(Pid, DynamicsDb)}}; -save_dynamic_child(RestartType, Pid, Args, #state{dynamics = Dynamics} = State) -> - DynamicsDb = dynamics_db(RestartType, Dynamics), - State#state{dynamics = {dict, ?DICTS:store(Pid, Args, DynamicsDb)}}. - -dynamics_db(temporary, undefined) -> - ?SETS:new(); -dynamics_db(_, undefined) -> - ?DICTS:new(); -dynamics_db(_, {_Tag, DynamicsDb}) -> - DynamicsDb. - -dynamic_child_args(_Pid, temporary, _DynamicsDb) -> - {ok, undefined}; -dynamic_child_args(Pid, _RT, {dict, DynamicsDb}) -> - ?DICTS:find(Pid, DynamicsDb); -dynamic_child_args(_Pid, _RT, undefined) -> - error. - -state_del_child(#child{pid = Pid, restart_type = temporary}, State) when ?is_simple(State) -> - NDynamics = ?SETS:del_element(Pid, dynamics_db(temporary, State#state.dynamics)), - State#state{dynamics = {set, NDynamics}}; -state_del_child(#child{pid = Pid, restart_type = RType}, State) when ?is_simple(State) -> - NDynamics = ?DICTS:erase(Pid, dynamics_db(RType, State#state.dynamics)), - State#state{dynamics = {dict, NDynamics}}; -state_del_child(Child, State) -> - NChildren = del_child(Child#child.name, State#state.children), - State#state{children = NChildren}. - -del_child(Name, [Ch|Chs]) when Ch#child.name =:= Name, Ch#child.restart_type =:= temporary -> - Chs; -del_child(Name, [Ch|Chs]) when Ch#child.name =:= Name -> - [Ch#child{pid = undefined} | Chs]; -del_child(Pid, [Ch|Chs]) when Ch#child.pid =:= Pid, Ch#child.restart_type =:= temporary -> - Chs; -del_child(Pid, [Ch|Chs]) when Ch#child.pid =:= Pid -> - [Ch#child{pid = undefined} | Chs]; -del_child(Name, [Ch|Chs]) -> - [Ch|del_child(Name, Chs)]; -del_child(_, []) -> - []. +-spec save_child(child_rec(), state()) -> state(). +save_child(#child{mfargs = {M, F, _}} = Child, State) when ?is_temporary(Child) -> + do_save_child(Child#child{mfargs = {M, F, undefined}}, State); +save_child(Child, State) -> + do_save_child(Child, State). + +-spec do_save_child(child_rec(), state()) -> state(). +do_save_child(#child{id = Id} = Child, #state{children = {Ids,Db}} = State) -> + State#state{children = {[Id|Ids],Db#{Id => Child}}}. + +-spec del_child(child_rec(), state()) -> state(); + (child_id(), children()) -> children(). +del_child(#child{pid = Pid}, State) when ?is_simple(State) -> + dyn_erase(Pid,State); +del_child(Child, State) when is_record(Child,child), is_record(State,state) -> + NChildren = del_child(Child#child.id, State#state.children), + State#state{children = NChildren}; +del_child(Id, {Ids,Db}) -> + case maps:get(Id, Db) of + Child when Child#child.restart_type =:= temporary -> + {lists:delete(Id, Ids), maps:remove(Id, Db)}; + Child -> + {Ids, Db#{Id=>Child#child{pid=undefined}}} + end. -%% Chs = [S4, S3, Ch, S1, S0] -%% Ret: {[S4, S3, Ch], [S1, S0]} -split_child(Name, Chs) -> - split_child(Name, Chs, []). - -split_child(Name, [Ch|Chs], After) when Ch#child.name =:= Name -> - {lists:reverse([Ch#child{pid = undefined} | After]), Chs}; -split_child(Pid, [Ch|Chs], After) when Ch#child.pid =:= Pid -> - {lists:reverse([Ch#child{pid = undefined} | After]), Chs}; -split_child(Name, [Ch|Chs], After) -> - split_child(Name, Chs, [Ch | After]); -split_child(_, [], After) -> - {lists:reverse(After), []}. - -get_child(Name, State) -> - get_child(Name, State, false). - -get_child(Pid, State, AllowPid) when AllowPid, is_pid(Pid) -> - get_dynamic_child(Pid, State); -get_child(Name, State, _) -> - lists:keysearch(Name, #child.name, State#state.children). - -get_dynamic_child(Pid, #state{children=[Child], dynamics=Dynamics}) -> - case is_dynamic_pid(Pid, Dynamics) of - true -> - {value, Child#child{pid=Pid}}; - false -> - RPid = restarting(Pid), - case is_dynamic_pid(RPid, Dynamics) of - true -> - {value, Child#child{pid=RPid}}; - false -> +%% In: {[S4, S3, Ch, S1, S0],Db} +%% Ret: {{[S4, S3, Ch],Db1}, {[S1, S0],Db2}} +%% Db1 and Db2 contain the keys in the lists they are associated with. +-spec split_child(child_id(), children()) -> {children(), children()}. +split_child(Id, {Ids,Db}) -> + {IdsAfter,IdsBefore} = split_ids(Id, Ids, []), + DbBefore = maps:with(IdsBefore,Db), + #{Id:=Ch} = DbAfter = maps:with(IdsAfter,Db), + {{IdsAfter,DbAfter#{Id=>Ch#child{pid=undefined}}},{IdsBefore,DbBefore}}. + +split_ids(Id, [Id|Ids], After) -> + {lists:reverse([Id|After]), Ids}; +split_ids(Id, [Other|Ids], After) -> + split_ids(Id, Ids, [Other | After]). + +%% Find the child record for a given Pid (dynamic child) or Id +%% (non-dynamic child). This is called from the API functions. +-spec find_child(pid() | child_id(), state()) -> {ok,child_rec()} | error. +find_child(Pid, State) when is_pid(Pid), ?is_simple(State) -> + case find_dynamic_child(Pid, State) of + error -> + case find_dynamic_child(restarting(Pid), State) of + error -> case erlang:is_process_alive(Pid) of - true -> false; - false -> {value, Child} - end - end + true -> error; + false -> {ok, get_dynamic_child(State)} + end; + Other -> + Other + end; + Other -> + Other + end; +find_child(Id, #state{children = {_Ids,Db}}) -> + maps:find(Id, Db). + +%% Get the child record - either by child id or by pid. If +%% simple_one_for_one, then insert the pid and args into the returned +%% child record. This is called when trying to restart the child. +-spec find_child_and_args(IdOrPid, state()) -> {ok, child_rec()} | error when + IdOrPid :: pid() | {restarting,pid()} | child_id(). +find_child_and_args(Pid, State) when ?is_simple(State) -> + case find_dynamic_child(Pid, State) of + {ok,#child{mfargs={M,F,_}} = Child} -> + {ok, Args} = dyn_args(Pid, State), + {ok, Child#child{mfargs = {M, F, Args}}}; + error -> + error + end; +find_child_and_args(Pid, State) when is_pid(Pid) -> + find_child_by_pid(Pid, State); +find_child_and_args(Id, #state{children={_Ids,Db}}) -> + maps:find(Id, Db). + +%% Given the pid, find the child record for a dynamic child, and +%% include the pid in the returned record. +-spec find_dynamic_child(IdOrPid, state()) -> {ok, child_rec()} | error when + IdOrPid :: pid() | {restarting,pid()} | child_id(). +find_dynamic_child(Pid, State) -> + case dyn_exists(Pid, State) of + true -> + Child = get_dynamic_child(State), + {ok, Child#child{pid=Pid}}; + false -> + error end. -is_dynamic_pid(Pid, {dict, Dynamics}) -> - ?DICTS:is_key(Pid, Dynamics); -is_dynamic_pid(Pid, {set, Dynamics}) -> - ?SETS:is_element(Pid, Dynamics); -is_dynamic_pid(_Pid, undefined) -> - false. - -replace_child(Child, State) -> - Chs = do_replace_child(Child, State#state.children), - State#state{children = Chs}. - -do_replace_child(Child, [Ch|Chs]) when Ch#child.name =:= Child#child.name -> - [Child | Chs]; -do_replace_child(Child, [Ch|Chs]) -> - [Ch|do_replace_child(Child, Chs)]. +%% Given the pid, find the child record for a non-dyanamic child. +-spec find_child_by_pid(IdOrPid, state()) -> {ok,child_rec()} | error when + IdOrPid :: pid() | {restarting,pid()}. +find_child_by_pid(Pid,#state{children={_Ids,Db}}) -> + Fun = fun(_Id,#child{pid=P}=Ch,_) when P =:= Pid -> + throw(Ch); + (_,_,error) -> + error + end, + try maps:fold(Fun,error,Db) + catch throw:Child -> {ok,Child} + end. -remove_child(Child, State) -> - Chs = lists:keydelete(Child#child.name, #child.name, State#state.children), - State#state{children = Chs}. +%% Get the child record from a simple_one_for_one supervisor - no pid +%% It is assumed that the child can always be found +-spec get_dynamic_child(state()) -> child_rec(). +get_dynamic_child(#state{children={[Id],Db}}) -> + #{Id := Child} = Db, + Child. + +%% Update pid in the given child record and store it in the process state +-spec set_pid(term(), child_id(), state()) -> state(); + (term(), child_id(), children()) -> children(). +set_pid(Pid, Id, #state{children=Children} = State) -> + State#state{children = set_pid(Pid, Id, Children)}; +set_pid(Pid, Id, {Ids, Db}) -> + NewDb = maps:update_with(Id, fun(Child) -> Child#child{pid=Pid} end, Db), + {Ids,NewDb}. + +%% Remove the Id and the child record from the process state +-spec remove_child(child_id(), state()) -> state(). +remove_child(Id, #state{children={Ids,Db}} = State) -> + NewIds = lists:delete(Id,Ids), + NewDb = maps:remove(Id,Db), + State#state{children = {NewIds,NewDb}}. + +%% In the order of Ids, traverse the children and update each child +%% according to the return value of the Fun. +%% On error, abort and return the merge of the old and the updated map. +%% NOTE: The returned list of Ids is reverted compared to the input. +-spec children_map(Fun, children()) -> {ok, children()} | + {error,children(),Reason} when + Fun :: fun((child_id(),child_rec()) -> {update,child_rec()} | + remove | + {abort, Reason}), + Reason :: term(). +children_map(Fun,{Ids,Db}) -> + children_map(Fun, Ids, Db, []). + +children_map(Fun,[Id|Ids],Db,Acc) -> + case Fun(Id,maps:get(Id,Db)) of + {update,Child} -> + children_map(Fun,Ids,Db#{Id => Child},[Id|Acc]); + remove -> + children_map(Fun,Ids,maps:remove(Id,Db),Acc); + {abort,Reason} -> + {error,{lists:reverse(Ids)++[Id|Acc],Db},Reason} + end; +children_map(_Fun,[],Db,Acc) -> + {ok,{Acc,Db}}. + +%% In the order of Ids, map over all children and return the list +-spec children_to_list(Fun, children()) -> List when + Fun :: fun((child_id(), child_rec()) -> Elem), + List :: list(Elem), + Elem :: term(). +children_to_list(Fun,{Ids,Db}) -> + children_to_list(Fun, Ids, Db, []). +children_to_list(Fun,[Id|Ids],Db,Acc) -> + children_to_list(Fun,Ids,Db,[Fun(Id,maps:get(Id,Db))|Acc]); +children_to_list(_Fun,[],_Db,Acc) -> + lists:reverse(Acc). + +%% The order is not important - so ignore Ids +-spec children_fold(Fun, Acc0, children()) -> Acc1 when + Fun :: fun((child_id(), child_rec(), AccIn) -> AccOut), + Acc0 :: term(), + Acc1 :: term(), + AccIn :: term(), + AccOut :: term(). +children_fold(Fun,Init,{_Ids,Db}) -> + maps:fold(Fun, Init, Db). + +-spec append(children(), children()) -> children(). +append({Ids1,Db1},{Ids2,Db2}) -> + {Ids1++Ids2,maps:merge(Db1,Db2)}. %%----------------------------------------------------------------- %% Func: init_state/4 @@ -1290,27 +1244,27 @@ supname(N, _) -> N. %%% Returns: {ok, [child_rec()]} | Error %%% ------------------------------------------------------ -check_startspec(Children) -> check_startspec(Children, []). +check_startspec(Children) -> check_startspec(Children, [], #{}). -check_startspec([ChildSpec|T], Res) -> +check_startspec([ChildSpec|T], Ids, Db) -> case check_childspec(ChildSpec) of - {ok, Child} -> - case lists:keymember(Child#child.name, #child.name, Res) of + {ok, #child{id=Id}=Child} -> + case maps:is_key(Id, Db) of %% The error message duplicate_child_name is kept for %% backwards compatibility, although %% duplicate_child_id would be more correct. - true -> {duplicate_child_name, Child#child.name}; - false -> check_startspec(T, [Child | Res]) + true -> {duplicate_child_name, Id}; + false -> check_startspec(T, [Id | Ids], Db#{Id=>Child}) end; Error -> Error end; -check_startspec([], Res) -> - {ok, lists:reverse(Res)}. +check_startspec([], Ids, Db) -> + {ok, {lists:reverse(Ids),Db}}. check_childspec(ChildSpec) when is_map(ChildSpec) -> catch do_check_childspec(maps:merge(?default_child_spec,ChildSpec)); -check_childspec({Name, Func, RestartType, Shutdown, ChildType, Mods}) -> - check_childspec(#{id => Name, +check_childspec({Id, Func, RestartType, Shutdown, ChildType, Mods}) -> + check_childspec(#{id => Id, start => Func, restart => RestartType, shutdown => Shutdown, @@ -1320,15 +1274,15 @@ check_childspec(X) -> {invalid_child_spec, X}. do_check_childspec(#{restart := RestartType, type := ChildType} = ChildSpec)-> - Name = case ChildSpec of - #{id := N} -> N; + Id = case ChildSpec of + #{id := I} -> I; _ -> throw(missing_id) end, Func = case ChildSpec of #{start := F} -> F; _ -> throw(missing_start) end, - validName(Name), + validId(Id), validFunc(Func), validRestartType(RestartType), validChildType(ChildType), @@ -1343,14 +1297,14 @@ do_check_childspec(#{restart := RestartType, _ -> {M,_,_} = Func, [M] end, validMods(Mods), - {ok, #child{name = Name, mfargs = Func, restart_type = RestartType, + {ok, #child{id = Id, mfargs = Func, restart_type = RestartType, shutdown = Shutdown, child_type = ChildType, modules = Mods}}. validChildType(supervisor) -> true; validChildType(worker) -> true; validChildType(What) -> throw({invalid_child_type, What}). -validName(_Name) -> true. +validId(_Id) -> true. validFunc({M, F, A}) when is_atom(M), is_atom(F), @@ -1379,13 +1333,13 @@ validMods(Mods) when is_list(Mods) -> Mods); validMods(Mods) -> throw({invalid_modules, Mods}). -child_to_spec(#child{name = Name, +child_to_spec(#child{id = Id, mfargs = Func, restart_type = RestartType, shutdown = Shutdown, child_type = ChildType, modules = Mods}) -> - #{id => Name, + #{id => Id, start => Func, restart => RestartType, shutdown => Shutdown, @@ -1439,17 +1393,16 @@ report_error(Error, Reason, Child, SupName) -> {offender, extract_child(Child)}], error_logger:error_report(supervisor_report, ErrorMsg). - extract_child(Child) when is_list(Child#child.pid) -> [{nb_children, length(Child#child.pid)}, - {id, Child#child.name}, + {id, Child#child.id}, {mfargs, Child#child.mfargs}, {restart_type, Child#child.restart_type}, {shutdown, Child#child.shutdown}, {child_type, Child#child.child_type}]; extract_child(Child) -> [{pid, Child#child.pid}, - {id, Child#child.name}, + {id, Child#child.id}, {mfargs, Child#child.mfargs}, {restart_type, Child#child.restart_type}, {shutdown, Child#child.shutdown}, @@ -1465,3 +1418,46 @@ format_status(terminate, [_PDict, State]) -> format_status(_, [_PDict, State]) -> [{data, [{"State", State}]}, {supervisor, [{"Callback", State#state.module}]}]. + +%%%----------------------------------------------------------------- +%%% Dynamics database access +dyn_size(#state{dynamics = {Mod,Db}}) -> + Mod:size(Db). + +dyn_erase(Pid,#state{dynamics={sets,Db}}=State) -> + State#state{dynamics={sets,sets:del_element(Pid,Db)}}; +dyn_erase(Pid,#state{dynamics={maps,Db}}=State) -> + State#state{dynamics={maps,maps:remove(Pid,Db)}}. + +dyn_store(Pid,_,#state{dynamics={sets,Db}}=State) -> + State#state{dynamics={sets,sets:add_element(Pid,Db)}}; +dyn_store(Pid,Args,#state{dynamics={maps,Db}}=State) -> + State#state{dynamics={maps,Db#{Pid => Args}}}. + +dyn_fold(Fun,Init,#state{dynamics={sets,Db}}) -> + sets:fold(Fun,Init,Db); +dyn_fold(Fun,Init,#state{dynamics={maps,Db}}) -> + maps:fold(fun(Pid,_,Acc) -> Fun(Pid,Acc) end, Init, Db). + +dyn_map(Fun, #state{dynamics={sets,Db}}) -> + lists:map(Fun, sets:to_list(Db)); +dyn_map(Fun, #state{dynamics={maps,Db}}) -> + lists:map(Fun, maps:keys(Db)). + +dyn_exists(Pid, #state{dynamics={sets, Db}}) -> + sets:is_element(Pid, Db); +dyn_exists(Pid, #state{dynamics={maps, Db}}) -> + maps:is_key(Pid, Db). + +dyn_args(_Pid, #state{dynamics={sets, _Db}}) -> + {ok,undefined}; +dyn_args(Pid, #state{dynamics={maps, Db}}) -> + maps:find(Pid, Db). + +dyn_init(State) -> + dyn_init(get_dynamic_child(State),State). + +dyn_init(Child,State) when ?is_temporary(Child) -> + State#state{dynamics={sets,sets:new()}}; +dyn_init(_Child,State) -> + State#state{dynamics={maps,maps:new()}}. diff --git a/lib/stdlib/src/uri_string.erl b/lib/stdlib/src/uri_string.erl new file mode 100644 index 0000000000..a84679c595 --- /dev/null +++ b/lib/stdlib/src/uri_string.erl @@ -0,0 +1,2069 @@ +%% +%% %CopyrightBegin% +%% +%% Copyright Ericsson AB 2017. All Rights Reserved. +%% +%% Licensed under the Apache License, Version 2.0 (the "License"); +%% you may not use this file except in compliance with the License. +%% You may obtain a copy of the License at +%% +%% http://www.apache.org/licenses/LICENSE-2.0 +%% +%% Unless required by applicable law or agreed to in writing, software +%% distributed under the License is distributed on an "AS IS" BASIS, +%% WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +%% See the License for the specific language governing permissions and +%% limitations under the License. +%% +%% %CopyrightEnd% +%% +%% +%% [RFC 3986, Chapter 2.2. Reserved Characters] +%% +%% reserved = gen-delims / sub-delims +%% +%% gen-delims = ":" / "/" / "?" / "#" / "[" / "]" / "@" +%% +%% sub-delims = "!" / "$" / "&" / "'" / "(" / ")" +%% / "*" / "+" / "," / ";" / "=" +%% +%% +%% [RFC 3986, Chapter 2.3. Unreserved Characters] +%% +%% unreserved = ALPHA / DIGIT / "-" / "." / "_" / "~" +%% +%% +%% [RFC 3986, Chapter 3. Syntax Components] +%% +%% The generic URI syntax consists of a hierarchical sequence of +%% components referred to as the scheme, authority, path, query, and +%% fragment. +%% +%% URI = scheme ":" hier-part [ "?" query ] [ "#" fragment ] +%% +%% hier-part = "//" authority path-abempty +%% / path-absolute +%% / path-rootless +%% / path-empty +%% +%% The scheme and path components are required, though the path may be +%% empty (no characters). When authority is present, the path must +%% either be empty or begin with a slash ("/") character. When +%% authority is not present, the path cannot begin with two slash +%% characters ("//"). These restrictions result in five different ABNF +%% rules for a path (Section 3.3), only one of which will match any +%% given URI reference. +%% +%% The following are two example URIs and their component parts: +%% +%% foo://example.com:8042/over/there?name=ferret#nose +%% \_/ \______________/\_________/ \_________/ \__/ +%% | | | | | +%% scheme authority path query fragment +%% | _____________________|__ +%% / \ / \ +%% urn:example:animal:ferret:nose +%% +%% +%% [RFC 3986, Chapter 3.1. Scheme] +%% +%% Each URI begins with a scheme name that refers to a specification for +%% assigning identifiers within that scheme. +%% +%% scheme = ALPHA *( ALPHA / DIGIT / "+" / "-" / "." ) +%% +%% +%% [RFC 3986, Chapter 3.2. Authority] +%% +%% Many URI schemes include a hierarchical element for a naming +%% authority so that governance of the name space defined by the +%% remainder of the URI is delegated to that authority (which may, in +%% turn, delegate it further). +%% +%% authority = [ userinfo "@" ] host [ ":" port ] +%% +%% +%% [RFC 3986, Chapter 3.2.1. User Information] +%% +%% The userinfo subcomponent may consist of a user name and, optionally, +%% scheme-specific information about how to gain authorization to access +%% the resource. The user information, if present, is followed by a +%% commercial at-sign ("@") that delimits it from the host. +%% +%% userinfo = *( unreserved / pct-encoded / sub-delims / ":" ) +%% +%% +%% [RFC 3986, Chapter 3.2.2. Host] +%% +%% The host subcomponent of authority is identified by an IP literal +%% encapsulated within square brackets, an IPv4 address in dotted- +%% decimal form, or a registered name. +%% +%% host = IP-literal / IPv4address / reg-name +%% +%% IP-literal = "[" ( IPv6address / IPvFuture ) "]" +%% +%% IPvFuture = "v" 1*HEXDIG "." 1*( unreserved / sub-delims / ":" ) +%% +%% IPv6address = 6( h16 ":" ) ls32 +%% / "::" 5( h16 ":" ) ls32 +%% / [ h16 ] "::" 4( h16 ":" ) ls32 +%% / [ *1( h16 ":" ) h16 ] "::" 3( h16 ":" ) ls32 +%% / [ *2( h16 ":" ) h16 ] "::" 2( h16 ":" ) ls32 +%% / [ *3( h16 ":" ) h16 ] "::" h16 ":" ls32 +%% / [ *4( h16 ":" ) h16 ] "::" ls32 +%% / [ *5( h16 ":" ) h16 ] "::" h16 +%% / [ *6( h16 ":" ) h16 ] "::" +%% +%% ls32 = ( h16 ":" h16 ) / IPv4address +%% ; least-significant 32 bits of address +%% +%% h16 = 1*4HEXDIG +%% ; 16 bits of address represented in hexadecimal +%% +%% IPv4address = dec-octet "." dec-octet "." dec-octet "." dec-octet +%% +%% dec-octet = DIGIT ; 0-9 +%% / %x31-39 DIGIT ; 10-99 +%% / "1" 2DIGIT ; 100-199 +%% / "2" %x30-34 DIGIT ; 200-249 +%% / "25" %x30-35 ; 250-255 +%% +%% reg-name = *( unreserved / pct-encoded / sub-delims ) +%% +%% +%% [RFC 3986, Chapter 3.2.2. Port] +%% +%% The port subcomponent of authority is designated by an optional port +%% number in decimal following the host and delimited from it by a +%% single colon (":") character. +%% +%% port = *DIGIT +%% +%% +%% [RFC 3986, Chapter 3.3. Path] +%% +%% The path component contains data, usually organized in hierarchical +%% form, that, along with data in the non-hierarchical query component +%% (Section 3.4), serves to identify a resource within the scope of the +%% URI's scheme and naming authority (if any). The path is terminated +%% by the first question mark ("?") or number sign ("#") character, or +%% by the end of the URI. +%% +%% path = path-abempty ; begins with "/" or is empty +%% / path-absolute ; begins with "/" but not "//" +%% / path-noscheme ; begins with a non-colon segment +%% / path-rootless ; begins with a segment +%% / path-empty ; zero characters +%% +%% path-abempty = *( "/" segment ) +%% path-absolute = "/" [ segment-nz *( "/" segment ) ] +%% path-noscheme = segment-nz-nc *( "/" segment ) +%% path-rootless = segment-nz *( "/" segment ) +%% path-empty = 0<pchar> +%% segment = *pchar +%% segment-nz = 1*pchar +%% segment-nz-nc = 1*( unreserved / pct-encoded / sub-delims / "@" ) +%% ; non-zero-length segment without any colon ":" +%% +%% pchar = unreserved / pct-encoded / sub-delims / ":" / "@" +%% +%% +%% [RFC 3986, Chapter 3.4. Query] +%% +%% The query component contains non-hierarchical data that, along with +%% data in the path component (Section 3.3), serves to identify a +%% resource within the scope of the URI's scheme and naming authority +%% (if any). The query component is indicated by the first question +%% mark ("?") character and terminated by a number sign ("#") character +%% or by the end of the URI. +%% +%% query = *( pchar / "/" / "?" ) +%% +%% +%% [RFC 3986, Chapter 3.5. Fragment] +%% +%% The fragment identifier component of a URI allows indirect +%% identification of a secondary resource by reference to a primary +%% resource and additional identifying information. +%% +%% fragment = *( pchar / "/" / "?" ) +%% +%% +%% [RFC 3986, Chapter 4.1. URI Reference] +%% +%% URI-reference is used to denote the most common usage of a resource +%% identifier. +%% +%% URI-reference = URI / relative-ref +%% +%% +%% [RFC 3986, Chapter 4.2. Relative Reference] +%% +%% A relative reference takes advantage of the hierarchical syntax +%% (Section 1.2.3) to express a URI reference relative to the name space +%% of another hierarchical URI. +%% +%% relative-ref = relative-part [ "?" query ] [ "#" fragment ] +%% +%% relative-part = "//" authority path-abempty +%% / path-absolute +%% / path-noscheme +%% / path-empty +%% +%% +%% [RFC 3986, Chapter 4.3. Absolute URI] +%% +%% Some protocol elements allow only the absolute form of a URI without +%% a fragment identifier. For example, defining a base URI for later +%% use by relative references calls for an absolute-URI syntax rule that +%% does not allow a fragment. +%% +%% absolute-URI = scheme ":" hier-part [ "?" query ] +%% +-module(uri_string). + +%%------------------------------------------------------------------------- +%% External API +%%------------------------------------------------------------------------- +-export([compose_query/1, compose_query/2, + dissect_query/1, normalize/1, parse/1, + recompose/1, transcode/2]). +-export_type([error/0, uri_map/0, uri_string/0]). + + +%%------------------------------------------------------------------------- +%% Internal API +%%------------------------------------------------------------------------- +-export([is_host/1, is_path/1]). % suppress warnings + + +%%------------------------------------------------------------------------- +%% Macros +%%------------------------------------------------------------------------- +-define(CHAR(Char), <<Char/utf8>>). +-define(STRING_EMPTY, <<>>). +-define(STRING(MatchStr), <<MatchStr/binary>>). +-define(STRING_REST(MatchStr, Rest), <<MatchStr/utf8, Rest/binary>>). + +-define(DEC2HEX(X), + if ((X) >= 0) andalso ((X) =< 9) -> (X) + $0; + ((X) >= 10) andalso ((X) =< 15) -> (X) + $A - 10 + end). + +-define(HEX2DEC(X), + if ((X) >= $0) andalso ((X) =< $9) -> (X) - $0; + ((X) >= $A) andalso ((X) =< $F) -> (X) - $A + 10; + ((X) >= $a) andalso ((X) =< $f) -> (X) - $a + 10 + end). + + +%%%========================================================================= +%%% API +%%%========================================================================= + +%%------------------------------------------------------------------------- +%% URI compliant with RFC 3986 +%% ASCII %x21 - %x7A ("!" - "z") except +%% %x34 " double quote +%% %x60 < less than +%% %x62 > greater than +%% %x92 \ backslash +%% %x94 ^ caret / circumflex +%% %x96 ` grave / accent +%%------------------------------------------------------------------------- +-type uri_string() :: iodata(). +-type error() :: {error, atom(), term()}. + + +%%------------------------------------------------------------------------- +%% RFC 3986, Chapter 3. Syntax Components +%%------------------------------------------------------------------------- +-type uri_map() :: + #{fragment => unicode:chardata(), + host => unicode:chardata(), + path => unicode:chardata(), + port => non_neg_integer() | undefined, + query => unicode:chardata(), + scheme => unicode:chardata(), + userinfo => unicode:chardata()} | #{}. + + +%%------------------------------------------------------------------------- +%% Normalize URIs +%%------------------------------------------------------------------------- +-spec normalize(URIString) -> NormalizedURI when + URIString :: uri_string(), + NormalizedURI :: uri_string(). +normalize(URIString) -> + %% Percent-encoding normalization and case normalization for + %% percent-encoded triplets are achieved by running parse and + %% recompose on the input URI string. + recompose( + normalize_path_segment( + normalize_scheme_based( + normalize_case( + parse(URIString))))). + + +%%------------------------------------------------------------------------- +%% Parse URIs +%%------------------------------------------------------------------------- +-spec parse(URIString) -> URIMap when + URIString :: uri_string(), + URIMap :: uri_map() + | error(). +parse(URIString) when is_binary(URIString) -> + try parse_uri_reference(URIString, #{}) + catch + throw:{error, Atom, RestData} -> {error, Atom, RestData} + end; +parse(URIString) when is_list(URIString) -> + try + Binary = unicode:characters_to_binary(URIString), + Map = parse_uri_reference(Binary, #{}), + convert_mapfields_to_list(Map) + catch + throw:{error, Atom, RestData} -> {error, Atom, RestData} + end. + + +%%------------------------------------------------------------------------- +%% Recompose URIs +%%------------------------------------------------------------------------- +-spec recompose(URIMap) -> URIString when + URIMap :: uri_map(), + URIString :: uri_string() + | error(). +recompose(Map) -> + case is_valid_map(Map) of + false -> + {error, invalid_map, Map}; + true -> + try + T0 = update_scheme(Map, empty), + T1 = update_userinfo(Map, T0), + T2 = update_host(Map, T1), + T3 = update_port(Map, T2), + T4 = update_path(Map, T3), + T5 = update_query(Map, T4), + update_fragment(Map, T5) + catch + throw:{error, Atom, RestData} -> {error, Atom, RestData} + end + end. + + +%%------------------------------------------------------------------------- +%% Transcode URIs +%%------------------------------------------------------------------------- +-spec transcode(URIString, Options) -> Result when + URIString :: uri_string(), + Options :: [{in_encoding, unicode:encoding()}|{out_encoding, unicode:encoding()}], + Result :: uri_string() + | error(). +transcode(URIString, Options) when is_binary(URIString) -> + try + InEnc = proplists:get_value(in_encoding, Options, utf8), + OutEnc = proplists:get_value(out_encoding, Options, utf8), + List = convert_to_list(URIString, InEnc), + Output = transcode(List, [], InEnc, OutEnc), + convert_to_binary(Output, utf8, OutEnc) + catch + throw:{error, Atom, RestData} -> {error, Atom, RestData} + end; +transcode(URIString, Options) when is_list(URIString) -> + InEnc = proplists:get_value(in_encoding, Options, utf8), + OutEnc = proplists:get_value(out_encoding, Options, utf8), + Flattened = flatten_list(URIString, InEnc), + try transcode(Flattened, [], InEnc, OutEnc) + catch + throw:{error, Atom, RestData} -> {error, Atom, RestData} + end. + + +%%------------------------------------------------------------------------- +%% Functions for working with the query part of a URI as a list +%% of key/value pairs. +%% HTML5 - 4.10.22.6 URL-encoded form data +%%------------------------------------------------------------------------- + +%%------------------------------------------------------------------------- +%% Compose urlencoded query string from a list of unescaped key/value pairs. +%% (application/x-www-form-urlencoded encoding algorithm) +%%------------------------------------------------------------------------- +-spec compose_query(QueryList) -> QueryString when + QueryList :: [{uri_string(), uri_string()}], + QueryString :: uri_string() + | error(). +compose_query(List) -> + compose_query(List, [{encoding, utf8}]). + + +-spec compose_query(QueryList, Options) -> QueryString when + QueryList :: [{uri_string(), uri_string()}], + Options :: [{encoding, atom()}], + QueryString :: uri_string() + | error(). +compose_query([],_Options) -> + []; +compose_query(List, Options) -> + try compose_query(List, Options, false, <<>>) + catch + throw:{error, Atom, RestData} -> {error, Atom, RestData} + end. +%% +compose_query([{Key,Value}|Rest], Options, IsList, Acc) -> + Separator = get_separator(Rest), + K = form_urlencode(Key, Options), + V = form_urlencode(Value, Options), + IsListNew = IsList orelse is_list(Key) orelse is_list(Value), + compose_query(Rest, Options, IsListNew, <<Acc/binary,K/binary,"=",V/binary,Separator/binary>>); +compose_query([], _Options, IsList, Acc) -> + case IsList of + true -> convert_to_list(Acc, utf8); + false -> Acc + end. + + +%%------------------------------------------------------------------------- +%% Dissect a query string into a list of unescaped key/value pairs. +%% (application/x-www-form-urlencoded decoding algorithm) +%%------------------------------------------------------------------------- +-spec dissect_query(QueryString) -> QueryList when + QueryString :: uri_string(), + QueryList :: [{uri_string(), uri_string()}] + | error(). +dissect_query(<<>>) -> + []; +dissect_query([]) -> + []; +dissect_query(QueryString) when is_list(QueryString) -> + try + B = convert_to_binary(QueryString, utf8, utf8), + dissect_query_key(B, true, [], <<>>, <<>>) + catch + throw:{error, Atom, RestData} -> {error, Atom, RestData} + end; +dissect_query(QueryString) -> + try dissect_query_key(QueryString, false, [], <<>>, <<>>) + catch + throw:{error, Atom, RestData} -> {error, Atom, RestData} + end. + + +%%%======================================================================== +%%% Internal functions +%%%======================================================================== + +%%------------------------------------------------------------------------- +%% Converts Map fields to lists +%%------------------------------------------------------------------------- +convert_mapfields_to_list(Map) -> + Fun = fun (_, V) when is_binary(V) -> unicode:characters_to_list(V); + (_, V) -> V end, + maps:map(Fun, Map). + + +%%------------------------------------------------------------------------- +%% [RFC 3986, Chapter 4.1. URI Reference] +%% +%% URI-reference is used to denote the most common usage of a resource +%% identifier. +%% +%% URI-reference = URI / relative-ref +%%------------------------------------------------------------------------- +-spec parse_uri_reference(binary(), uri_map()) -> uri_map(). +parse_uri_reference(<<>>, _) -> #{path => <<>>}; +parse_uri_reference(URIString, URI) -> + try parse_scheme_start(URIString, URI) + catch + throw:{_,_,_} -> + parse_relative_part(URIString, URI) + end. + + +%%------------------------------------------------------------------------- +%% [RFC 3986, Chapter 4.2. Relative Reference] +%% +%% A relative reference takes advantage of the hierarchical syntax +%% (Section 1.2.3) to express a URI reference relative to the name space +%% of another hierarchical URI. +%% +%% relative-ref = relative-part [ "?" query ] [ "#" fragment ] +%% +%% relative-part = "//" authority path-abempty +%% / path-absolute +%% / path-noscheme +%% / path-empty +%%------------------------------------------------------------------------- +-spec parse_relative_part(binary(), uri_map()) -> uri_map(). +parse_relative_part(?STRING_REST("//", Rest), URI) -> + %% Parse userinfo - "//" is NOT part of authority + try parse_userinfo(Rest, URI) of + {T, URI1} -> + Userinfo = calculate_parsed_userinfo(Rest, T), + URI2 = maybe_add_path(URI1), + URI2#{userinfo => decode_userinfo(Userinfo)} + catch + throw:{_,_,_} -> + {T, URI1} = parse_host(Rest, URI), + Host = calculate_parsed_host_port(Rest, T), + URI2 = maybe_add_path(URI1), + URI2#{host => decode_host(remove_brackets(Host))} + end; +parse_relative_part(?STRING_REST($/, Rest), URI) -> + {T, URI1} = parse_segment(Rest, URI), % path-absolute + Path = calculate_parsed_part(Rest, T), + URI1#{path => decode_path(?STRING_REST($/, Path))}; +parse_relative_part(?STRING_REST($?, Rest), URI) -> + {T, URI1} = parse_query(Rest, URI), % path-empty ?query + Query = calculate_parsed_query_fragment(Rest, T), + URI2 = maybe_add_path(URI1), + URI2#{query => decode_query(Query)}; +parse_relative_part(?STRING_REST($#, Rest), URI) -> + {T, URI1} = parse_fragment(Rest, URI), % path-empty + Fragment = calculate_parsed_query_fragment(Rest, T), + URI2 = maybe_add_path(URI1), + URI2#{fragment => decode_fragment(Fragment)}; +parse_relative_part(?STRING_REST(Char, Rest), URI) -> + case is_segment_nz_nc(Char) of + true -> + {T, URI1} = parse_segment_nz_nc(Rest, URI), % path-noscheme + Path = calculate_parsed_part(Rest, T), + URI1#{path => decode_path(?STRING_REST(Char, Path))}; + false -> throw({error,invalid_uri,[Char]}) + end. + + +%%------------------------------------------------------------------------- +%% [RFC 3986, Chapter 3.3. Path] +%% +%% The path component contains data, usually organized in hierarchical +%% form, that, along with data in the non-hierarchical query component +%% (Section 3.4), serves to identify a resource within the scope of the +%% URI's scheme and naming authority (if any). The path is terminated +%% by the first question mark ("?") or number sign ("#") character, or +%% by the end of the URI. +%% +%% path = path-abempty ; begins with "/" or is empty +%% / path-absolute ; begins with "/" but not "//" +%% / path-noscheme ; begins with a non-colon segment +%% / path-rootless ; begins with a segment +%% / path-empty ; zero characters +%% +%% path-abempty = *( "/" segment ) +%% path-absolute = "/" [ segment-nz *( "/" segment ) ] +%% path-noscheme = segment-nz-nc *( "/" segment ) +%% path-rootless = segment-nz *( "/" segment ) +%% path-empty = 0<pchar> +%% segment = *pchar +%% segment-nz = 1*pchar +%% segment-nz-nc = 1*( unreserved / pct-encoded / sub-delims / "@" ) +%% ; non-zero-length segment without any colon ":" +%% +%% pchar = unreserved / pct-encoded / sub-delims / ":" / "@" +%%------------------------------------------------------------------------- + +%%------------------------------------------------------------------------- +%% path-abempty +%%------------------------------------------------------------------------- +-spec parse_segment(binary(), uri_map()) -> {binary(), uri_map()}. +parse_segment(?STRING_REST($/, Rest), URI) -> + parse_segment(Rest, URI); % segment +parse_segment(?STRING_REST($?, Rest), URI) -> + {T, URI1} = parse_query(Rest, URI), % ?query + Query = calculate_parsed_query_fragment(Rest, T), + {Rest, URI1#{query => decode_query(Query)}}; +parse_segment(?STRING_REST($#, Rest), URI) -> + {T, URI1} = parse_fragment(Rest, URI), + Fragment = calculate_parsed_query_fragment(Rest, T), + {Rest, URI1#{fragment => decode_fragment(Fragment)}}; +parse_segment(?STRING_REST(Char, Rest), URI) -> + case is_pchar(Char) of + true -> parse_segment(Rest, URI); + false -> throw({error,invalid_uri,[Char]}) + end; +parse_segment(?STRING_EMPTY, URI) -> + {?STRING_EMPTY, URI}. + + +%%------------------------------------------------------------------------- +%% path-noscheme +%%------------------------------------------------------------------------- +-spec parse_segment_nz_nc(binary(), uri_map()) -> {binary(), uri_map()}. +parse_segment_nz_nc(?STRING_REST($/, Rest), URI) -> + parse_segment(Rest, URI); % segment +parse_segment_nz_nc(?STRING_REST($?, Rest), URI) -> + {T, URI1} = parse_query(Rest, URI), % ?query + Query = calculate_parsed_query_fragment(Rest, T), + {Rest, URI1#{query => decode_query(Query)}}; +parse_segment_nz_nc(?STRING_REST($#, Rest), URI) -> + {T, URI1} = parse_fragment(Rest, URI), + Fragment = calculate_parsed_query_fragment(Rest, T), + {Rest, URI1#{fragment => decode_fragment(Fragment)}}; +parse_segment_nz_nc(?STRING_REST(Char, Rest), URI) -> + case is_segment_nz_nc(Char) of + true -> parse_segment_nz_nc(Rest, URI); + false -> throw({error,invalid_uri,[Char]}) + end; +parse_segment_nz_nc(?STRING_EMPTY, URI) -> + {?STRING_EMPTY, URI}. + + +%% Check if char is pchar. +-spec is_pchar(char()) -> boolean(). +is_pchar($%) -> true; % pct-encoded +is_pchar($:) -> true; +is_pchar($@) -> true; +is_pchar(Char) -> is_unreserved(Char) orelse is_sub_delim(Char). + +%% Check if char is segment_nz_nc. +-spec is_segment_nz_nc(char()) -> boolean(). +is_segment_nz_nc($%) -> true; % pct-encoded +is_segment_nz_nc($@) -> true; +is_segment_nz_nc(Char) -> is_unreserved(Char) orelse is_sub_delim(Char). + + +%%------------------------------------------------------------------------- +%% [RFC 3986, Chapter 3.1. Scheme] +%% +%% Each URI begins with a scheme name that refers to a specification for +%% assigning identifiers within that scheme. +%% +%% scheme = ALPHA *( ALPHA / DIGIT / "+" / "-" / "." ) +%%------------------------------------------------------------------------- +-spec parse_scheme_start(binary(), uri_map()) -> uri_map(). +parse_scheme_start(?STRING_REST(Char, Rest), URI) -> + case is_alpha(Char) of + true -> {T, URI1} = parse_scheme(Rest, URI), + Scheme = calculate_parsed_scheme(Rest, T), + URI2 = maybe_add_path(URI1), + URI2#{scheme => ?STRING_REST(Char, Scheme)}; + false -> throw({error,invalid_uri,[Char]}) + end. + +%% Add path component if it missing after parsing the URI. +%% According to the URI specification there is always a +%% path component in every URI-reference and it can be +%% empty. +maybe_add_path(Map) -> + case maps:is_key(path, Map) of + false -> + Map#{path => <<>>}; + _Else -> + Map + end. + + + +-spec parse_scheme(binary(), uri_map()) -> {binary(), uri_map()}. +parse_scheme(?STRING_REST($:, Rest), URI) -> + {_, URI1} = parse_hier(Rest, URI), + {Rest, URI1}; +parse_scheme(?STRING_REST(Char, Rest), URI) -> + case is_scheme(Char) of + true -> parse_scheme(Rest, URI); + false -> throw({error,invalid_uri,[Char]}) + end; +parse_scheme(?STRING_EMPTY, _URI) -> + throw({error,invalid_uri,<<>>}). + + +%% Check if char is allowed in scheme +-spec is_scheme(char()) -> boolean(). +is_scheme($+) -> true; +is_scheme($-) -> true; +is_scheme($.) -> true; +is_scheme(Char) -> is_alpha(Char) orelse is_digit(Char). + + +%%------------------------------------------------------------------------- +%% hier-part = "//" authority path-abempty +%% / path-absolute +%% / path-rootless +%% / path-empty +%%------------------------------------------------------------------------- +-spec parse_hier(binary(), uri_map()) -> {binary(), uri_map()}. +parse_hier(?STRING_REST("//", Rest), URI) -> + % Parse userinfo - "//" is NOT part of authority + try parse_userinfo(Rest, URI) of + {T, URI1} -> + Userinfo = calculate_parsed_userinfo(Rest, T), + {Rest, URI1#{userinfo => decode_userinfo(Userinfo)}} + catch + throw:{_,_,_} -> + {T, URI1} = parse_host(Rest, URI), + Host = calculate_parsed_host_port(Rest, T), + {Rest, URI1#{host => decode_host(remove_brackets(Host))}} + end; +parse_hier(?STRING_REST($/, Rest), URI) -> + {T, URI1} = parse_segment(Rest, URI), % path-absolute + Path = calculate_parsed_part(Rest, T), + {Rest, URI1#{path => decode_path(?STRING_REST($/, Path))}}; +parse_hier(?STRING_REST($?, Rest), URI) -> + {T, URI1} = parse_query(Rest, URI), % path-empty ?query + Query = calculate_parsed_query_fragment(Rest, T), + {Rest, URI1#{query => decode_query(Query)}}; +parse_hier(?STRING_REST($#, Rest), URI) -> + {T, URI1} = parse_fragment(Rest, URI), % path-empty + Fragment = calculate_parsed_query_fragment(Rest, T), + {Rest, URI1#{fragment => decode_fragment(Fragment)}}; +parse_hier(?STRING_REST(Char, Rest), URI) -> % path-rootless + case is_pchar(Char) of + true -> % segment_nz + {T, URI1} = parse_segment(Rest, URI), + Path = calculate_parsed_part(Rest, T), + {Rest, URI1#{path => decode_path(?STRING_REST(Char, Path))}}; + false -> throw({error,invalid_uri,[Char]}) + end; +parse_hier(?STRING_EMPTY, URI) -> + {<<>>, URI}. + + +%%------------------------------------------------------------------------- +%% [RFC 3986, Chapter 3.2. Authority] +%% +%% Many URI schemes include a hierarchical element for a naming +%% authority so that governance of the name space defined by the +%% remainder of the URI is delegated to that authority (which may, in +%% turn, delegate it further). +%% +%% The authority component is preceded by a double slash ("//") and is +%% terminated by the next slash ("/"), question mark ("?"), or number +%% sign ("#") character, or by the end of the URI. +%% +%% authority = [ userinfo "@" ] host [ ":" port ] +%% +%% +%% [RFC 3986, Chapter 3.2.1. User Information] +%% +%% The userinfo subcomponent may consist of a user name and, optionally, +%% scheme-specific information about how to gain authorization to access +%% the resource. The user information, if present, is followed by a +%% commercial at-sign ("@") that delimits it from the host. +%% +%% userinfo = *( unreserved / pct-encoded / sub-delims / ":" ) +%%------------------------------------------------------------------------- +-spec parse_userinfo(binary(), uri_map()) -> {binary(), uri_map()}. +parse_userinfo(?CHAR($@), URI) -> + {?STRING_EMPTY, URI#{host => <<>>}}; +parse_userinfo(?STRING_REST($@, Rest), URI) -> + {T, URI1} = parse_host(Rest, URI), + Host = calculate_parsed_host_port(Rest, T), + {Rest, URI1#{host => decode_host(remove_brackets(Host))}}; +parse_userinfo(?STRING_REST(Char, Rest), URI) -> + case is_userinfo(Char) of + true -> parse_userinfo(Rest, URI); + false -> throw({error,invalid_uri,[Char]}) + end; +parse_userinfo(?STRING_EMPTY, _URI) -> + %% URI cannot end in userinfo state + throw({error,invalid_uri,<<>>}). + + +%% Check if char is allowed in userinfo +-spec is_userinfo(char()) -> boolean(). +is_userinfo($%) -> true; % pct-encoded +is_userinfo($:) -> true; +is_userinfo(Char) -> is_unreserved(Char) orelse is_sub_delim(Char). + + +%%------------------------------------------------------------------------- +%% [RFC 3986, Chapter 3.2.2. Host] +%% +%% The host subcomponent of authority is identified by an IP literal +%% encapsulated within square brackets, an IPv4 address in dotted- +%% decimal form, or a registered name. +%% +%% host = IP-literal / IPv4address / reg-name +%% +%% IP-literal = "[" ( IPv6address / IPvFuture ) "]" +%% +%% IPvFuture = "v" 1*HEXDIG "." 1*( unreserved / sub-delims / ":" ) +%% +%% IPv6address = 6( h16 ":" ) ls32 +%% / "::" 5( h16 ":" ) ls32 +%% / [ h16 ] "::" 4( h16 ":" ) ls32 +%% / [ *1( h16 ":" ) h16 ] "::" 3( h16 ":" ) ls32 +%% / [ *2( h16 ":" ) h16 ] "::" 2( h16 ":" ) ls32 +%% / [ *3( h16 ":" ) h16 ] "::" h16 ":" ls32 +%% / [ *4( h16 ":" ) h16 ] "::" ls32 +%% / [ *5( h16 ":" ) h16 ] "::" h16 +%% / [ *6( h16 ":" ) h16 ] "::" +%% +%% ls32 = ( h16 ":" h16 ) / IPv4address +%% ; least-significant 32 bits of address +%% +%% h16 = 1*4HEXDIG +%% ; 16 bits of address represented in hexadecimal +%% +%% IPv4address = dec-octet "." dec-octet "." dec-octet "." dec-octet +%% +%% dec-octet = DIGIT ; 0-9 +%% / %x31-39 DIGIT ; 10-99 +%% / "1" 2DIGIT ; 100-199 +%% / "2" %x30-34 DIGIT ; 200-249 +%% / "25" %x30-35 ; 250-255 +%% +%% reg-name = *( unreserved / pct-encoded / sub-delims ) +%%------------------------------------------------------------------------- +-spec parse_host(binary(), uri_map()) -> {binary(), uri_map()}. +parse_host(?STRING_REST($:, Rest), URI) -> + {T, URI1} = parse_port(Rest, URI), + H = calculate_parsed_host_port(Rest, T), + Port = get_port(H), + {Rest, URI1#{port => Port}}; +parse_host(?STRING_REST($/, Rest), URI) -> + {T, URI1} = parse_segment(Rest, URI), % path-abempty + Path = calculate_parsed_part(Rest, T), + {Rest, URI1#{path => decode_path(?STRING_REST($/, Path))}}; +parse_host(?STRING_REST($?, Rest), URI) -> + {T, URI1} = parse_query(Rest, URI), % path-empty ?query + Query = calculate_parsed_query_fragment(Rest, T), + {Rest, URI1#{query => decode_query(Query)}}; +parse_host(?STRING_REST($[, Rest), URI) -> + parse_ipv6_bin(Rest, [], URI); +parse_host(?STRING_REST($#, Rest), URI) -> + {T, URI1} = parse_fragment(Rest, URI), % path-empty + Fragment = calculate_parsed_query_fragment(Rest, T), + {Rest, URI1#{fragment => decode_fragment(Fragment)}}; +parse_host(?STRING_REST(Char, Rest), URI) -> + case is_digit(Char) of + true -> parse_ipv4_bin(Rest, [Char], URI); + false -> parse_reg_name(?STRING_REST(Char, Rest), URI) + end; +parse_host(?STRING_EMPTY, URI) -> + {?STRING_EMPTY, URI}. + + +-spec parse_reg_name(binary(), uri_map()) -> {binary(), uri_map()}. +parse_reg_name(?STRING_REST($:, Rest), URI) -> + {T, URI1} = parse_port(Rest, URI), + H = calculate_parsed_host_port(Rest, T), + Port = get_port(H), + {Rest, URI1#{port => Port}}; +parse_reg_name(?STRING_REST($/, Rest), URI) -> + {T, URI1} = parse_segment(Rest, URI), % path-abempty + Path = calculate_parsed_part(Rest, T), + {Rest, URI1#{path => decode_path(?STRING_REST($/, Path))}}; +parse_reg_name(?STRING_REST($?, Rest), URI) -> + {T, URI1} = parse_query(Rest, URI), % path-empty ?query + Query = calculate_parsed_query_fragment(Rest, T), + {Rest, URI1#{query => decode_query(Query)}}; +parse_reg_name(?STRING_REST($#, Rest), URI) -> + {T, URI1} = parse_fragment(Rest, URI), % path-empty + Fragment = calculate_parsed_query_fragment(Rest, T), + {Rest, URI1#{fragment => decode_fragment(Fragment)}}; +parse_reg_name(?STRING_REST(Char, Rest), URI) -> + case is_reg_name(Char) of + true -> parse_reg_name(Rest, URI); + false -> throw({error,invalid_uri,[Char]}) + end; +parse_reg_name(?STRING_EMPTY, URI) -> + {?STRING_EMPTY, URI}. + +%% Check if char is allowed in reg-name +-spec is_reg_name(char()) -> boolean(). +is_reg_name($%) -> true; +is_reg_name(Char) -> is_unreserved(Char) orelse is_sub_delim(Char). + + +-spec parse_ipv4_bin(binary(), list(), uri_map()) -> {binary(), uri_map()}. +parse_ipv4_bin(?STRING_REST($:, Rest), Acc, URI) -> + _ = validate_ipv4_address(lists:reverse(Acc)), + {T, URI1} = parse_port(Rest, URI), + H = calculate_parsed_host_port(Rest, T), + Port = get_port(H), + {Rest, URI1#{port => Port}}; +parse_ipv4_bin(?STRING_REST($/, Rest), Acc, URI) -> + _ = validate_ipv4_address(lists:reverse(Acc)), + {T, URI1} = parse_segment(Rest, URI), % path-abempty + Path = calculate_parsed_part(Rest, T), + {Rest, URI1#{path => decode_path(?STRING_REST($/, Path))}}; +parse_ipv4_bin(?STRING_REST($?, Rest), Acc, URI) -> + _ = validate_ipv4_address(lists:reverse(Acc)), + {T, URI1} = parse_query(Rest, URI), % path-empty ?query + Query = calculate_parsed_query_fragment(Rest, T), + {Rest, URI1#{query => decode_query(Query)}}; +parse_ipv4_bin(?STRING_REST($#, Rest), Acc, URI) -> + _ = validate_ipv4_address(lists:reverse(Acc)), + {T, URI1} = parse_fragment(Rest, URI), % path-empty + Fragment = calculate_parsed_query_fragment(Rest, T), + {Rest, URI1#{fragment => decode_fragment(Fragment)}}; +parse_ipv4_bin(?STRING_REST(Char, Rest), Acc, URI) -> + case is_ipv4(Char) of + true -> parse_ipv4_bin(Rest, [Char|Acc], URI); + false -> throw({error,invalid_uri,[Char]}) + end; +parse_ipv4_bin(?STRING_EMPTY, Acc, URI) -> + _ = validate_ipv4_address(lists:reverse(Acc)), + {?STRING_EMPTY, URI}. + + +%% Check if char is allowed in IPv4 addresses +-spec is_ipv4(char()) -> boolean(). +is_ipv4($.) -> true; +is_ipv4(Char) -> is_digit(Char). + +-spec validate_ipv4_address(list()) -> list(). +validate_ipv4_address(Addr) -> + case inet:parse_ipv4strict_address(Addr) of + {ok, _} -> Addr; + {error, _} -> throw({error,invalid_uri,Addr}) + end. + + +-spec parse_ipv6_bin(binary(), list(), uri_map()) -> {binary(), uri_map()}. +parse_ipv6_bin(?STRING_REST($], Rest), Acc, URI) -> + _ = validate_ipv6_address(lists:reverse(Acc)), + parse_ipv6_bin_end(Rest, URI); +parse_ipv6_bin(?STRING_REST(Char, Rest), Acc, URI) -> + case is_ipv6(Char) of + true -> parse_ipv6_bin(Rest, [Char|Acc], URI); + false -> throw({error,invalid_uri,[Char]}) + end; +parse_ipv6_bin(?STRING_EMPTY, _Acc, _URI) -> + throw({error,invalid_uri,<<>>}). + +%% Check if char is allowed in IPv6 addresses +-spec is_ipv6(char()) -> boolean(). +is_ipv6($:) -> true; +is_ipv6($.) -> true; +is_ipv6(Char) -> is_hex_digit(Char). + + +-spec parse_ipv6_bin_end(binary(), uri_map()) -> {binary(), uri_map()}. +parse_ipv6_bin_end(?STRING_REST($:, Rest), URI) -> + {T, URI1} = parse_port(Rest, URI), + H = calculate_parsed_host_port(Rest, T), + Port = get_port(H), + {Rest, URI1#{port => Port}}; +parse_ipv6_bin_end(?STRING_REST($/, Rest), URI) -> + {T, URI1} = parse_segment(Rest, URI), % path-abempty + Path = calculate_parsed_part(Rest, T), + {Rest, URI1#{path => decode_path(?STRING_REST($/, Path))}}; +parse_ipv6_bin_end(?STRING_REST($?, Rest), URI) -> + {T, URI1} = parse_query(Rest, URI), % path-empty ?query + Query = calculate_parsed_query_fragment(Rest, T), + {Rest, URI1#{query => decode_query(Query)}}; +parse_ipv6_bin_end(?STRING_REST($#, Rest), URI) -> + {T, URI1} = parse_fragment(Rest, URI), % path-empty + Fragment = calculate_parsed_query_fragment(Rest, T), + {Rest, URI1#{fragment => decode_fragment(Fragment)}}; +parse_ipv6_bin_end(?STRING_REST(Char, Rest), URI) -> + case is_ipv6(Char) of + true -> parse_ipv6_bin_end(Rest, URI); + false -> throw({error,invalid_uri,[Char]}) + end; +parse_ipv6_bin_end(?STRING_EMPTY, URI) -> + {?STRING_EMPTY, URI}. + +-spec validate_ipv6_address(list()) -> list(). +validate_ipv6_address(Addr) -> + case inet:parse_ipv6strict_address(Addr) of + {ok, _} -> Addr; + {error, _} -> throw({error,invalid_uri,Addr}) + end. + + +%%------------------------------------------------------------------------- +%% [RFC 3986, Chapter 3.2.2. Port] +%% +%% The port subcomponent of authority is designated by an optional port +%% number in decimal following the host and delimited from it by a +%% single colon (":") character. +%% +%% port = *DIGIT +%%------------------------------------------------------------------------- +-spec parse_port(binary(), uri_map()) -> {binary(), uri_map()}. +parse_port(?STRING_REST($/, Rest), URI) -> + {T, URI1} = parse_segment(Rest, URI), % path-abempty + Path = calculate_parsed_part(Rest, T), + {Rest, URI1#{path => decode_path(?STRING_REST($/, Path))}}; +parse_port(?STRING_REST($?, Rest), URI) -> + {T, URI1} = parse_query(Rest, URI), % path-empty ?query + Query = calculate_parsed_query_fragment(Rest, T), + {Rest, URI1#{query => decode_query(Query)}}; +parse_port(?STRING_REST($#, Rest), URI) -> + {T, URI1} = parse_fragment(Rest, URI), % path-empty + Fragment = calculate_parsed_query_fragment(Rest, T), + {Rest, URI1#{fragment => decode_fragment(Fragment)}}; +parse_port(?STRING_REST(Char, Rest), URI) -> + case is_digit(Char) of + true -> parse_port(Rest, URI); + false -> throw({error,invalid_uri,[Char]}) + end; +parse_port(?STRING_EMPTY, URI) -> + {?STRING_EMPTY, URI}. + + +%%------------------------------------------------------------------------- +%% [RFC 3986, Chapter 3.4. Query] +%% +%% The query component contains non-hierarchical data that, along with +%% data in the path component (Section 3.3), serves to identify a +%% resource within the scope of the URI's scheme and naming authority +%% (if any). The query component is indicated by the first question +%% mark ("?") character and terminated by a number sign ("#") character +%% or by the end of the URI. +%% +%% query = *( pchar / "/" / "?" ) +%%------------------------------------------------------------------------- +-spec parse_query(binary(), uri_map()) -> {binary(), uri_map()}. +parse_query(?STRING_REST($#, Rest), URI) -> + {T, URI1} = parse_fragment(Rest, URI), + Fragment = calculate_parsed_query_fragment(Rest, T), + {Rest, URI1#{fragment => decode_fragment(Fragment)}}; +parse_query(?STRING_REST(Char, Rest), URI) -> + case is_query(Char) of + true -> parse_query(Rest, URI); + false -> throw({error,invalid_uri,[Char]}) + end; +parse_query(?STRING_EMPTY, URI) -> + {?STRING_EMPTY, URI}. + + +%% Check if char is allowed in query +-spec is_query(char()) -> boolean(). +is_query($/) -> true; +is_query($?) -> true; +is_query(Char) -> is_pchar(Char). + + +%%------------------------------------------------------------------------- +%% [RFC 3986, Chapter 3.5. Fragment] +%% +%% The fragment identifier component of a URI allows indirect +%% identification of a secondary resource by reference to a primary +%% resource and additional identifying information. +%% +%% fragment = *( pchar / "/" / "?" ) +%%------------------------------------------------------------------------- +-spec parse_fragment(binary(), uri_map()) -> {binary(), uri_map()}. +parse_fragment(?STRING_REST(Char, Rest), URI) -> + case is_fragment(Char) of + true -> parse_fragment(Rest, URI); + false -> throw({error,invalid_uri,[Char]}) + end; +parse_fragment(?STRING_EMPTY, URI) -> + {?STRING_EMPTY, URI}. + + +%% Check if char is allowed in fragment +-spec is_fragment(char()) -> boolean(). +is_fragment($/) -> true; +is_fragment($?) -> true; +is_fragment(Char) -> is_pchar(Char). + + +%%------------------------------------------------------------------------- +%% [RFC 3986, Chapter 2.2. Reserved Characters] +%% +%% reserved = gen-delims / sub-delims +%% +%% gen-delims = ":" / "/" / "?" / "#" / "[" / "]" / "@" +%% +%% sub-delims = "!" / "$" / "&" / "'" / "(" / ")" +%% / "*" / "+" / "," / ";" / "=" +%% +%%------------------------------------------------------------------------- + +%% Check if char is sub-delim. +-spec is_sub_delim(char()) -> boolean(). +is_sub_delim($!) -> true; +is_sub_delim($$) -> true; +is_sub_delim($&) -> true; +is_sub_delim($') -> true; +is_sub_delim($() -> true; +is_sub_delim($)) -> true; + +is_sub_delim($*) -> true; +is_sub_delim($+) -> true; +is_sub_delim($,) -> true; +is_sub_delim($;) -> true; +is_sub_delim($=) -> true; +is_sub_delim(_) -> false. + + +%%------------------------------------------------------------------------- +%% [RFC 3986, Chapter 2.3. Unreserved Characters] +%% +%% unreserved = ALPHA / DIGIT / "-" / "." / "_" / "~" +%% +%%------------------------------------------------------------------------- +-spec is_unreserved(char()) -> boolean(). +is_unreserved($-) -> true; +is_unreserved($.) -> true; +is_unreserved($_) -> true; +is_unreserved($~) -> true; +is_unreserved(Char) -> is_alpha(Char) orelse is_digit(Char). + +-spec is_alpha(char()) -> boolean(). +is_alpha(C) + when $A =< C, C =< $Z; + $a =< C, C =< $z -> true; +is_alpha(_) -> false. + +-spec is_digit(char()) -> boolean(). +is_digit(C) + when $0 =< C, C =< $9 -> true; +is_digit(_) -> false. + +-spec is_hex_digit(char()) -> boolean(). +is_hex_digit(C) + when $0 =< C, C =< $9;$a =< C, C =< $f;$A =< C, C =< $F -> true; +is_hex_digit(_) -> false. + + +%% Remove enclosing brackets from binary +-spec remove_brackets(binary()) -> binary(). +remove_brackets(<<$[/utf8, Rest/binary>>) -> + {H,T} = split_binary(Rest, byte_size(Rest) - 1), + case T =:= <<$]/utf8>> of + true -> H; + false -> Rest + end; +remove_brackets(Addr) -> Addr. + + +%%------------------------------------------------------------------------- +%% Helper functions for calculating the parsed binary. +%%------------------------------------------------------------------------- +-spec calculate_parsed_scheme(binary(), binary()) -> binary(). +calculate_parsed_scheme(Input, <<>>) -> + strip_last_char(Input, [$:]); +calculate_parsed_scheme(Input, Unparsed) -> + get_parsed_binary(Input, Unparsed). + + +-spec calculate_parsed_part(binary(), binary()) -> binary(). +calculate_parsed_part(Input, <<>>) -> + strip_last_char(Input, [$?,$#]); +calculate_parsed_part(Input, Unparsed) -> + get_parsed_binary(Input, Unparsed). + + +-spec calculate_parsed_userinfo(binary(), binary()) -> binary(). +calculate_parsed_userinfo(Input, <<>>) -> + strip_last_char(Input, [$?,$#,$@]); +calculate_parsed_userinfo(Input, Unparsed) -> + get_parsed_binary(Input, Unparsed). + + +-spec calculate_parsed_host_port(binary(), binary()) -> binary(). +calculate_parsed_host_port(Input, <<>>) -> + strip_last_char(Input, [$:,$?,$#,$/]); +calculate_parsed_host_port(Input, Unparsed) -> + get_parsed_binary(Input, Unparsed). + + +calculate_parsed_query_fragment(Input, <<>>) -> + strip_last_char(Input, [$#]); +calculate_parsed_query_fragment(Input, Unparsed) -> + get_parsed_binary(Input, Unparsed). + + +get_port(<<>>) -> + undefined; +get_port(B) -> + try binary_to_integer(B) + catch + error:badarg -> + throw({error, invalid_uri, B}) + end. + + +%% Strip last char if it is in list +%% +%% This function is optimized for speed: parse/1 is about 10% faster than +%% with an alternative implementation based on lists and sets. +strip_last_char(<<>>, _) -> <<>>; +strip_last_char(Input, [C0]) -> + case binary:last(Input) of + C0 -> + init_binary(Input); + _Else -> + Input + end; +strip_last_char(Input, [C0,C1]) -> + case binary:last(Input) of + C0 -> + init_binary(Input); + C1 -> + init_binary(Input); + _Else -> + Input + end; +strip_last_char(Input, [C0,C1,C2]) -> + case binary:last(Input) of + C0 -> + init_binary(Input); + C1 -> + init_binary(Input); + C2 -> + init_binary(Input); + _Else -> + Input + end; +strip_last_char(Input, [C0,C1,C2,C3]) -> + case binary:last(Input) of + C0 -> + init_binary(Input); + C1 -> + init_binary(Input); + C2 -> + init_binary(Input); + C3 -> + init_binary(Input); + _Else -> + Input + end. + + +%% Get parsed binary +get_parsed_binary(Input, Unparsed) -> + {First, _} = split_binary(Input, byte_size(Input) - byte_size_exl_head(Unparsed)), + First. + + +%% Return all bytes of the binary except the last one. The binary must be non-empty. +init_binary(B) -> + {Init, _} = + split_binary(B, byte_size(B) - 1), + Init. + + +%% Returns the size of a binary exluding the first element. +%% Used in calls to split_binary(). +-spec byte_size_exl_head(binary()) -> number(). +byte_size_exl_head(<<>>) -> 0; +byte_size_exl_head(Binary) -> byte_size(Binary) + 1. + + +%%------------------------------------------------------------------------- +%% [RFC 3986, Chapter 2.1. Percent-Encoding] +%% +%% A percent-encoding mechanism is used to represent a data octet in a +%% component when that octet's corresponding character is outside the +%% allowed set or is being used as a delimiter of, or within, the +%% component. A percent-encoded octet is encoded as a character +%% triplet, consisting of the percent character "%" followed by the two +%% hexadecimal digits representing that octet's numeric value. For +%% example, "%20" is the percent-encoding for the binary octet +%% "00100000" (ABNF: %x20), which in US-ASCII corresponds to the space +%% character (SP). Section 2.4 describes when percent-encoding and +%% decoding is applied. +%% +%% pct-encoded = "%" HEXDIG HEXDIG +%%------------------------------------------------------------------------- +-spec decode_userinfo(binary()) -> binary(). +decode_userinfo(Cs) -> + check_utf8(decode(Cs, fun is_userinfo/1, <<>>)). + +-spec decode_host(binary()) -> binary(). +decode_host(Cs) -> + check_utf8(decode(Cs, fun is_host/1, <<>>)). + +-spec decode_path(binary()) -> binary(). +decode_path(Cs) -> + check_utf8(decode(Cs, fun is_path/1, <<>>)). + +-spec decode_query(binary()) -> binary(). +decode_query(Cs) -> + check_utf8(decode(Cs, fun is_query/1, <<>>)). + +-spec decode_fragment(binary()) -> binary(). +decode_fragment(Cs) -> + check_utf8(decode(Cs, fun is_fragment/1, <<>>)). + + +%% Returns Cs if it is utf8 encoded. +check_utf8(Cs) -> + case unicode:characters_to_list(Cs) of + {incomplete,_,_} -> + throw({error,invalid_utf8,Cs}); + {error,_,_} -> + throw({error,invalid_utf8,Cs}); + _ -> Cs + end. + +%%------------------------------------------------------------------------- +%% Percent-encode +%%------------------------------------------------------------------------- + +%% Only validates as scheme cannot have percent-encoded characters +-spec encode_scheme(list()|binary()) -> list() | binary(). +encode_scheme([]) -> + throw({error,invalid_scheme,""}); +encode_scheme(<<>>) -> + throw({error,invalid_scheme,<<>>}); +encode_scheme(Scheme) -> + case validate_scheme(Scheme) of + true -> Scheme; + false -> throw({error,invalid_scheme,Scheme}) + end. + +-spec encode_userinfo(list()|binary()) -> list() | binary(). +encode_userinfo(Cs) -> + encode(Cs, fun is_userinfo/1). + +-spec encode_host(list()|binary()) -> list() | binary(). +encode_host(Cs) -> + case classify_host(Cs) of + regname -> Cs; + ipv4 -> Cs; + ipv6 -> bracket_ipv6(Cs); + other -> encode(Cs, fun is_reg_name/1) + end. + +-spec encode_path(list()|binary()) -> list() | binary(). +encode_path(Cs) -> + encode(Cs, fun is_path/1). + +-spec encode_query(list()|binary()) -> list() | binary(). +encode_query(Cs) -> + encode(Cs, fun is_query/1). + +-spec encode_fragment(list()|binary()) -> list() | binary(). +encode_fragment(Cs) -> + encode(Cs, fun is_fragment/1). + +%%------------------------------------------------------------------------- +%% Helper funtions for percent-decode +%%------------------------------------------------------------------------- +decode(<<$%,C0,C1,Cs/binary>>, Fun, Acc) -> + case is_hex_digit(C0) andalso is_hex_digit(C1) of + true -> + B = ?HEX2DEC(C0)*16+?HEX2DEC(C1), + decode(Cs, Fun, <<Acc/binary, B>>); + false -> throw({error,invalid_percent_encoding,<<$%,C0,C1>>}) + end; +decode(<<C,Cs/binary>>, Fun, Acc) -> + case Fun(C) of + true -> decode(Cs, Fun, <<Acc/binary, C>>); + false -> throw({error,invalid_percent_encoding,<<C,Cs/binary>>}) + end; +decode(<<>>, _Fun, Acc) -> + Acc. + +%% Check if char is allowed in host +-spec is_host(char()) -> boolean(). +is_host($:) -> true; +is_host(Char) -> is_unreserved(Char) orelse is_sub_delim(Char). + +%% Check if char is allowed in path +-spec is_path(char()) -> boolean(). +is_path($/) -> true; +is_path(Char) -> is_pchar(Char). + + +%%------------------------------------------------------------------------- +%% Helper functions for percent-encode +%%------------------------------------------------------------------------- +-spec encode(list()|binary(), fun()) -> list() | binary(). +encode(Component, Fun) when is_list(Component) -> + B = unicode:characters_to_binary(Component), + unicode:characters_to_list(encode(B, Fun, <<>>)); +encode(Component, Fun) when is_binary(Component) -> + encode(Component, Fun, <<>>). +%% +encode(<<Char/utf8, Rest/binary>>, Fun, Acc) -> + C = encode_codepoint_binary(Char, Fun), + encode(Rest, Fun, <<Acc/binary,C/binary>>); +encode(<<Char, Rest/binary>>, _Fun, _Acc) -> + throw({error,invalid_input,<<Char,Rest/binary>>}); +encode(<<>>, _Fun, Acc) -> + Acc. + + +-spec encode_codepoint_binary(integer(), fun()) -> binary(). +encode_codepoint_binary(C, Fun) -> + case Fun(C) of + false -> percent_encode_binary(C); + true -> <<C>> + end. + + +-spec percent_encode_binary(integer()) -> binary(). +percent_encode_binary(Code) -> + percent_encode_binary(<<Code/utf8>>, <<>>). + + +percent_encode_binary(<<A:4,B:4,Rest/binary>>, Acc) -> + percent_encode_binary(Rest, <<Acc/binary,$%,(?DEC2HEX(A)),(?DEC2HEX(B))>>); +percent_encode_binary(<<>>, Acc) -> + Acc. + + +%%------------------------------------------------------------------------- +%%------------------------------------------------------------------------- +validate_scheme([]) -> true; +validate_scheme([H|T]) -> + case is_scheme(H) of + true -> validate_scheme(T); + false -> false + end; +validate_scheme(<<>>) -> true; +validate_scheme(<<H, Rest/binary>>) -> + case is_scheme(H) of + true -> validate_scheme(Rest); + false -> false + end. + + +%%------------------------------------------------------------------------- +%% Classifies hostname into the following categories: +%% regname, ipv4 - address does not contain reserved characters to be +%% percent-encoded +%% ipv6 - address does not contain reserved characters but it shall be +%% encolsed in brackets +%% other - address shall be percent-encoded +%%------------------------------------------------------------------------- +classify_host([]) -> other; +classify_host(Addr) when is_binary(Addr) -> + A = unicode:characters_to_list(Addr), + classify_host_ipv6(A); +classify_host(Addr) -> + classify_host_ipv6(Addr). + +classify_host_ipv6(Addr) -> + case is_ipv6_address(Addr) of + true -> ipv6; + false -> classify_host_ipv4(Addr) + end. + +classify_host_ipv4(Addr) -> + case is_ipv4_address(Addr) of + true -> ipv4; + false -> classify_host_regname(Addr) + end. + +classify_host_regname([]) -> regname; +classify_host_regname([H|T]) -> + case is_reg_name(H) of + true -> classify_host_regname(T); + false -> other + end. + +is_ipv4_address(Addr) -> + case inet:parse_ipv4strict_address(Addr) of + {ok, _} -> true; + {error, _} -> false + end. + +is_ipv6_address(Addr) -> + case inet:parse_ipv6strict_address(Addr) of + {ok, _} -> true; + {error, _} -> false + end. + +bracket_ipv6(Addr) when is_binary(Addr) -> + concat(<<$[,Addr/binary>>,<<$]>>); +bracket_ipv6(Addr) when is_list(Addr) -> + [$[|Addr] ++ "]". + + +%%------------------------------------------------------------------------- +%% Helper funtions for recompose +%%------------------------------------------------------------------------- + +%%------------------------------------------------------------------------- +%% Checks if input Map has valid combination of fields that can be +%% recomposed into a URI. +%% +%% The implementation is based on a decision tree that fulfills the +%% following rules: +%% - 'path' shall always be present in the input map +%% URI = scheme ":" hier-part [ "?" query ] [ "#" fragment ] +%% hier-part = "//" authority path-abempty +%% / path-absolute +%% / path-rootless +%% / path-empty +%% - 'host' shall be present in the input map when 'path' starts with +%% two slashes ("//") +%% path = path-abempty ; begins with "/" or is empty +%% / path-absolute ; begins with "/" but not "//" +%% / path-noscheme ; begins with a non-colon segment +%% / path-rootless ; begins with a segment +%% / path-empty ; zero characters +%% path-abempty = *( "/" segment ) +%% segment = *pchar +%% - 'host' shall be present if userinfo or port is present in input map +%% authority = [ userinfo "@" ] host [ ":" port ] +%% - All fields shall be valid (scheme, userinfo, host, port, path, query +%% or fragment). +%%------------------------------------------------------------------------- +is_valid_map(#{path := Path} = Map) -> + ((starts_with_two_slash(Path) andalso is_valid_map_host(Map)) + orelse + (maps:is_key(userinfo, Map) andalso is_valid_map_host(Map)) + orelse + (maps:is_key(port, Map) andalso is_valid_map_host(Map)) + orelse + all_fields_valid(Map)); +is_valid_map(#{}) -> + false. + + +is_valid_map_host(Map) -> + maps:is_key(host, Map) andalso all_fields_valid(Map). + + +all_fields_valid(Map) -> + Fun = fun(scheme, _, Acc) -> Acc; + (userinfo, _, Acc) -> Acc; + (host, _, Acc) -> Acc; + (port, _, Acc) -> Acc; + (path, _, Acc) -> Acc; + (query, _, Acc) -> Acc; + (fragment, _, Acc) -> Acc; + (_, _, _) -> false + end, + maps:fold(Fun, true, Map). + + +starts_with_two_slash([$/,$/|_]) -> + true; +starts_with_two_slash(?STRING_REST("//", _)) -> + true; +starts_with_two_slash(_) -> false. + + +update_scheme(#{scheme := Scheme}, _) -> + add_colon_postfix(encode_scheme(Scheme)); +update_scheme(#{}, _) -> + empty. + + +update_userinfo(#{userinfo := Userinfo}, empty) -> + add_auth_prefix(encode_userinfo(Userinfo)); +update_userinfo(#{userinfo := Userinfo}, URI) -> + concat(URI,add_auth_prefix(encode_userinfo(Userinfo))); +update_userinfo(#{}, empty) -> + empty; +update_userinfo(#{}, URI) -> + URI. + + +update_host(#{host := Host}, empty) -> + add_auth_prefix(encode_host(Host)); +update_host(#{host := Host} = Map, URI) -> + concat(URI,add_host_prefix(Map, encode_host(Host))); +update_host(#{}, empty) -> + empty; +update_host(#{}, URI) -> + URI. + + +%% URI cannot be empty for ports. E.g. ":8080" is not a valid URI +update_port(#{port := undefined}, URI) -> + concat(URI, <<":">>); +update_port(#{port := Port}, URI) -> + concat(URI,add_colon(encode_port(Port))); +update_port(#{}, URI) -> + URI. + + +update_path(#{path := Path}, empty) -> + encode_path(Path); +update_path(#{path := Path}, URI) -> + concat(URI,encode_path(Path)); +update_path(#{}, empty) -> + empty; +update_path(#{}, URI) -> + URI. + + +update_query(#{query := Query}, empty) -> + encode_query(Query); +update_query(#{query := Query}, URI) -> + concat(URI,add_question_mark(encode_query(Query))); +update_query(#{}, empty) -> + empty; +update_query(#{}, URI) -> + URI. + + +update_fragment(#{fragment := Fragment}, empty) -> + add_hashmark(encode_fragment(Fragment)); +update_fragment(#{fragment := Fragment}, URI) -> + concat(URI,add_hashmark(encode_fragment(Fragment))); +update_fragment(#{}, empty) -> + ""; +update_fragment(#{}, URI) -> + URI. + +%%------------------------------------------------------------------------- +%% Concatenates its arguments that can be lists and binaries. +%% The result is a list if at least one of its argument is a list and +%% binary otherwise. +%%------------------------------------------------------------------------- +concat(A, B) when is_binary(A), is_binary(B) -> + <<A/binary, B/binary>>; +concat(A, B) when is_binary(A), is_list(B) -> + unicode:characters_to_list(A) ++ B; +concat(A, B) when is_list(A) -> + A ++ maybe_to_list(B). + +add_hashmark(Comp) when is_binary(Comp) -> + <<$#, Comp/binary>>; +add_hashmark(Comp) when is_list(Comp) -> + [$#|Comp]. + +add_question_mark(Comp) when is_binary(Comp) -> + <<$?, Comp/binary>>; +add_question_mark(Comp) when is_list(Comp) -> + [$?|Comp]. + +add_colon(Comp) when is_binary(Comp) -> + <<$:, Comp/binary>>. + +add_colon_postfix(Comp) when is_binary(Comp) -> + <<Comp/binary,$:>>; +add_colon_postfix(Comp) when is_list(Comp) -> + Comp ++ ":". + +add_auth_prefix(Comp) when is_binary(Comp) -> + <<"//", Comp/binary>>; +add_auth_prefix(Comp) when is_list(Comp) -> + [$/,$/|Comp]. + +add_host_prefix(#{userinfo := _}, Host) when is_binary(Host) -> + <<$@,Host/binary>>; +add_host_prefix(#{}, Host) when is_binary(Host) -> + <<"//",Host/binary>>; +add_host_prefix(#{userinfo := _}, Host) when is_list(Host) -> + [$@|Host]; +add_host_prefix(#{}, Host) when is_list(Host) -> + [$/,$/|Host]. + +maybe_to_list(Comp) when is_binary(Comp) -> unicode:characters_to_list(Comp); +maybe_to_list(Comp) -> Comp. + +encode_port(Port) -> + integer_to_binary(Port). + +%%------------------------------------------------------------------------- +%% Helper functions for transcode +%%------------------------------------------------------------------------- + +%%------------------------------------------------------------------------- +%% uri_string:transcode(<<"x%00%00%00%F6"/utf32>>). +%% 1. Convert (transcode/2) input to list form (list of unicode codepoints) +%% "x%00%00%00%F6" +%% 2. Accumulate characters until percent-encoded segment (transcode/4). +%% Acc = "x" +%% 3. Convert percent-encoded triplets to binary form (transcode_pct/4) +%% <<0,0,0,246>> +%% 4. Transcode in-encoded binary to out-encoding (utf32 -> utf8): +%% <<195,182>> +%% 5. Percent-encode out-encoded binary: +%% <<"%C3%B6"/utf8>> = <<37,67,51,37,66,54>> +%% 6. Convert binary to list form, reverse it and append the accumulator +%% "6B%3C%" + "x" +%% 7. Reverse Acc and return it +%%------------------------------------------------------------------------- +transcode([$%,_C0,_C1|_Rest] = L, Acc, InEnc, OutEnc) -> + transcode_pct(L, Acc, <<>>, InEnc, OutEnc); +transcode([_C|_Rest] = L, Acc, InEnc, OutEnc) -> + transcode(L, Acc, [], InEnc, OutEnc). +%% +transcode([$%,_C0,_C1|_Rest] = L, Acc, List, InEncoding, OutEncoding) -> + transcode_pct(L, List ++ Acc, <<>>, InEncoding, OutEncoding); +transcode([C|Rest], Acc, List, InEncoding, OutEncoding) -> + transcode(Rest, Acc, [C|List], InEncoding, OutEncoding); +transcode([], Acc, List, _InEncoding, _OutEncoding) -> + lists:reverse(List ++ Acc). + + +%% Transcode percent-encoded segment +transcode_pct([$%,C0,C1|Rest] = L, Acc, B, InEncoding, OutEncoding) -> + case is_hex_digit(C0) andalso is_hex_digit(C1) of + true -> + Int = ?HEX2DEC(C0)*16+?HEX2DEC(C1), + transcode_pct(Rest, Acc, <<B/binary, Int>>, InEncoding, OutEncoding); + false -> throw({error, invalid_percent_encoding,L}) + end; +transcode_pct([_C|_Rest] = L, Acc, B, InEncoding, OutEncoding) -> + OutBinary = convert_to_binary(B, InEncoding, OutEncoding), + PctEncUtf8 = percent_encode_segment(OutBinary), + Out = lists:reverse(convert_to_list(PctEncUtf8, utf8)), + transcode(L, Out ++ Acc, [], InEncoding, OutEncoding); +transcode_pct([], Acc, B, InEncoding, OutEncoding) -> + OutBinary = convert_to_binary(B, InEncoding, OutEncoding), + PctEncUtf8 = percent_encode_segment(OutBinary), + Out = convert_to_list(PctEncUtf8, utf8), + lists:reverse(Acc) ++ Out. + + +%% Convert to binary +convert_to_binary(Binary, InEncoding, OutEncoding) -> + case unicode:characters_to_binary(Binary, InEncoding, OutEncoding) of + {error, _List, RestData} -> + throw({error, invalid_input, RestData}); + {incomplete, _List, RestData} -> + throw({error, invalid_input, RestData}); + Result -> + Result + end. + + +%% Convert to list +convert_to_list(Binary, InEncoding) -> + case unicode:characters_to_list(Binary, InEncoding) of + {error, _List, RestData} -> + throw({error, invalid_input, RestData}); + {incomplete, _List, RestData} -> + throw({error, invalid_input, RestData}); + Result -> + Result + end. + + +%% Flatten input list +flatten_list([], _) -> + []; +flatten_list(L, InEnc) -> + flatten_list(L, InEnc, []). +%% +flatten_list([H|T], InEnc, Acc) when is_binary(H) -> + L = convert_to_list(H, InEnc), + flatten_list(T, InEnc, lists:reverse(L) ++ Acc); +flatten_list([H|T], InEnc, Acc) when is_list(H) -> + flatten_list(H ++ T, InEnc, Acc); +flatten_list([H|T], InEnc, Acc) -> + flatten_list(T, InEnc, [H|Acc]); +flatten_list([], _InEnc, Acc) -> + lists:reverse(Acc); +flatten_list(Arg, _, _) -> + throw({error, invalid_input, Arg}). + + +percent_encode_segment(Segment) -> + percent_encode_binary(Segment, <<>>). + + +%%------------------------------------------------------------------------- +%% Helper functions for compose_query +%%------------------------------------------------------------------------- + +%% Returns separator to be used between key-value pairs +get_separator(L) when length(L) =:= 0 -> + <<>>; +get_separator(_L) -> + <<"&">>. + + +%% HTML5 - 4.10.22.6 URL-encoded form data - encoding +form_urlencode(Cs, [{encoding, latin1}]) when is_list(Cs) -> + B = convert_to_binary(Cs, utf8, utf8), + html5_byte_encode(base10_encode(B)); +form_urlencode(Cs, [{encoding, latin1}]) when is_binary(Cs) -> + html5_byte_encode(base10_encode(Cs)); +form_urlencode(Cs, [{encoding, Encoding}]) + when is_list(Cs), Encoding =:= utf8; Encoding =:= unicode -> + B = convert_to_binary(Cs, utf8, Encoding), + html5_byte_encode(B); +form_urlencode(Cs, [{encoding, Encoding}]) + when is_binary(Cs), Encoding =:= utf8; Encoding =:= unicode -> + html5_byte_encode(Cs); +form_urlencode(Cs, [{encoding, Encoding}]) when is_list(Cs); is_binary(Cs) -> + throw({error,invalid_encoding, Encoding}); +form_urlencode(Cs, _) -> + throw({error,invalid_input, Cs}). + + +%% For each character in the entry's name and value that cannot be expressed using +%% the selected character encoding, replace the character by a string consisting of +%% a U+0026 AMPERSAND character (&), a "#" (U+0023) character, one or more ASCII +%% digits representing the Unicode code point of the character in base ten, and +%% finally a ";" (U+003B) character. +base10_encode(Cs) -> + base10_encode(Cs, <<>>). +%% +base10_encode(<<>>, Acc) -> + Acc; +base10_encode(<<H/utf8,T/binary>>, Acc) when H > 255 -> + Base10 = convert_to_binary(integer_to_list(H,10), utf8, utf8), + base10_encode(T, <<Acc/binary,"&#",Base10/binary,$;>>); +base10_encode(<<H/utf8,T/binary>>, Acc) -> + base10_encode(T, <<Acc/binary,H>>). + + +html5_byte_encode(B) -> + html5_byte_encode(B, <<>>). +%% +html5_byte_encode(<<>>, Acc) -> + Acc; +html5_byte_encode(<<$ ,T/binary>>, Acc) -> + html5_byte_encode(T, <<Acc/binary,$+>>); +html5_byte_encode(<<H,T/binary>>, Acc) -> + case is_url_char(H) of + true -> + html5_byte_encode(T, <<Acc/binary,H>>); + false -> + <<A:4,B:4>> = <<H>>, + html5_byte_encode(T, <<Acc/binary,$%,(?DEC2HEX(A)),(?DEC2HEX(B))>>) + end; +html5_byte_encode(H, _Acc) -> + throw({error,invalid_input, H}). + + +%% Return true if input char can appear in form-urlencoded string +%% Allowed chararacters: +%% 0x2A, 0x2D, 0x2E, 0x30 to 0x39, 0x41 to 0x5A, +%% 0x5F, 0x61 to 0x7A +is_url_char(C) + when C =:= 16#2A; C =:= 16#2D; + C =:= 16#2E; C =:= 16#5F; + 16#30 =< C, C =< 16#39; + 16#41 =< C, C =< 16#5A; + 16#61 =< C, C =< 16#7A -> true; +is_url_char(_) -> false. + + +%%------------------------------------------------------------------------- +%% Helper functions for dissect_query +%%------------------------------------------------------------------------- +dissect_query_key(<<$=,T/binary>>, IsList, Acc, Key, Value) -> + dissect_query_value(T, IsList, Acc, Key, Value); +dissect_query_key(<<"&#",T/binary>>, IsList, Acc, Key, Value) -> + dissect_query_key(T, IsList, Acc, <<Key/binary,"&#">>, Value); +dissect_query_key(<<$&,_T/binary>>, _IsList, _Acc, _Key, _Value) -> + throw({error, missing_value, "&"}); +dissect_query_key(<<H,T/binary>>, IsList, Acc, Key, Value) -> + dissect_query_key(T, IsList, Acc, <<Key/binary,H>>, Value); +dissect_query_key(B, _, _, _, _) -> + throw({error, missing_value, B}). + + +dissect_query_value(<<$&,T/binary>>, IsList, Acc, Key, Value) -> + K = form_urldecode(IsList, Key), + V = form_urldecode(IsList, Value), + dissect_query_key(T, IsList, [{K,V}|Acc], <<>>, <<>>); +dissect_query_value(<<H,T/binary>>, IsList, Acc, Key, Value) -> + dissect_query_value(T, IsList, Acc, Key, <<Value/binary,H>>); +dissect_query_value(<<>>, IsList, Acc, Key, Value) -> + K = form_urldecode(IsList, Key), + V = form_urldecode(IsList, Value), + lists:reverse([{K,V}|Acc]). + + +%% Form-urldecode input based on RFC 1866 [8.2.1] +form_urldecode(true, B) -> + Result = base10_decode(form_urldecode(B, <<>>)), + convert_to_list(Result, utf8); +form_urldecode(false, B) -> + base10_decode(form_urldecode(B, <<>>)); +form_urldecode(<<>>, Acc) -> + Acc; +form_urldecode(<<$+,T/binary>>, Acc) -> + form_urldecode(T, <<Acc/binary,$ >>); +form_urldecode(<<$%,C0,C1,T/binary>>, Acc) -> + case is_hex_digit(C0) andalso is_hex_digit(C1) of + true -> + V = ?HEX2DEC(C0)*16+?HEX2DEC(C1), + form_urldecode(T, <<Acc/binary, V>>); + false -> + L = convert_to_list(<<$%,C0,C1,T/binary>>, utf8), + throw({error, invalid_percent_encoding, L}) + end; +form_urldecode(<<H/utf8,T/binary>>, Acc) -> + form_urldecode(T, <<Acc/binary,H/utf8>>); +form_urldecode(<<H,_/binary>>, _Acc) -> + throw({error, invalid_character, [H]}). + +base10_decode(Cs) -> + base10_decode(Cs, <<>>). +% +base10_decode(<<>>, Acc) -> + Acc; +base10_decode(<<"&#",T/binary>>, Acc) -> + base10_decode_unicode(T, Acc); +base10_decode(<<H/utf8,T/binary>>, Acc) -> + base10_decode(T,<<Acc/binary,H/utf8>>); +base10_decode(<<H,_/binary>>, _) -> + throw({error, invalid_input, [H]}). + + +base10_decode_unicode(B, Acc) -> + base10_decode_unicode(B, 0, Acc). +%% +base10_decode_unicode(<<H/utf8,T/binary>>, Codepoint, Acc) when $0 =< H, H =< $9 -> + Res = Codepoint * 10 + (H - $0), + base10_decode_unicode(T, Res, Acc); +base10_decode_unicode(<<$;,T/binary>>, Codepoint, Acc) -> + base10_decode(T, <<Acc/binary,Codepoint/utf8>>); +base10_decode_unicode(<<H,_/binary>>, _, _) -> + throw({error, invalid_input, [H]}). + + +%%------------------------------------------------------------------------- +%% Helper functions for normalize +%%------------------------------------------------------------------------- + +%% 6.2.2.1. Case Normalization +normalize_case(#{scheme := Scheme, host := Host} = Map) -> + Map#{scheme => to_lower(Scheme), + host => to_lower(Host)}; +normalize_case(#{host := Host} = Map) -> + Map#{host => to_lower(Host)}; +normalize_case(#{scheme := Scheme} = Map) -> + Map#{scheme => to_lower(Scheme)}; +normalize_case(#{} = Map) -> + Map. + + +to_lower(Cs) when is_list(Cs) -> + B = convert_to_binary(Cs, utf8, utf8), + convert_to_list(to_lower(B), utf8); +to_lower(Cs) when is_binary(Cs) -> + to_lower(Cs, <<>>). +%% +to_lower(<<C,Cs/binary>>, Acc) when $A =< C, C =< $Z -> + to_lower(Cs, <<Acc/binary,(C + 32)>>); +to_lower(<<C,Cs/binary>>, Acc) -> + to_lower(Cs, <<Acc/binary,C>>); +to_lower(<<>>, Acc) -> + Acc. + + +%% 6.2.2.3. Path Segment Normalization +%% 5.2.4. Remove Dot Segments +normalize_path_segment(Map) -> + Path = maps:get(path, Map, undefined), + Map#{path => remove_dot_segments(Path)}. + + +remove_dot_segments(Path) when is_binary(Path) -> + remove_dot_segments(Path, <<>>); +remove_dot_segments(Path) when is_list(Path) -> + B = convert_to_binary(Path, utf8, utf8), + B1 = remove_dot_segments(B, <<>>), + convert_to_list(B1, utf8). +%% +remove_dot_segments(<<>>, Output) -> + Output; +remove_dot_segments(<<"../",T/binary>>, Output) -> + remove_dot_segments(T, Output); +remove_dot_segments(<<"./",T/binary>>, Output) -> + remove_dot_segments(T, Output); +remove_dot_segments(<<"/./",T/binary>>, Output) -> + remove_dot_segments(<<$/,T/binary>>, Output); +remove_dot_segments(<<"/.">>, Output) -> + remove_dot_segments(<<$/>>, Output); +remove_dot_segments(<<"/../",T/binary>>, Output) -> + Out1 = remove_last_segment(Output), + remove_dot_segments(<<$/,T/binary>>, Out1); +remove_dot_segments(<<"/..">>, Output) -> + Out1 = remove_last_segment(Output), + remove_dot_segments(<<$/>>, Out1); +remove_dot_segments(<<$.>>, Output) -> + remove_dot_segments(<<>>, Output); +remove_dot_segments(<<"..">>, Output) -> + remove_dot_segments(<<>>, Output); +remove_dot_segments(Input, Output) -> + {First, Rest} = first_path_segment(Input), + remove_dot_segments(Rest, <<Output/binary,First/binary>>). + + +first_path_segment(Input) -> + F = first_path_segment(Input, <<>>), + split_binary(Input, byte_size(F)). +%% +first_path_segment(<<$/,T/binary>>, Acc) -> + first_path_segment_end(<<T/binary>>, <<Acc/binary,$/>>); +first_path_segment(<<C,T/binary>>, Acc) -> + first_path_segment_end(<<T/binary>>, <<Acc/binary,C>>). + + +first_path_segment_end(<<>>, Acc) -> + Acc; +first_path_segment_end(<<$/,_/binary>>, Acc) -> + Acc; +first_path_segment_end(<<C,T/binary>>, Acc) -> + first_path_segment_end(<<T/binary>>, <<Acc/binary,C>>). + + +remove_last_segment(<<>>) -> + <<>>; +remove_last_segment(B) -> + {Init, Last} = split_binary(B, byte_size(B) - 1), + case Last of + <<$/>> -> + Init; + _Char -> + remove_last_segment(Init) + end. + + +%% RFC 3986, 6.2.3. Scheme-Based Normalization +normalize_scheme_based(Map) -> + Scheme = maps:get(scheme, Map, undefined), + Port = maps:get(port, Map, undefined), + Path= maps:get(path, Map, undefined), + normalize_scheme_based(Map, Scheme, Port, Path). +%% +normalize_scheme_based(Map, Scheme, Port, Path) + when Scheme =:= "http"; Scheme =:= <<"http">> -> + normalize_http(Map, Port, Path); +normalize_scheme_based(Map, Scheme, Port, Path) + when Scheme =:= "https"; Scheme =:= <<"https">> -> + normalize_https(Map, Port, Path); +normalize_scheme_based(Map, Scheme, Port, _Path) + when Scheme =:= "ftp"; Scheme =:= <<"ftp">> -> + normalize_ftp(Map, Port); +normalize_scheme_based(Map, Scheme, Port, _Path) + when Scheme =:= "ssh"; Scheme =:= <<"ssh">> -> + normalize_ssh_sftp(Map, Port); +normalize_scheme_based(Map, Scheme, Port, _Path) + when Scheme =:= "sftp"; Scheme =:= <<"sftp">> -> + normalize_ssh_sftp(Map, Port); +normalize_scheme_based(Map, Scheme, Port, _Path) + when Scheme =:= "tftp"; Scheme =:= <<"tftp">> -> + normalize_tftp(Map, Port); +normalize_scheme_based(Map, _, _, _) -> + Map. + + +normalize_http(Map, Port, Path) -> + M1 = normalize_port(Map, Port, 80), + normalize_http_path(M1, Path). + + +normalize_https(Map, Port, Path) -> + M1 = normalize_port(Map, Port, 443), + normalize_http_path(M1, Path). + + +normalize_ftp(Map, Port) -> + normalize_port(Map, Port, 21). + + +normalize_ssh_sftp(Map, Port) -> + normalize_port(Map, Port, 22). + + +normalize_tftp(Map, Port) -> + normalize_port(Map, Port, 69). + + +normalize_port(Map, Port, Default) -> + case Port of + Default -> + maps:remove(port, Map); + _Else -> + Map + end. + + +normalize_http_path(Map, Path) -> + case Path of + "" -> + Map#{path => "/"}; + <<>> -> + Map#{path => <<"/">>}; + _Else -> + Map + end. diff --git a/lib/stdlib/test/Makefile b/lib/stdlib/test/Makefile index 523cb95065..8490770f3d 100644 --- a/lib/stdlib/test/Makefile +++ b/lib/stdlib/test/Makefile @@ -87,6 +87,7 @@ MODULES= \ timer_simple_SUITE \ unicode_SUITE \ unicode_util_SUITE \ + uri_string_SUITE \ win32reg_SUITE \ y2k_SUITE \ select_SUITE \ diff --git a/lib/stdlib/test/base64_SUITE.erl b/lib/stdlib/test/base64_SUITE.erl index 48b3f5f959..1fc4c3fc0e 100644 --- a/lib/stdlib/test/base64_SUITE.erl +++ b/lib/stdlib/test/base64_SUITE.erl @@ -97,10 +97,9 @@ base64_otp_5635(Config) when is_list(Config) -> <<"===">> = base64:decode(base64:encode("===")), ok. %%------------------------------------------------------------------------- -%% OTP-6279: Guard needed so that function fails in a correct -%% way for faulty input, i.e. function_clause. +%% OTP-6279: Make sure illegal characters are rejected when decoding. base64_otp_6279(Config) when is_list(Config) -> - {'EXIT',{function_clause, _}} = (catch base64:decode("dGVzda==a")), + {'EXIT',_} = (catch base64:decode("dGVzda==a")), ok. %%------------------------------------------------------------------------- %% Encode and decode big binaries. @@ -115,48 +114,61 @@ big(Config) when is_list(Config) -> %%------------------------------------------------------------------------- %% Make sure illegal characters are rejected when decoding. illegal(Config) when is_list(Config) -> - {'EXIT',{function_clause, _}} = (catch base64:decode("()")), + %% A few samples with different error reasons. Nothing can be + %% assumed about the reason for the crash. + {'EXIT',_} = (catch base64:decode("()")), + {'EXIT',_} = (catch base64:decode(<<19:8,20:8,21:8,22:8>>)), + {'EXIT',_} = (catch base64:decode([19,20,21,22])), + {'EXIT',_} = (catch base64:decode_to_string(<<19:8,20:8,21:8,22:8>>)), + {'EXIT',_} = (catch base64:decode_to_string([19,20,21,22])), ok. %%------------------------------------------------------------------------- %% mime_decode and mime_decode_to_string have different implementations -%% so test both with the same input separately. Both functions have -%% the same implementation for binary/string arguments. +%% so test both with the same input separately. %% %% Test base64:mime_decode/1. mime_decode(Config) when is_list(Config) -> + MimeDecode = fun(In) -> + Out = base64:mime_decode(In), + Out = base64:mime_decode(binary_to_list(In)) + end, %% Test correct padding - <<"one">> = base64:mime_decode(<<"b25l">>), - <<"on">> = base64:mime_decode(<<"b24=">>), - <<"o">> = base64:mime_decode(<<"bw==">>), + <<"one">> = MimeDecode(<<"b25l">>), + <<"on">> = MimeDecode(<<"b24=">>), + <<"o">> = MimeDecode(<<"bw==">>), %% Test 1 extra padding - <<"one">> = base64:mime_decode(<<"b25l= =">>), - <<"on">> = base64:mime_decode(<<"b24== =">>), - <<"o">> = base64:mime_decode(<<"bw=== =">>), + <<"one">> = MimeDecode(<<"b25l= =">>), + <<"on">> = MimeDecode(<<"b24== =">>), + <<"o">> = MimeDecode(<<"bw=== =">>), %% Test 2 extra padding - <<"one">> = base64:mime_decode(<<"b25l===">>), - <<"on">> = base64:mime_decode(<<"b24====">>), - <<"o">> = base64:mime_decode(<<"bw=====">>), + <<"one">> = MimeDecode(<<"b25l===">>), + <<"on">> = MimeDecode(<<"b24====">>), + <<"o">> = MimeDecode(<<"bw=====">>), %% Test misc embedded padding - <<"one">> = base64:mime_decode(<<"b2=5l===">>), - <<"on">> = base64:mime_decode(<<"b=24====">>), - <<"o">> = base64:mime_decode(<<"b=w=====">>), + <<"one">> = MimeDecode(<<"b2=5l===">>), + <<"on">> = MimeDecode(<<"b=24====">>), + <<"o">> = MimeDecode(<<"b=w=====">>), %% Test misc white space and illegals with embedded padding - <<"one">> = base64:mime_decode(<<" b~2=\r\n5()l===">>), - <<"on">> = base64:mime_decode(<<"\tb =2\"¤4=¤= ==">>), - <<"o">> = base64:mime_decode(<<"\nb=w=====">>), + <<"one">> = MimeDecode(<<" b~2=\r\n5()l===">>), + <<"on">> = MimeDecode(<<"\tb =2\"¤4=¤= ==">>), + <<"o">> = MimeDecode(<<"\nb=w=====">>), %% Two pads <<"Aladdin:open sesame">> = - base64:mime_decode("QWxhZGRpbjpvc()GVuIHNlc2FtZQ=="), + MimeDecode(<<"QWxhZGRpbjpvc()GVuIHNlc2FtZQ==">>), %% One pad to ignore, followed by more text - <<"Hello World!!">> = base64:mime_decode(<<"SGVsb)(G8gV29ybGQ=h IQ= =">>), + <<"Hello World!!">> = MimeDecode(<<"SGVsb)(G8gV29ybGQ=h IQ= =">>), %% No pad <<"Aladdin:open sesam">> = - base64:mime_decode("QWxhZGRpbjpvcG¤\")(VuIHNlc2Ft"), + MimeDecode(<<"QWxhZGRpbjpvcG¤\")(VuIHNlc2Ft">>), %% Encoded base 64 strings may be divided by non base 64 chars. %% In this cases whitespaces. <<"0123456789!@#0^&*();:<>,. []{}">> = - base64:mime_decode( + MimeDecode( <<"MDEy MzQ1Njc4 \tOSFAIzBeJ \nio)(oKTs6 PD4sLi \r\nBbXXt9">>), + %% Zeroes + <<"012">> = MimeDecode(<<"\000M\000D\000E\000y=\000">>), + <<"o">> = MimeDecode(<<"bw==\000">>), + <<"o">> = MimeDecode(<<"bw=\000=">>), ok. %%------------------------------------------------------------------------- @@ -165,39 +177,48 @@ mime_decode(Config) when is_list(Config) -> %% Test base64:mime_decode_to_string/1. mime_decode_to_string(Config) when is_list(Config) -> + MimeDecodeToString = + fun(In) -> + Out = base64:mime_decode_to_string(In), + Out = base64:mime_decode_to_string(binary_to_list(In)) + end, %% Test correct padding - "one" = base64:mime_decode_to_string(<<"b25l">>), - "on" = base64:mime_decode_to_string(<<"b24=">>), - "o" = base64:mime_decode_to_string(<<"bw==">>), + "one" = MimeDecodeToString(<<"b25l">>), + "on" = MimeDecodeToString(<<"b24=">>), + "o" = MimeDecodeToString(<<"bw==">>), %% Test 1 extra padding - "one" = base64:mime_decode_to_string(<<"b25l= =">>), - "on" = base64:mime_decode_to_string(<<"b24== =">>), - "o" = base64:mime_decode_to_string(<<"bw=== =">>), + "one" = MimeDecodeToString(<<"b25l= =">>), + "on" = MimeDecodeToString(<<"b24== =">>), + "o" = MimeDecodeToString(<<"bw=== =">>), %% Test 2 extra padding - "one" = base64:mime_decode_to_string(<<"b25l===">>), - "on" = base64:mime_decode_to_string(<<"b24====">>), - "o" = base64:mime_decode_to_string(<<"bw=====">>), + "one" = MimeDecodeToString(<<"b25l===">>), + "on" = MimeDecodeToString(<<"b24====">>), + "o" = MimeDecodeToString(<<"bw=====">>), %% Test misc embedded padding - "one" = base64:mime_decode_to_string(<<"b2=5l===">>), - "on" = base64:mime_decode_to_string(<<"b=24====">>), - "o" = base64:mime_decode_to_string(<<"b=w=====">>), + "one" = MimeDecodeToString(<<"b2=5l===">>), + "on" = MimeDecodeToString(<<"b=24====">>), + "o" = MimeDecodeToString(<<"b=w=====">>), %% Test misc white space and illegals with embedded padding - "one" = base64:mime_decode_to_string(<<" b~2=\r\n5()l===">>), - "on" = base64:mime_decode_to_string(<<"\tb =2\"¤4=¤= ==">>), - "o" = base64:mime_decode_to_string(<<"\nb=w=====">>), + "one" = MimeDecodeToString(<<" b~2=\r\n5()l===">>), + "on" = MimeDecodeToString(<<"\tb =2\"¤4=¤= ==">>), + "o" = MimeDecodeToString(<<"\nb=w=====">>), %% Two pads "Aladdin:open sesame" = - base64:mime_decode_to_string("QWxhZGRpbjpvc()GVuIHNlc2FtZQ=="), + MimeDecodeToString(<<"QWxhZGRpbjpvc()GVuIHNlc2FtZQ==">>), %% One pad to ignore, followed by more text - "Hello World!!" = base64:mime_decode_to_string(<<"SGVsb)(G8gV29ybGQ=h IQ= =">>), + "Hello World!!" = MimeDecodeToString(<<"SGVsb)(G8gV29ybGQ=h IQ= =">>), %% No pad "Aladdin:open sesam" = - base64:mime_decode_to_string("QWxhZGRpbjpvcG¤\")(VuIHNlc2Ft"), + MimeDecodeToString(<<"QWxhZGRpbjpvcG¤\")(VuIHNlc2Ft">>), %% Encoded base 64 strings may be divided by non base 64 chars. %% In this cases whitespaces. "0123456789!@#0^&*();:<>,. []{}" = - base64:mime_decode_to_string( + MimeDecodeToString( <<"MDEy MzQ1Njc4 \tOSFAIzBeJ \nio)(oKTs6 PD4sLi \r\nBbXXt9">>), + %% Zeroes + "012" = MimeDecodeToString(<<"\000M\000D\000E\000y=\000">>), + "o" = MimeDecodeToString(<<"bw==\000">>), + "o" = MimeDecodeToString(<<"bw=\000=">>), ok. %%------------------------------------------------------------------------- diff --git a/lib/stdlib/test/erl_lint_SUITE.erl b/lib/stdlib/test/erl_lint_SUITE.erl index b76bece07f..5efffc6a5c 100644 --- a/lib/stdlib/test/erl_lint_SUITE.erl +++ b/lib/stdlib/test/erl_lint_SUITE.erl @@ -66,7 +66,8 @@ otp_11851/1,otp_11879/1,otp_13230/1, record_errors/1, otp_11879_cont/1, non_latin1_module/1, otp_14323/1, - get_stacktrace/1, otp_14285/1, otp_14378/1]). + get_stacktrace/1, stacktrace_syntax/1, + otp_14285/1, otp_14378/1]). suite() -> [{ct_hooks,[ts_install_cth]}, @@ -87,7 +88,7 @@ all() -> maps, maps_type, maps_parallel_match, otp_11851, otp_11879, otp_13230, record_errors, otp_11879_cont, non_latin1_module, otp_14323, - get_stacktrace, otp_14285, otp_14378]. + get_stacktrace, stacktrace_syntax, otp_14285, otp_14378]. groups() -> [{unused_vars_warn, [], @@ -4129,6 +4130,40 @@ get_stacktrace(Config) -> run(Config, Ts), ok. +stacktrace_syntax(Config) -> + Ts = [{guard, + <<"t1() -> + try error(foo) + catch _:_:Stk when is_number(Stk) -> ok + end. + ">>, + [], + {errors,[{3,erl_lint,{stacktrace_guard,'Stk'}}],[]}}, + {bound, + <<"t1() -> + Stk = [], + try error(foo) + catch _:_:Stk -> ok + end. + ">>, + [], + {errors,[{4,erl_lint,{stacktrace_bound,'Stk'}}],[]}}, + {guard_and_bound, + <<"t1() -> + Stk = [], + try error(foo) + catch _:_:Stk when is_integer(Stk) -> ok + end. + ">>, + [], + {errors,[{4,erl_lint,{stacktrace_bound,'Stk'}}, + {4,erl_lint,{stacktrace_guard,'Stk'}}],[]}} + ], + + run(Config, Ts), + ok. + + %% Unicode atoms. otp_14285(Config) -> %% A small sample of all the errors and warnings in module erl_lint. diff --git a/lib/stdlib/test/ets_SUITE.erl b/lib/stdlib/test/ets_SUITE.erl index 5a5e282998..07c8b60cbd 100644 --- a/lib/stdlib/test/ets_SUITE.erl +++ b/lib/stdlib/test/ets_SUITE.erl @@ -1682,7 +1682,7 @@ do_random_test() -> ets:delete(Set), verify_etsmem(EtsMem). -%% Ttest various variants of update_element. +%% Test various variants of update_element. update_element(Config) when is_list(Config) -> EtsMem = etsmem(), repeat_for_opts(fun update_element_opts/1), @@ -6369,7 +6369,7 @@ very_big_num(0, Result) -> Result. make_port() -> - open_port({spawn, "efile"}, [eof]). + hd(erlang:ports()). make_pid() -> spawn_link(fun sleeper/0). diff --git a/lib/stdlib/test/filelib_SUITE.erl b/lib/stdlib/test/filelib_SUITE.erl index 1236fe45f4..930cea347f 100644 --- a/lib/stdlib/test/filelib_SUITE.erl +++ b/lib/stdlib/test/filelib_SUITE.erl @@ -33,6 +33,8 @@ -include_lib("common_test/include/ct.hrl"). -include_lib("kernel/include/file.hrl"). +-define(PRIM_FILE, prim_file). + init_per_testcase(_Case, Config) -> Config. @@ -446,10 +448,10 @@ wildcard_symlink(Config) when is_list(Config) -> erl_prim_loader)), ["sub","symlink"] = basenames(Dir, filelib:wildcard(filename:join(Dir, "*"), - prim_file)), + ?PRIM_FILE)), ["symlink"] = basenames(Dir, filelib:wildcard(filename:join(Dir, "symlink"), - prim_file)), + ?PRIM_FILE)), ok = file:delete(AFile), %% The symlink should still be visible even when its target %% has been deleted. @@ -465,10 +467,10 @@ wildcard_symlink(Config) when is_list(Config) -> erl_prim_loader)), ["sub","symlink"] = basenames(Dir, filelib:wildcard(filename:join(Dir, "*"), - prim_file)), + ?PRIM_FILE)), ["symlink"] = basenames(Dir, filelib:wildcard(filename:join(Dir, "symlink"), - prim_file)), + ?PRIM_FILE)), ok end. @@ -497,17 +499,17 @@ is_file_symlink(Config) -> ok -> true = filelib:is_dir(DirAlias), true = filelib:is_dir(DirAlias, erl_prim_loader), - true = filelib:is_dir(DirAlias, prim_file), + true = filelib:is_dir(DirAlias, ?PRIM_FILE), true = filelib:is_file(DirAlias), true = filelib:is_file(DirAlias, erl_prim_loader), - true = filelib:is_file(DirAlias, prim_file), + true = filelib:is_file(DirAlias, ?PRIM_FILE), ok = file:make_symlink(AFile,FileAlias), true = filelib:is_file(FileAlias), true = filelib:is_file(FileAlias, erl_prim_loader), - true = filelib:is_file(FileAlias, prim_file), + true = filelib:is_file(FileAlias, ?PRIM_FILE), true = filelib:is_regular(FileAlias), true = filelib:is_regular(FileAlias, erl_prim_loader), - true = filelib:is_regular(FileAlias, prim_file), + true = filelib:is_regular(FileAlias, ?PRIM_FILE), ok end. @@ -528,11 +530,11 @@ file_props_symlink(Config) -> {_,_} = LastMod = filelib:last_modified(AFile), LastMod = filelib:last_modified(Alias), LastMod = filelib:last_modified(Alias, erl_prim_loader), - LastMod = filelib:last_modified(Alias, prim_file), + LastMod = filelib:last_modified(Alias, ?PRIM_FILE), FileSize = filelib:file_size(AFile), FileSize = filelib:file_size(Alias), FileSize = filelib:file_size(Alias, erl_prim_loader), - FileSize = filelib:file_size(Alias, prim_file) + FileSize = filelib:file_size(Alias, ?PRIM_FILE) end. find_source(Config) when is_list(Config) -> diff --git a/lib/stdlib/test/filename_SUITE.erl b/lib/stdlib/test/filename_SUITE.erl index 4c82ec1c22..f284eb1ed6 100644 --- a/lib/stdlib/test/filename_SUITE.erl +++ b/lib/stdlib/test/filename_SUITE.erl @@ -30,7 +30,6 @@ -export([pathtype_bin/1,rootname_bin/1,split_bin/1]). -export([t_basedir_api/1, t_basedir_xdg/1, t_basedir_windows/1]). -export([safe_relative_path/1]). --export([validate/1]). -include_lib("common_test/include/ct.hrl"). @@ -44,8 +43,7 @@ all() -> absname_bin, absname_bin_2, {group,p}, t_basedir_xdg, t_basedir_windows, - safe_relative_path, - validate]. + safe_relative_path]. groups() -> [{p, [parallel], @@ -109,6 +107,17 @@ absname(Config) when is_list(Config) -> [Drive|":/erlang/src"] = filename:absname([Drive|":erlang/src"]), "a:/erlang" = filename:absname("a:erlang"), + "//foo" = filename:absname("//foo"), + "//foo/bar" = filename:absname("//foo/bar"), + "//foo/\bar" = filename:absname("//foo/\bar"), + "//foo/bar/baz" = filename:absname("//foo/bar\\baz"), + "//foo/bar/baz" = filename:absname("//foo\\bar/baz"), + "//foo" = filename:absname("\\\\foo"), + "//foo/bar" = filename:absname("\\\\foo/bar"), + "//foo/\bar" = filename:absname("\\\\foo/\bar"), + "//foo/bar/baz" = filename:absname("\\\\foo/bar\\baz"), + "//foo/bar/baz" = filename:absname("\\\\foo\\bar/baz"), + file:set_cwd(Cwd), ok; {unix, _} -> @@ -169,6 +178,23 @@ absname_2(Config) when is_list(Config) -> [Drive|":/"]), "a:/erlang" = filename:absname("a:erlang", [Drive|":/"]), + "//foo" = filename:absname("foo","//"), + "//foo/bar" = filename:absname("foo/bar", "//"), + "//foo/bar" = filename:absname("bar", "//foo"), + "//bar" = filename:absname("/bar", "//foo"), + "//foo/bar/baz" = filename:absname("bar/baz", "//foo"), + "//bar/baz" = filename:absname("//bar/baz", "//foo"), + "//\bar" = filename:absname("/\bar", "//foo"), + "//foo" = filename:absname("foo","\\\\"), + "//foo/bar" = filename:absname("foo/bar", "\\\\"), + "//foo/bar" = filename:absname("bar", "\\\\foo"), + "//bar" = filename:absname("/bar", "\\\\foo"), + "//foo/bar/baz" = filename:absname("bar/baz", "\\\\foo"), + "//bar/baz" = filename:absname("\\\\bar/baz", "\\\\foo"), + "//\bar" = filename:absname("/\bar", "\\\\foo"), + "//bar/baz" = filename:absname("\\\\bar/baz", "//foo"), + "//bar/baz" = filename:absname("//bar/baz", "\\\\foo"), + ok; _ -> "/usr/foo" = filename:absname(foo, "/usr"), @@ -246,6 +272,18 @@ dirname(Config) when is_list(Config) -> "A:usr" = filename:dirname("A:usr/foo.erl"), "/usr" = filename:dirname("\\usr\\foo.erl"), "/" = filename:dirname("\\usr"), + "//foo/bar" = filename:dirname("//foo/bar/baz.erl"), + "//foo/\bar" = filename:dirname("//foo/\bar/baz.erl"), + "//foo/bar" = filename:dirname("//foo\\bar/baz.erl"), + "//foo/bar" = filename:dirname("\\\\foo/bar/baz.erl"), + "//foo/\bar" = filename:dirname("\\\\foo/\bar/baz.erl"), + "//foo/bar" = filename:dirname("\\\\foo\\bar/baz.erl"), + "//foo" = filename:dirname("//foo/baz.erl"), + "//foo" = filename:dirname("//foo/\baz.erl"), + "//foo" = filename:dirname("//foo\\baz.erl"), + "//foo" = filename:dirname("\\\\foo/baz.erl"), + "//foo" = filename:dirname("\\\\foo/\baz.erl"), + "//foo" = filename:dirname("\\\\foo\\baz.erl"), "A:" = filename:dirname("A:"); _ -> true end, @@ -291,7 +329,6 @@ join(Config) when is_list(Config) -> %% join/1 and join/2 (OTP-12158) by using help function %% filename_join/2. "/" = filename:join(["/"]), - "/" = filename:join(["//"]), "usr/foo.erl" = filename_join("usr","foo.erl"), "/src/foo.erl" = filename_join(usr, "/src/foo.erl"), "/src/foo.erl" = filename_join("/src/",'foo.erl'), @@ -303,7 +340,6 @@ join(Config) when is_list(Config) -> "a/b/c/d/e/f/g" = filename_join("a//b/c/", "d//e/f/g"), "a/b/c/d/e/f/g" = filename_join("a//b/c", "d//e/f/g"), "/d/e/f/g" = filename_join("a//b/c", "/d//e/f/g"), - "/d/e/f/g" = filename:join("a//b/c", "//d//e/f/g"), "foo/bar" = filename_join([$f,$o,$o,$/,[]], "bar"), @@ -334,6 +370,7 @@ join(Config) when is_list(Config) -> case os:type() of {win32, _} -> + "//" = filename:join(["//"]), "d:/" = filename:join(["D:/"]), "d:/" = filename:join(["D:\\"]), "d:/abc" = filename_join("D:/", "abc"), @@ -347,8 +384,35 @@ join(Config) when is_list(Config) -> "c:/usr/foo.erl" = filename:join(["A:","C:/usr","foo.erl"]), "c:usr/foo.erl" = filename:join(["A:","C:usr","foo.erl"]), "d:/foo" = filename:join([$D, $:, $/, []], "foo"), + "//" = filename:join("\\\\", ""), + "//foo" = filename:join("\\\\", "foo"), + "//foo/bar" = filename:join("\\\\", "foo\\\\bar"), + "//foo/bar/baz" = filename:join("\\\\foo", "bar\\\\baz"), + "//foo/bar/baz" = filename:join("\\\\foo", "bar\\baz"), + "//foo/bar/baz" = filename:join("\\\\foo\\bar", baz), + "//foo/\bar/baz" = filename:join("\\\\foo/\bar", baz), + "//foo/bar/baz" = filename:join("\\\\foo/bar", baz), + "//bar/baz" = filename:join("\\\\foo", "\\\\bar\\baz"), + "//bar/baz" = filename:join("\\\\foo", "//bar\\baz"), + "//bar/baz" = filename:join("\\\\foo", "//bar/baz"), + "//bar/baz" = filename:join("\\\\foo", "\\\\bar/baz"), + "//d/e/f/g" = filename:join("a//b/c", "//d//e/f/g"), + "//" = filename:join("//", ""), + "//foo" = filename:join("//", "foo"), + "//foo/bar" = filename:join("//", "foo\\\\bar"), + "//foo/bar/baz" = filename:join("//foo", "bar\\\\baz"), + "//foo/bar/baz" = filename:join("//foo", "bar\\baz"), + "//foo/bar/baz" = filename:join("//foo\\bar", baz), + "//foo/\bar/baz" = filename:join("//foo/\bar", baz), + "//foo/bar/baz" = filename:join("//foo/bar", baz), + "//bar/baz" = filename:join("//foo", "\\\\bar\\baz"), + "//bar/baz" = filename:join("//foo", "//bar\\baz"), + "//bar/baz" = filename:join("//foo", "//bar/baz"), + "//bar/baz" = filename:join("//foo", "\\\\bar/baz"), ok; _ -> + "/" = filename:join(["//"]), + "/d/e/f/g" = filename:join("a//b/c", "//d//e/f/g"), ok end. @@ -404,6 +468,16 @@ split(Config) when is_list(Config) -> filename:split("a:\\msdev\\include"), ["a:","msdev","include"] = filename:split("a:msdev\\include"), + ["//","foo"] = + filename:split("\\\\foo"), + ["//","foo"] = + filename:split("//foo"), + ["//","foo","bar"] = + filename:split("\\\\foo\\\\bar"), + ["//","foo","baz"] = + filename:split("\\\\foo\\baz"), + ["//","foo","baz"] = + filename:split("//foo\\baz"), ok; _ -> ok @@ -632,7 +706,6 @@ extension_bin(Config) when is_list(Config) -> join_bin(Config) when is_list(Config) -> <<"/">> = filename:join([<<"/">>]), - <<"/">> = filename:join([<<"//">>]), <<"usr/foo.erl">> = filename:join(<<"usr">>,<<"foo.erl">>), <<"/src/foo.erl">> = filename:join(usr, <<"/src/foo.erl">>), <<"/src/foo.erl">> = filename:join([<<"/src/">>,'foo.erl']), @@ -644,7 +717,6 @@ join_bin(Config) when is_list(Config) -> <<"a/b/c/d/e/f/g">> = filename:join([<<"a//b/c/">>, <<"d//e/f/g">>]), <<"a/b/c/d/e/f/g">> = filename:join([<<"a//b/c">>, <<"d//e/f/g">>]), <<"/d/e/f/g">> = filename:join([<<"a//b/c">>, <<"/d//e/f/g">>]), - <<"/d/e/f/g">> = filename:join([<<"a//b/c">>, <<"//d//e/f/g">>]), <<"foo/bar">> = filename:join([$f,$o,$o,$/,[]], <<"bar">>), @@ -697,6 +769,7 @@ join_bin(Config) when is_list(Config) -> case os:type() of {win32, _} -> + <<"//">> = filename:join([<<"//">>]), <<"d:/">> = filename:join([<<"D:/">>]), <<"d:/">> = filename:join([<<"D:\\">>]), <<"d:/abc">> = filename:join([<<"D:/">>, <<"abc">>]), @@ -710,8 +783,35 @@ join_bin(Config) when is_list(Config) -> <<"c:/usr/foo.erl">> = filename:join([<<"A:">>,<<"C:/usr">>,<<"foo.erl">>]), <<"c:usr/foo.erl">> = filename:join([<<"A:">>,<<"C:usr">>,<<"foo.erl">>]), <<"d:/foo">> = filename:join([$D, $:, $/, []], <<"foo">>), + <<"//">> = filename:join(<<"\\\\">>, <<"">>), + <<"//foo">> = filename:join(<<"\\\\">>, <<"foo">>), + <<"//foo/bar">> = filename:join(<<"\\\\">>, <<"foo\\\\bar">>), + <<"//foo/bar/baz">> = filename:join(<<"\\\\foo">>, <<"bar\\\\baz">>), + <<"//bar/baz">> = filename:join(<<"\\\\foo">>, <<"\\\\bar\\baz">>), + <<"//foo/bar/baz">> = filename:join(<<"\\\\foo\\bar">>, baz), + <<"//foo/\bar/baz">> = filename:join(<<"\\\\foo/\bar">>, baz), + <<"//foo/bar/baz">> = filename:join(<<"\\\\foo/bar">>, baz), + <<"//bar/baz">> = filename:join(<<"\\\\foo">>, <<"\\\\bar\\baz">>), + <<"//bar/baz">> = filename:join(<<"\\\\foo">>, <<"//bar\\baz">>), + <<"//bar/baz">> = filename:join(<<"\\\\foo">>, <<"//bar/baz">>), + <<"//bar/baz">> = filename:join(<<"\\\\foo">>, <<"\\\\bar/baz">>), + <<"//d/e/f/g">> = filename:join([<<"a//b/c">>, <<"//d//e/f/g">>]), + <<"//">> = filename:join(<<"//">>, <<"">>), + <<"//foo">> = filename:join(<<"//">>, <<"foo">>), + <<"//foo/bar">> = filename:join(<<"//">>, <<"foo\\\\bar">>), + <<"//foo/bar/baz">> = filename:join(<<"//foo">>, <<"bar\\\\baz">>), + <<"//bar/baz">> = filename:join(<<"//foo">>, <<"\\\\bar\\baz">>), + <<"//foo/bar/baz">> = filename:join(<<"//foo\\bar">>, baz), + <<"//foo/\bar/baz">> = filename:join(<<"//foo/\bar">>, baz), + <<"//foo/bar/baz">> = filename:join(<<"//foo/bar">>, baz), + <<"//bar/baz">> = filename:join(<<"//foo">>, <<"\\\\bar\\baz">>), + <<"//bar/baz">> = filename:join(<<"//foo">>, <<"//bar\\baz">>), + <<"//bar/baz">> = filename:join(<<"//foo">>, <<"//bar/baz">>), + <<"//bar/baz">> = filename:join(<<"//foo">>, <<"\\\\bar/baz">>), ok; _ -> + <<"/">> = filename:join([<<"//">>]), + <<"/d/e/f/g">> = filename:join([<<"a//b/c">>, <<"//d//e/f/g">>]), ok end. @@ -758,6 +858,16 @@ split_bin(Config) when is_list(Config) -> filename:split(<<"a:\\msdev\\include">>), [<<"a:">>,<<"msdev">>,<<"include">>] = filename:split(<<"a:msdev\\include">>), + [<<"//">>,<<"foo">>] = + filename:split(<<"\\\\foo">>), + [<<"//">>,<<"foo">>] = + filename:split(<<"//foo">>), + [<<"//">>,<<"foo">>,<<"bar">>] = + filename:split(<<"\\\\foo\\\\bar">>), + [<<"//">>,<<"foo">>,<<"baz">>] = + filename:split(<<"\\\\foo\\baz">>), + [<<"//">>,<<"foo">>,<<"baz">>] = + filename:split(<<"//foo\\baz">>), ok; _ -> ok @@ -1013,56 +1123,3 @@ basedir_xdg_def(Type,Home,Name) -> Dir <- ["/usr/local/share/","/usr/share/"]]; site_config -> [filename:join(["/etc/xdg",Name])] end. - -validate(Config) when is_list(Config) -> - true = filename:validate(blipp), - false = filename:validate('bli\0pp'), - false = filename:validate('blipp\0'), - true = filename:validate("blipp"), - false = filename:validate("bli"++[0]++"pp"), - false = filename:validate("blipp"++[0]), - true = filename:validate(["one ", blipp, "blopp"]), - false = filename:validate(["one ", 'bli\0pp', "blopp"]), - false = filename:validate(["one ", 'blipp\0', "blopp"]), - false = filename:validate(["one ", 'blipp', "blopp\0"]), - false = filename:validate([0]), - false = filename:validate([]), - false = filename:validate([[[]],[[[[],[[[[[[[[]]], '', [[[[[]]]]]]]]]]]]]]), - false = filename:validate([16#110000]), - false = filename:validate([16#110001]), - false = filename:validate([16#110000*2]), - case file:native_name_encoding() of - latin1 -> - true = filename:validate(lists:seq(1, 255)), - false = filename:validate([256]); - utf8 -> - true = filename:validate(lists:seq(1, 16#D7FF)), - true = filename:validate(lists:seq(16#E000, 16#FFFF)), - true = filename:validate([16#FFFF]), - case os:type() of - {win32, _} -> - false = filename:validate([16#10000]), - true = filename:validate(lists:seq(16#D800,16#DFFF)); - _ -> - true = filename:validate([16#10000]), - true = filename:validate([16#10FFFF]), - lists:foreach(fun (C) -> - false = filename:validate([C]) - end, - lists:seq(16#D800,16#DFFF)) - end - - end, - true = filename:validate(<<1,17,255>>), - false = filename:validate(<<1,0,17,255>>), - false = filename:validate(<<1,17,255,0>>), - false = filename:validate(<<>>), - lists:foreach(fun (N) -> - true = filename:validate(N) - end, - code:get_path()), - ok. - - - - diff --git a/lib/stdlib/test/io_SUITE.erl b/lib/stdlib/test/io_SUITE.erl index e2c73371cd..13929bdbb6 100644 --- a/lib/stdlib/test/io_SUITE.erl +++ b/lib/stdlib/test/io_SUITE.erl @@ -2138,8 +2138,8 @@ otp_14175(_Config) -> "#{}" = p(#{}, 1), "#{...}" = p(#{a => 1}, 1), "#{#{} => a}" = p(#{#{} => a}, 2), - "#{a => 1,...}" = p(#{a => 1, b => 2}, 2), - "#{a => 1,b => 2}" = p(#{a => 1, b => 2}, -1), + mt("#{a => 1,...}", p(#{a => 1, b => 2}, 2)), + mt("#{a => 1,b => 2}", p(#{a => 1, b => 2}, -1)), M = #{kaaaaaaaaaaaaaaaaaaa => v1,kbbbbbbbbbbbbbbbbbbb => v2, kccccccccccccccccccc => v3,kddddddddddddddddddd => v4, diff --git a/lib/stdlib/test/maps_SUITE.erl b/lib/stdlib/test/maps_SUITE.erl index 42e669a799..a75751b31d 100644 --- a/lib/stdlib/test/maps_SUITE.erl +++ b/lib/stdlib/test/maps_SUITE.erl @@ -30,6 +30,7 @@ -export([t_update_with_3/1, t_update_with_4/1, t_get_3/1, t_filter_2/1, t_fold_3/1,t_map_2/1,t_size_1/1, + t_iterator_1/1, t_with_2/1,t_without_2/1]). %%-define(badmap(V,F,Args), {'EXIT', {{badmap,V}, [{maps,F,Args,_}|_]}}). @@ -47,6 +48,7 @@ all() -> [t_update_with_3,t_update_with_4, t_get_3,t_filter_2, t_fold_3,t_map_2,t_size_1, + t_iterator_1, t_with_2,t_without_2]. t_update_with_3(Config) when is_list(Config) -> @@ -127,6 +129,8 @@ t_filter_2(Config) when is_list(Config) -> Pred2 = fun(K,V) -> is_list(K) andalso (V rem 2) =:= 0 end, #{a := 2,c := 4} = maps:filter(Pred1,M), #{"b" := 2,"c" := 4} = maps:filter(Pred2,M), + #{a := 2,c := 4} = maps:filter(Pred1,maps:iterator(M)), + #{"b" := 2,"c" := 4} = maps:filter(Pred2,maps:iterator(M)), %% error case ?badmap(a,filter,[_,a]) = (catch maps:filter(fun(_,_) -> ok end,id(a))), ?badarg(filter,[<<>>,#{}]) = (catch maps:filter(id(<<>>),#{})), @@ -139,6 +143,8 @@ t_fold_3(Config) when is_list(Config) -> Tot0 = lists:sum(Vs), Tot1 = maps:fold(fun({k,_},V,A) -> A + V end, 0, M0), true = Tot0 =:= Tot1, + Tot2 = maps:fold(fun({k,_},V,A) -> A + V end, 0, maps:iterator(M0)), + true = Tot0 =:= Tot2, %% error case ?badmap(a,fold,[_,0,a]) = (catch maps:fold(fun(_,_,_) -> ok end,0,id(a))), @@ -151,12 +157,48 @@ t_map_2(Config) when is_list(Config) -> #{ {k,1} := 1, {k,200} := 200} = M0, M1 = maps:map(fun({k,_},V) -> V + 42 end, M0), #{ {k,1} := 43, {k,200} := 242} = M1, + M2 = maps:map(fun({k,_},V) -> V + 42 end, maps:iterator(M0)), + #{ {k,1} := 43, {k,200} := 242} = M2, %% error case ?badmap(a,map,[_,a]) = (catch maps:map(fun(_,_) -> ok end, id(a))), ?badarg(map,[<<>>,#{}]) = (catch maps:map(id(<<>>),#{})), ok. +t_iterator_1(Config) when is_list(Config) -> + + %% Small map test + M0 = #{ a => 1, b => 2 }, + I0 = maps:iterator(M0), + {K1,V1,I1} = maps:next(I0), + {K2,V2,I2} = maps:next(I1), + none = maps:next(I2), + + KVList = lists:sort([{K1,V1},{K2,V2}]), + KVList = lists:sort(maps:to_list(M0)), + + %% Large map test + + Vs2 = lists:seq(1,200), + M2 = maps:from_list([{{k,I},I}||I<-Vs2]), + KVList2 = lists:sort(iter_kv(maps:iterator(M2))), + KVList2 = lists:sort(maps:to_list(M2)), + + %% Larger map test + + Vs3 = lists:seq(1,10000), + M3 = maps:from_list([{{k,I},I}||I<-Vs3]), + KVList3 = lists:sort(iter_kv(maps:iterator(M3))), + KVList3 = lists:sort(maps:to_list(M3)), + ok. + +iter_kv(I) -> + case maps:next(I) of + none -> + []; + {K,V,NI} -> + [{K,V} | iter_kv(NI)] + end. t_size_1(Config) when is_list(Config) -> 0 = maps:size(#{}), diff --git a/lib/stdlib/test/property_test/README b/lib/stdlib/test/property_test/README new file mode 100644 index 0000000000..57602bf719 --- /dev/null +++ b/lib/stdlib/test/property_test/README @@ -0,0 +1,12 @@ + +%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% +%%% %%% +%%% WARNING %%% +%%% %%% +%%% This is experimental code which may be changed or removed %%% +%%% anytime without any warning. %%% +%%% %%% +%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% + +The test in this directory are written assuming that the user has a QuickCheck license. They are to be run manually. Some may be possible to be run with other tools, e.g. PropEr. + diff --git a/lib/stdlib/test/property_test/uri_string_recompose.erl b/lib/stdlib/test/property_test/uri_string_recompose.erl new file mode 100644 index 0000000000..e51a671172 --- /dev/null +++ b/lib/stdlib/test/property_test/uri_string_recompose.erl @@ -0,0 +1,361 @@ +%% +%% %CopyrightBegin% +%% +%% Copyright Ericsson AB 2008-2017. All Rights Reserved. +%% +%% Licensed under the Apache License, Version 2.0 (the "License"); +%% you may not use this file except in compliance with the License. +%% You may obtain a copy of the License at +%% +%% http://www.apache.org/licenses/LICENSE-2.0 +%% +%% Unless required by applicable law or agreed to in writing, software +%% distributed under the License is distributed on an "AS IS" BASIS, +%% WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +%% See the License for the specific language governing permissions and +%% limitations under the License. +%% +%% %CopyrightEnd% +%% +-module(uri_string_recompose). + +-compile(export_all). + +-proptest(eqc). +-proptest([triq,proper]). + +-ifndef(EQC). +-ifndef(PROPER). +-ifndef(TRIQ). +-define(EQC,true). +-endif. +-endif. +-endif. + +-ifdef(EQC). +-include_lib("eqc/include/eqc.hrl"). +-define(MOD_eqc,eqc). + +-else. +-ifdef(PROPER). +-include_lib("proper/include/proper.hrl"). +-define(MOD_eqc,proper). + +-else. +-ifdef(TRIQ). +-define(MOD_eqc,triq). +-include_lib("triq/include/triq.hrl"). + +-endif. +-endif. +-endif. + + +-define(STRING_REST(MatchStr, Rest), <<MatchStr/utf8, Rest/binary>>). + +-define(SCHEME, {scheme, scheme()}). +-define(USER, {userinfo, unicode()}). +-define(HOST, {host, host_map()}). +-define(PORT, {port, port()}). +-define(PATH_ABE, {path, path_abempty_map()}). +-define(PATH_ABS, {path, path_absolute_map()}). +-define(PATH_NOS, {path, path_noscheme_map()}). +-define(PATH_ROO, {path, path_rootless_map()}). +-define(PATH_EMP, {path, path_empty_map()}). +-define(QUERY, {query, query_map()}). +-define(FRAGMENT, {fragment, fragment_map()}). + + +%%%======================================================================== +%%% Properties +%%%======================================================================== + +prop_recompose() -> + ?FORALL(Map, map(), + Map =:= uri_string:parse(uri_string:recompose(Map)) + ). + +%% Stats +prop_map_key_length_collect() -> + ?FORALL(List, map(), + collect(length(maps:keys(List)), true)). + +prop_map_collect() -> + ?FORALL(List, map(), + collect(lists:sort(maps:keys(List)), true)). + +prop_scheme_collect() -> + ?FORALL(List, scheme(), + collect(length(List), true)). + + +%%%======================================================================== +%%% Generators +%%%======================================================================== + +map() -> + ?LET(Gen, comp_proplist(), proplist_to_map(Gen)). + +comp_proplist() -> + frequency([ + {2, [?SCHEME,?PATH_ABS]}, + {2, [?SCHEME,?PATH_ROO]}, + {2, [?SCHEME,?PATH_EMP]}, + {2, [?SCHEME,?HOST,?PATH_ABE]}, + {2, [?SCHEME,?USER,?HOST,?PATH_ABE]}, + {2, [?SCHEME,?HOST,?PORT,?PATH_ABE]}, + {2, [?SCHEME,?USER,?HOST,?PORT,?PATH_ABE]}, + + {2, [?PATH_ABS]}, + {2, [?PATH_NOS]}, + {2, [?PATH_EMP]}, + {2, [?HOST,?PATH_ABE]}, + {2, [?USER,?HOST,?PATH_ABE]}, + {2, [?HOST,?PORT,?PATH_ABE]}, + {2, [?USER,?HOST,?PORT,?PATH_ABE]}, + + + {2, [?SCHEME,?PATH_ABS,?QUERY]}, + {2, [?SCHEME,?PATH_ROO,?QUERY]}, + {2, [?SCHEME,?PATH_EMP,?QUERY]}, + {2, [?SCHEME,?HOST,?PATH_ABE,?QUERY]}, + {2, [?SCHEME,?USER,?HOST,?PATH_ABE,?QUERY]}, + {2, [?SCHEME,?HOST,?PORT,?PATH_ABE,?QUERY]}, + {2, [?SCHEME,?USER,?HOST,?PORT,?PATH_ABE,?QUERY]}, + + {2, [?PATH_ABS,?QUERY]}, + {2, [?PATH_NOS,?QUERY]}, + {2, [?PATH_EMP,?QUERY]}, + {2, [?HOST,?PATH_ABE,?QUERY]}, + {2, [?USER,?HOST,?PATH_ABE,?QUERY]}, + {2, [?HOST,?PORT,?PATH_ABE,?QUERY]}, + {2, [?USER,?HOST,?PORT,?PATH_ABE,?QUERY]}, + + + {2, [?SCHEME,?PATH_ABS,?FRAGMENT]}, + {2, [?SCHEME,?PATH_ROO,?FRAGMENT]}, + {2, [?SCHEME,?PATH_EMP,?FRAGMENT]}, + {2, [?SCHEME,?HOST,?PATH_ABE,?FRAGMENT]}, + {2, [?SCHEME,?USER,?HOST,?PATH_ABE,?FRAGMENT]}, + {2, [?SCHEME,?HOST,?PORT,?PATH_ABE,?FRAGMENT]}, + {2, [?SCHEME,?USER,?HOST,?PORT,?PATH_ABE,?FRAGMENT]}, + + {2, [?PATH_ABS,?FRAGMENT]}, + {2, [?PATH_NOS,?FRAGMENT]}, + {2, [?PATH_EMP,?FRAGMENT]}, + {2, [?HOST,?PATH_ABE,?FRAGMENT]}, + {2, [?USER,?HOST,?PATH_ABE,?FRAGMENT]}, + {2, [?HOST,?PORT,?PATH_ABE,?FRAGMENT]}, + {2, [?USER,?HOST,?PORT,?PATH_ABE,?FRAGMENT]}, + + + {2, [?SCHEME,?PATH_ABS,?QUERY,?FRAGMENT]}, + {2, [?SCHEME,?PATH_ROO,?QUERY,?FRAGMENT]}, + {2, [?SCHEME,?PATH_EMP,?QUERY,?FRAGMENT]}, + {2, [?SCHEME,?HOST,?PATH_ABE,?QUERY,?FRAGMENT]}, + {2, [?SCHEME,?USER,?HOST,?PATH_ABE,?QUERY,?FRAGMENT]}, + {2, [?SCHEME,?HOST,?PORT,?PATH_ABE,?QUERY,?FRAGMENT]}, + {2, [?SCHEME,?USER,?HOST,?PORT,?PATH_ABE,?QUERY,?FRAGMENT]}, + + {2, [?PATH_ABS,?QUERY,?FRAGMENT]}, + {2, [?PATH_NOS,?QUERY,?FRAGMENT]}, + {2, [?PATH_EMP,?QUERY,?FRAGMENT]}, + {2, [?HOST,?PATH_ABE,?QUERY,?FRAGMENT]}, + {2, [?USER,?HOST,?PATH_ABE,?QUERY,?FRAGMENT]}, + {2, [?HOST,?PORT,?PATH_ABE,?QUERY,?FRAGMENT]}, + {2, [?USER,?HOST,?PORT,?PATH_ABE,?QUERY,?FRAGMENT]} + ]). + + +%%------------------------------------------------------------------------- +%% Path +%%------------------------------------------------------------------------- +path_abempty_map() -> + frequency([{90, path_abe_map()}, + {10, path_empty_map()}]). + +path_abe_map() -> + ?SIZED(Length, path_abe_map(Length, [])). +%% +path_abe_map(0, Segments) -> + ?LET(Gen, Segments, lists:append(Gen)); +path_abe_map(N, Segments) -> + path_abe_map(N-1, [slash(),segment()|Segments]). + + +path_absolute_map() -> + ?SIZED(Length, path_absolute_map(Length, [])). +%% +path_absolute_map(0, Segments) -> + ?LET(Gen, [slash(),segment_nz()|Segments], lists:append(Gen)); +path_absolute_map(N, Segments) -> + path_absolute_map(N-1, [slash(),segment()|Segments]). + + +path_noscheme_map() -> + ?SIZED(Length, path_noscheme_map(Length, [])). +%% +path_noscheme_map(0, Segments) -> + ?LET(Gen, [segment_nz_nc()|Segments], lists:append(Gen)); +path_noscheme_map(N, Segments) -> + path_noscheme_map(N-1, [slash(),segment()|Segments]). + +path_rootless_map() -> + ?SIZED(Length, path_rootless_map(Length, [])). +%% +path_rootless_map(0, Segments) -> + ?LET(Gen, [segment_nz()|Segments], lists:append(Gen)); +path_rootless_map(N, Segments) -> + path_rootless_map(N-1, [slash(),segment()|Segments]). + + +segment_nz() -> + non_empty(segment()). + +segment_nz_nc() -> + non_empty(list(frequency([{30, unreserved()}, + {10, sub_delims()}, + {10, unicode_char()}, + {5, oneof([$@])} + ]))). + + +segment() -> + list(frequency([{30, unreserved()}, + {10, sub_delims()}, + {10, unicode_char()}, + {5, oneof([$:, $@])} + ])). + +slash() -> + "/". + +path_empty_map() -> + "". + + +%%------------------------------------------------------------------------- +%% Path +%%------------------------------------------------------------------------- +host_map() -> + frequency([{30, reg_name()}, + {30, ip_address()} + ]). + + +reg_name() -> + list(frequency([{30, alpha()}, + {10, sub_delims()}, + {10, unicode_char()} + ])). + +ip_address() -> + oneof(["127.0.0.1", "::127.0.0.1", + "2001:0db8:0000:0000:0000:0000:1428:07ab", + "2001:0db8:0000:0000:0000::1428:07ab", + "2001:0db8:0:0:0:0:1428:07ab", + "2001:0db8:0::0:1428:07ab"]). + +%% Generating only reg-names +host_uri() -> + non_empty(list(frequency([{30, unreserved()}, + {10, sub_delims()}, + {10, pct_encoded()} + ]))). + +%%------------------------------------------------------------------------- +%% Port, Query, Fragment +%%------------------------------------------------------------------------- +port() -> + frequency([{10, undefined}, + {10, range(1,65535)} + ]). + +query_map() -> + unicode(). + + +query_uri() -> + [$?| non_empty(list(frequency([{20, pchar()}, + {5, oneof([$/, $?])} % punctuation + ])))]. + +fragment_map() -> + unicode(). + +fragment_uri() -> + [$?| non_empty(list(frequency([{20, pchar()}, + {5, oneof([$/, $?])} % punctuation + ])))]. + + +%%------------------------------------------------------------------------- +%% Scheme +%%------------------------------------------------------------------------- +scheme() -> + ?SIZED(Length, scheme_start(Length, [])). +%% +scheme_start(0, L) -> + ?LET(Gen, L, lists:reverse(Gen)); +scheme_start(N, L) -> + scheme(N-1,[alpha()|L]). + +scheme(0, L) -> + ?LET(Gen, L, lists:reverse(Gen)); +scheme(N, L) -> + scheme(N-1, [scheme_char()|L]). + + +%%------------------------------------------------------------------------- +%% Misc +%%------------------------------------------------------------------------- +unicode() -> + list(frequency([{20, alpha()}, % alpha + {10, digit()}, % digit + {10, unicode_char()} % unicode + ])). + +scheme_char() -> + frequency([{20, alpha()}, % alpha + {20, digit()}, % digit + {5, oneof([$+, $-, $.])} % punctuation + ]). + +sub_delims() -> + oneof([$!, $$, $&, $', $(, $), + $*, $+, $,,$;, $=]). + +pchar() -> + frequency([{20, unreserved()}, + {5, pct_encoded()}, + {5, sub_delims()}, + {1, oneof([$:, $@])} % punctuation + ]). + +unreserved() -> + frequency([{20, alpha()}, + {5, digit()}, + {1, oneof([$-, $., $_, $~])} % punctuation + ]). + +unicode_char() -> + range(913, 1023). + +alpha() -> + frequency([{20, range($a, $z)}, % letters + {20, range($A, $Z)}]). % letters + +digit() -> + range($0, $9). % numbers + +pct_encoded() -> + oneof(["%C3%A4", "%C3%A5", "%C3%B6"]). + + +%%%======================================================================== +%%% Helpers +%%%======================================================================== +proplist_to_map(L) -> + lists:foldl(fun({K,V},M) -> M#{K => V}; + (_,M) -> M + end, #{}, L). diff --git a/lib/stdlib/test/qlc_SUITE.erl b/lib/stdlib/test/qlc_SUITE.erl index 5e9e03e410..8f8a0f6e73 100644 --- a/lib/stdlib/test/qlc_SUITE.erl +++ b/lib/stdlib/test/qlc_SUITE.erl @@ -1695,28 +1695,7 @@ sort(Config) when is_list(Config) -> [true || I <- lists:seq(1, 50000), not ets:insert(E, {I, I})], H = qlc:q([{X,Y} || X <- [a,b], Y <- qlc:sort(ets:table(E))]), 100000 = length(qlc:e(H)), - ets:delete(E)">>, - - begin - TmpDir = ?privdir, - [<<"TE = process_flag(trap_exit, true), - E = ets:new(foo, []), - [true || I <- lists:seq(1, 50000), not ets:insert(E, {I, I})], - Ports = erlang:ports(), - H = qlc:q([{X,Y} || X <- [a,b], - begin - [P] = erlang:ports() -- Ports, - exit(P, port_exit), - true - end, - Y <- qlc:sort(ets:table(E), - [{tmpdir,\"">>, - TmpDir, <<"\"}])]), - {error, qlc, {file_error, _, _}} = (catch qlc:e(H)), - receive {'EXIT', _, port_exit} -> ok end, - ets:delete(E), - process_flag(trap_exit, TE)">>] - end + ets:delete(E)">> ], run(Config, Ts), @@ -7871,7 +7850,7 @@ run_test(Config, Extra, {cres, Body, Opts, ExpectedCompileReturn}) -> {module, _} = code:load_abs(AbsFile, Mod), Ms0 = erlang:process_info(self(),messages), - Before = {{get(), lists:sort(ets:all()), Ms0}, pps()}, + Before = {{lget(), lists:sort(ets:all()), Ms0}, pps()}, %% Prepare the check that the qlc module does not call qlc_pt. _ = [unload_pt() || {file, Name} <- [code:is_loaded(qlc_pt)], @@ -7903,7 +7882,7 @@ run_test(Config, Extra, Body) -> wait_for_expected(R, {Strict0,PPS0}=Before, SourceFile, Wait) -> Ms = erlang:process_info(self(),messages), - After = {_,PPS1} = {{get(), lists:sort(ets:all()), Ms}, pps()}, + After = {_,PPS1} = {{lget(), lists:sort(ets:all()), Ms}, pps()}, case {R, After} of {ok, Before} -> ok; @@ -7931,6 +7910,18 @@ wait_for_expected(R, {Strict0,PPS0}=Before, SourceFile, Wait) -> expected({ok,Before}, {R,After}, SourceFile) end. +%% The qlc modules uses the process dictionary for storing names of files. +lget() -> + lists:sort([T || {K, _} = T <- get(), is_qlc_key(K)]). + +%% Copied from the qlc module. +-define(LCACHE_FILE(Ref), {Ref, '$_qlc_cache_tmpfiles_'}). +-define(MERGE_JOIN_FILE, '$_qlc_merge_join_tmpfiles_'). + +is_qlc_key(?LCACHE_FILE(_)) -> true; +is_qlc_key(?MERGE_JOIN_FILE) -> true; +is_qlc_key(_) -> false. + unload_pt() -> erlang:garbage_collect(), % get rid of references to qlc_pt... _ = code:purge(qlc_pt), diff --git a/lib/stdlib/test/stdlib_bench_SUITE.erl b/lib/stdlib/test/stdlib_bench_SUITE.erl index 8670e7029c..2a9981bb9e 100644 --- a/lib/stdlib/test/stdlib_bench_SUITE.erl +++ b/lib/stdlib/test/stdlib_bench_SUITE.erl @@ -1,7 +1,7 @@ %% %% %CopyrightBegin% %% -%% Copyright Ericsson AB 2012-2016. All Rights Reserved. +%% Copyright Ericsson AB 2012-2017. All Rights Reserved. %% %% Licensed under the Apache License, Version 2.0 (the "License"); %% you may not use this file except in compliance with the License. @@ -28,13 +28,20 @@ suite() -> [{ct_hooks,[{ts_install_cth,[{nodenames,2}]}]}]. all() -> - [{group,unicode}]. + [{group,unicode}, {group,base64}]. groups() -> [{unicode,[{repeat,5}], [norm_nfc_list, norm_nfc_deep_l, norm_nfc_binary, string_lexemes_list, string_lexemes_binary - ]}]. + ]}, + {base64,[{repeat,5}], + [decode_binary, decode_binary_to_string, + decode_list, decode_list_to_string, + encode_binary, encode_binary_to_string, + encode_list, encode_list_to_string, + mime_binary_decode, mime_binary_decode_to_string, + mime_list_decode, mime_list_decode_to_string]}]. init_per_group(_GroupName, Config) -> Config. @@ -105,3 +112,97 @@ norm_data(Config) -> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% +decode_binary(_Config) -> + test(decode, encoded_binary()). + +decode_binary_to_string(_Config) -> + test(decode_to_string, encoded_binary()). + +decode_list(_Config) -> + test(decode, encoded_list()). + +decode_list_to_string(_Config) -> + test(decode_to_string, encoded_list()). + +encode_binary(_Config) -> + test(encode, binary()). + +encode_binary_to_string(_Config) -> + test(encode_to_string, binary()). + +encode_list(_Config) -> + test(encode, list()). + +encode_list_to_string(_Config) -> + test(encode_to_string, list()). + +mime_binary_decode(_Config) -> + test(mime_decode, encoded_binary()). + +mime_binary_decode_to_string(_Config) -> + test(mime_decode_to_string, encoded_binary()). + +mime_list_decode(_Config) -> + test(mime_decode, encoded_list()). + +mime_list_decode_to_string(_Config) -> + test(mime_decode_to_string, encoded_list()). + +-define(SIZE, 10000). +-define(N, 1000). + +encoded_binary() -> + list_to_binary(encoded_list()). + +encoded_list() -> + L = random_byte_list(round(?SIZE*0.75)), + base64:encode_to_string(L). + +binary() -> + list_to_binary(list()). + +list() -> + random_byte_list(?SIZE). + +test(Func, Data) -> + F = fun() -> loop(?N, Func, Data) end, + {Time, ok} = timer:tc(fun() -> lspawn(F) end), + report_base64(Time). + +loop(0, _F, _D) -> garbage_collect(), ok; +loop(N, F, D) -> + _ = base64:F(D), + loop(N - 1, F, D). + +lspawn(Fun) -> + {Pid, Ref} = spawn_monitor(fun() -> exit(Fun()) end), + receive + {'DOWN', Ref, process, Pid, Rep} -> Rep + end. + +report_base64(Time) -> + Tps = round((?N*1000000)/Time), + ct_event:notify(#event{name = benchmark_data, + data = [{suite, "stdlib_base64"}, + {value, Tps}]}), + Tps. + +%% Copied from base64_SUITE.erl. + +random_byte_list(N) -> + random_byte_list(N, []). + +random_byte_list(0, Acc) -> + Acc; +random_byte_list(N, Acc) -> + random_byte_list(N-1, [rand:uniform(255)|Acc]). + +make_big_binary(N) -> + list_to_binary(mbb(N, [])). + +mbb(N, Acc) when N > 256 -> + B = list_to_binary(lists:seq(0, 255)), + mbb(N - 256, [B | Acc]); +mbb(N, Acc) -> + B = list_to_binary(lists:seq(0, N-1)), + lists:reverse(Acc, B). diff --git a/lib/stdlib/test/string_SUITE.erl b/lib/stdlib/test/string_SUITE.erl index 05f18ef238..d02a6eac0a 100644 --- a/lib/stdlib/test/string_SUITE.erl +++ b/lib/stdlib/test/string_SUITE.erl @@ -48,6 +48,7 @@ %% Run tests when debugging them -export([debug/0, time_func/4]). +-compile([nowarn_deprecated_function]). suite() -> [{ct_hooks,[ts_install_cth]}, @@ -92,14 +93,11 @@ end_per_testcase(_Case, _Config) -> ok. debug() -> - Config = [{data_dir, ?MODULE_STRING++"_data"}], + Config = [{data_dir, "./" ++ ?MODULE_STRING++"_data"}], [io:format("~p:~p~n",[Test,?MODULE:Test(Config)]) || {_,Tests} <- groups(), Test <- Tests]. -define(TEST(B,C,D), test(?LINE,?FUNCTION_NAME,B,C,D, true)). --define(TEST_EQ(B,C,D), - test(?LINE,?FUNCTION_NAME,B,C,D, true), - test(?LINE,?FUNCTION_NAME,hd(C),[B|tl(C),D, true)). -define(TEST_NN(B,C,D), test(?LINE,?FUNCTION_NAME,B,C,D, false), @@ -294,6 +292,7 @@ trim(_) -> ?TEST(["..h", ".e", <<"j..">>], [both, ". "], "h.ej"), ?TEST(["..h", <<".ejsa"/utf8>>, "n.."], [both, ". "], "h.ejsan"), %% Test that it behaves with graphemes (i.e. nfd tests are the hard part) + ?TEST([1013,101,778,101,101], [trailing, [101]], [1013,101,778]), ?TEST("aaåaa", [both, "a"], "å"), ?TEST(["aaa",778,"äöoo"], [both, "ao"], "åäö"), ?TEST([<<"aaa">>,778,"äöoo"], [both, "ao"], "åäö"), @@ -353,6 +352,7 @@ take(_) -> ?TEST([<<>>,<<"..">>, " h.ej", <<" ..">>], [Chars, true, leading], {".. ", "h.ej .."}), ?TEST(["..h", <<".ejsa"/utf8>>, "n.."], [Chars, true, leading], {"..", "h.ejsan.."}), %% Test that it behaves with graphemes (i.e. nfd tests are the hard part) + ?TEST([101,778], [[[101, 779]], true], {[101,778], []}), ?TEST(["aaee",778,"äöoo"], [[[$e,778]], true, leading], {"aae", [$e,778|"äöoo"]}), ?TEST([<<"aae">>,778,"äöoo"], [[[$e,778]],true,leading], {"aa", [$e,778|"äöoo"]}), ?TEST([<<"e">>,778,"åäöe", <<778/utf8>>], [[[$e,778]], true, leading], {[], [$e,778]++"åäöe"++[778]}), @@ -713,29 +713,123 @@ nth_lexeme(_) -> meas(Config) -> + Parent = self(), + Exec = fun() -> + DataDir0 = proplists:get_value(data_dir, Config), + DataDir = filename:join(lists:droplast(filename:split(DataDir0))), + case proplists:get_value(profile, Config, false) of + false -> + do_measure(DataDir); + eprof -> + eprof:profile(fun() -> do_measure(DataDir) end, [set_on_spawn]), + eprof:stop_profiling(), + eprof:analyze(), + eprof:stop() + end, + Parent ! {test_done, self()}, + normal + end, + ct:timetrap({minutes,2}), case ct:get_timetrap_info() of {_,{_,Scale}} when Scale > 1 -> {skip,{will_not_run_in_debug,Scale}}; - _ -> % No scaling - DataDir = proplists:get_value(data_dir, Config), - TestDir = filename:dirname(string:trim(DataDir, trailing, "/")), - do_measure(TestDir) + _ -> % No scaling, run at most 1.5 min + Tester = spawn(Exec), + receive {test_done, Tester} -> ok + after 90000 -> + io:format("Timelimit reached stopping~n",[]), + exit(Tester, die) + end, + ok end. -do_measure(TestDir) -> - File = filename:join(TestDir, ?MODULE_STRING ++ ".erl"), +do_measure(DataDir) -> + File = filename:join([DataDir,"unicode_util_SUITE_data","NormalizationTest.txt"]), io:format("File ~s ",[File]), {ok, Bin} = file:read_file(File), io:format("~p~n",[byte_size(Bin)]), Do = fun(Name, Func, Mode) -> - {N, Mean, Stddev, _} = time_func(Func, Mode, Bin, 50), - io:format("~10w ~6w ~6.2fms ±~4.2fms #~.2w gc included~n", + {N, Mean, Stddev, _} = time_func(Func, Mode, Bin, 20), + io:format("~15w ~6w ~6.2fms ±~5.2fms #~.2w gc included~n", [Name, Mode, Mean/1000, Stddev/1000, N]) end, + Do2 = fun(Name, Func, Mode) -> + {N, Mean, Stddev, _} = time_func(Func, binary, <<>>, 20), + io:format("~15w ~6w ~6.2fms ±~5.2fms #~.2w gc included~n", + [Name, Mode, Mean/1000, Stddev/1000, N]) + end, io:format("----------------------~n"), - Do(tokens, fun(Str) -> string:tokens(Str, [$\n,$\r]) end, list), + + Do(old_tokens, fun(Str) -> string:tokens(Str, [$\n,$\r]) end, list), Tokens = {lexemes, fun(Str) -> string:lexemes(Str, [$\n,$\r]) end}, [Do(Name,Fun,Mode) || {Name,Fun} <- [Tokens], Mode <- [list, binary]], + + S0 = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxy.....", + S0B = <<"xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxy.....">>, + Do2(old_strip_l, repeat(fun() -> string:strip(S0, left, $x) end), list), + Do2(trim_l, repeat(fun() -> string:trim(S0, leading, [$x]) end), list), + Do2(trim_l, repeat(fun() -> string:trim(S0B, leading, [$x]) end), binary), + Do2(old_strip_r, repeat(fun() -> string:strip(S0, right, $.) end), list), + Do2(trim_t, repeat(fun() -> string:trim(S0, trailing, [$.]) end), list), + Do2(trim_t, repeat(fun() -> string:trim(S0B, trailing, [$.]) end), binary), + + Do2(old_chr_sub, repeat(fun() -> string:sub_string(S0, string:chr(S0, $.)) end), list), + Do2(old_str_sub, repeat(fun() -> string:sub_string(S0, string:str(S0, [$.])) end), list), + Do2(find, repeat(fun() -> string:find(S0, [$.]) end), list), + Do2(find, repeat(fun() -> string:find(S0B, [$.]) end), binary), + Do2(old_str_sub2, repeat(fun() -> N = string:str(S0, "xy.."), + {string:sub_string(S0,1,N), string:sub_string(S0,N+4)} end), list), + Do2(split, repeat(fun() -> string:split(S0, "xy..") end), list), + Do2(split, repeat(fun() -> string:split(S0B, "xy..") end), binary), + + Do2(old_rstr_sub, repeat(fun() -> string:sub_string(S0, string:rstr(S0, [$y])) end), list), + Do2(find_t, repeat(fun() -> string:find(S0, [$y], trailing) end), list), + Do2(find_t, repeat(fun() -> string:find(S0B, [$y], trailing) end), binary), + Do2(old_rstr_sub2, repeat(fun() -> N = string:rstr(S0, "y.."), + {string:sub_string(S0,1,N), string:sub_string(S0,N+3)} end), list), + Do2(split_t, repeat(fun() -> string:split(S0, "y..", trailing) end), list), + Do2(split_t, repeat(fun() -> string:split(S0B, "y..", trailing) end), binary), + + Do2(old_span, repeat(fun() -> N=string:span(S0, [$x, $y]), + {string:sub_string(S0,1,N),string:sub_string(S0,N+1)} + end), list), + Do2(take, repeat(fun() -> string:take(S0, [$x, $y]) end), list), + Do2(take, repeat(fun() -> string:take(S0B, [$x, $y]) end), binary), + + Do2(old_cspan, repeat(fun() -> N=string:cspan(S0, [$.,$y]), + {string:sub_string(S0,1,N),string:sub_string(S0,N+1)} + end), list), + Do2(take_c, repeat(fun() -> string:take(S0, [$.,$y], true) end), list), + Do2(take_c, repeat(fun() -> string:take(S0B, [$.,$y], true) end), binary), + + Do2(old_substr, repeat(fun() -> string:substr(S0, 21, 15) end), list), + Do2(slice, repeat(fun() -> string:slice(S0, 20, 15) end), list), + Do2(slice, repeat(fun() -> string:slice(S0B, 20, 15) end), binary), + + io:format("--~n",[]), + NthTokens = {nth_lexemes, fun(Str) -> string:nth_lexeme(Str, 18000, [$\n,$\r]) end}, + [Do(Name,Fun,Mode) || {Name,Fun} <- [NthTokens], Mode <- [list, binary]], + Do2(take_t, repeat(fun() -> string:take(S0, [$.,$y], false, trailing) end), list), + Do2(take_t, repeat(fun() -> string:take(S0B, [$.,$y], false, trailing) end), binary), + Do2(take_tc, repeat(fun() -> string:take(S0, [$x], true, trailing) end), list), + Do2(take_tc, repeat(fun() -> string:take(S0B, [$x], true, trailing) end), binary), + + Length = {length, fun(Str) -> string:length(Str) end}, + [Do(Name,Fun,Mode) || {Name,Fun} <- [Length], Mode <- [list, binary]], + + Reverse = {reverse, fun(Str) -> string:reverse(Str) end}, + [Do(Name,Fun,Mode) || {Name,Fun} <- [Reverse], Mode <- [list, binary]], + + ok. + +repeat(F) -> + fun(_) -> repeat_1(F,20000) end. + +repeat_1(F, N) when N > 0 -> + F(), + repeat_1(F, N-1); +repeat_1(_, _) -> + erlang:garbage_collect(), ok. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% @@ -865,8 +959,6 @@ check_types_1({list, _},{list, undefined}) -> ok; check_types_1({list, _},{list, codepoints}) -> ok; -check_types_1({list, _},{list, {list, codepoints}}) -> - ok; check_types_1({list, {list, _}},{list, {list, codepoints}}) -> ok; check_types_1(mixed,_) -> diff --git a/lib/stdlib/test/supervisor_1.erl b/lib/stdlib/test/supervisor_1.erl index 419026749b..c3ccacc587 100644 --- a/lib/stdlib/test/supervisor_1.erl +++ b/lib/stdlib/test/supervisor_1.erl @@ -42,6 +42,8 @@ start_child(error) -> set -> gen_server:start_link(?MODULE, error, []) end; +start_child({return, Term}) -> + Term; start_child(Extra) -> {ok, Pid} = gen_server:start_link(?MODULE, normal, []), diff --git a/lib/stdlib/test/supervisor_SUITE.erl b/lib/stdlib/test/supervisor_SUITE.erl index cd2c6b0cbb..761df8eb40 100644 --- a/lib/stdlib/test/supervisor_SUITE.erl +++ b/lib/stdlib/test/supervisor_SUITE.erl @@ -39,6 +39,9 @@ sup_start_ignore_temporary_child_start_child_simple/1, sup_start_ignore_permanent_child_start_child_simple/1, sup_start_error_return/1, sup_start_fail/1, + sup_start_child_returns_error/1, + sup_start_restart_child_returns_error/1, + sup_start_child_returns_error_simple/1, sup_start_map/1, sup_start_map_simple/1, sup_start_map_faulty_specs/1, sup_stop_infinity/1, sup_stop_timeout/1, sup_stop_brutal_kill/1, @@ -65,14 +68,16 @@ simple_one_for_one_extra/1, simple_one_for_one_shutdown/1]). %% Misc tests --export([child_unlink/1, tree/1, count_children/1, +-export([child_unlink/1, tree/1, count_children/1, count_children_supervisor/1, count_restarting_children/1, get_callback_module/1, do_not_save_start_parameters_for_temporary_children/1, do_not_save_child_specs_for_temporary_children/1, simple_one_for_one_scale_many_temporary_children/1, simple_global_supervisor/1, hanging_restart_loop/1, + hanging_restart_loop_rest_for_one/1, hanging_restart_loop_simple/1, code_change/1, code_change_map/1, - code_change_simple/1, code_change_simple_map/1]). + code_change_simple/1, code_change_simple_map/1, + order_of_children/1, scale_start_stop_many_children/1]). %%------------------------------------------------------------------------- @@ -91,12 +96,15 @@ all() -> {group, normal_termination}, {group, shutdown_termination}, {group, abnormal_termination}, child_unlink, tree, - count_children, count_restarting_children, get_callback_module, + count_children, count_children_supervisor, count_restarting_children, + get_callback_module, do_not_save_start_parameters_for_temporary_children, do_not_save_child_specs_for_temporary_children, simple_one_for_one_scale_many_temporary_children, temporary_bystander, - simple_global_supervisor, hanging_restart_loop, hanging_restart_loop_simple, - code_change, code_change_map, code_change_simple, code_change_simple_map]. + simple_global_supervisor, hanging_restart_loop, + hanging_restart_loop_rest_for_one, hanging_restart_loop_simple, + code_change, code_change_map, code_change_simple, code_change_simple_map, + order_of_children, scale_start_stop_many_children]. groups() -> [{sup_start, [], @@ -105,7 +113,10 @@ groups() -> sup_start_ignore_temporary_child_start_child, sup_start_ignore_temporary_child_start_child_simple, sup_start_ignore_permanent_child_start_child_simple, - sup_start_error_return, sup_start_fail]}, + sup_start_error_return, sup_start_fail, + sup_start_child_returns_error, sup_start_restart_child_returns_error, + sup_start_child_returns_error_simple + ]}, {sup_start_map, [], [sup_start_map, sup_start_map_simple, sup_start_map_faulty_specs]}, {sup_stop, [], @@ -147,6 +158,15 @@ init_per_testcase(_Case, Config) -> Config. end_per_testcase(_Case, _Config) -> + %% Clean up to avoid unnecessary error reports in the shell + case whereis(sup_test) of + SupPid when is_pid(SupPid) -> + unlink(SupPid), + exit(SupPid,shutdown), + ok; + _ -> + error + end, ok. start_link(InitResult) -> @@ -274,6 +294,7 @@ sup_start_ignore_permanent_child_start_child_simple(Config) %% Regression test: check that the supervisor terminates without error. exit(Pid, shutdown), check_exit_reason(Pid, shutdown). + %%------------------------------------------------------------------------- %% Tests what happens if init-callback returns a invalid value. sup_start_error_return(Config) when is_list(Config) -> @@ -289,6 +310,53 @@ sup_start_fail(Config) when is_list(Config) -> check_exit_reason(Term). %%------------------------------------------------------------------------- +%% Test what happens when the start function for a child returns +%% {error,Reason} or some other term(). +sup_start_restart_child_returns_error(_Config) -> + process_flag(trap_exit, true), + Child = {child1, {supervisor_1, start_child, [error]}, + permanent, 1000, worker, []}, + {ok, _Pid} = start_link({ok, {{one_for_one, 2, 3600}, [Child]}}), + + ok = supervisor:terminate_child(sup_test, child1), + {error,{function_clause,_}} = supervisor:restart_child(sup_test,child1), + + [{child1,undefined,worker,[]}] = supervisor:which_children(sup_test), + ok. + +%%------------------------------------------------------------------------- +%% Test what happens when the start function for a child returns +%% {error,Reason} or some other term(). +sup_start_child_returns_error(_Config) -> + process_flag(trap_exit, true), + Child1 = {child1, {supervisor_1, start_child, [{return,{error,reason}}]}, + permanent, 1000, worker, []}, + Child2 = {child2, {supervisor_1, start_child, [{return,error_reason}]}, + permanent, 1000, worker, []}, + {ok, _Pid} = start_link({ok, {{one_for_one, 2, 3600}, []}}), + + {error,{reason,_}} = supervisor:start_child(sup_test,Child1), + {error,{error_reason,_}} = supervisor:start_child(sup_test,Child2), + + [] = supervisor:which_children(sup_test), + ok. + +%%------------------------------------------------------------------------- +%% Test what happens when the start function for a child returns +%% {error,Reason} - simple_one_for_one +sup_start_child_returns_error_simple(_Config) -> + process_flag(trap_exit, true), + Child = {child1, {supervisor_1, start_child, []}, + permanent, 1000, worker, []}, + {ok, _Pid} = start_link({ok, {{simple_one_for_one, 2, 3600}, [Child]}}), + + {error,reason} = supervisor:start_child(sup_test,[{return,{error,reason}}]), + {error,error_reason} = supervisor:start_child(sup_test,[{return,error_reason}]), + + [] = supervisor:which_children(sup_test), + ok. + +%%------------------------------------------------------------------------- %% Tests that the supervisor process starts correctly with map %% startspec, and that the full childspec can be read. sup_start_map(Config) when is_list(Config) -> @@ -468,7 +536,16 @@ extra_return(Config) when is_list(Config) -> [{child1, CPid3, worker, []}] = supervisor:which_children(sup_test), [1,1,0,1] = get_child_counts(sup_test), - ok. + %% Check that it can be automatically restarted + terminate(CPid3, abnormal), + [{child1, CPid4, worker, []}] = supervisor:which_children(sup_test), + [1,1,0,1] = get_child_counts(sup_test), + if (not is_pid(CPid4)) orelse CPid4=:=CPid3 -> + ct:fail({not_restarted,CPid3,CPid4}); + true -> + ok + end. + %%------------------------------------------------------------------------- %% Test API functions start_child/2, terminate_child/2, delete_child/2 %% restart_child/2, which_children/1, count_children/1. Only correct @@ -1140,7 +1217,7 @@ simple_one_for_one(Config) when is_list(Config) -> [{Id4, Pid4, _, _}|_] = supervisor:which_children(sup_test), terminate(SupPid, Pid4, Id4, abnormal), - check_exit([SupPid]). + check_exit_reason(SupPid,shutdown). %%------------------------------------------------------------------------- @@ -1378,6 +1455,11 @@ tree(Config) when is_list(Config) -> [?MODULE, {ok, {{one_for_one, 4, 3600}, []}}]}, permanent, infinity, supervisor, []}, + ChildSup3 = {supchild3, + {supervisor, start_link, + [?MODULE, {ok, {{one_for_one, 4, 3600}, []}}]}, + transient, infinity, + supervisor, []}, %% Top supervisor {ok, SupPid} = start_link({ok, {{one_for_all, 4, 3600}, []}}), @@ -1385,7 +1467,9 @@ tree(Config) when is_list(Config) -> %% Child supervisors {ok, Sup1} = supervisor:start_child(SupPid, ChildSup1), {ok, Sup2} = supervisor:start_child(SupPid, ChildSup2), - [2,2,2,0] = get_child_counts(SupPid), + {ok, _Sup3} = supervisor:start_child(SupPid, ChildSup3), + ok = supervisor:terminate_child(SupPid, supchild3), + [3,2,3,0] = get_child_counts(SupPid), %% Workers [{_, CPid2, _, _},{_, CPid1, _, _}] = @@ -1417,16 +1501,21 @@ tree(Config) when is_list(Config) -> timer:sleep(1000), - [{supchild2, NewSup2, _, _},{supchild1, NewSup1, _, _}] = + [{supchild3, NewSup3, _, _}, + {supchild2, NewSup2, _, _}, + {supchild1, NewSup1, _, _}] = supervisor:which_children(SupPid), - [2,2,2,0] = get_child_counts(SupPid), + [3,3,3,0] = get_child_counts(SupPid), [{child2, _, _, _},{child1, _, _, _}] = supervisor:which_children(NewSup1), [2,2,0,2] = get_child_counts(NewSup1), [] = supervisor:which_children(NewSup2), - [0,0,0,0] = get_child_counts(NewSup2). + [0,0,0,0] = get_child_counts(NewSup2), + + [] = supervisor:which_children(NewSup3), + [0,0,0,0] = get_child_counts(NewSup3). %%------------------------------------------------------------------------- %% Test count_children @@ -1459,6 +1548,36 @@ count_children(Config) when is_list(Config) -> [1,0,0,0] = get_child_counts(sup_test). %%------------------------------------------------------------------------- +%% Test count_children for simple_one_for_one, when children are supervisors +count_children_supervisor(Config) when is_list(Config) -> + process_flag(trap_exit, true), + Child = {child, {supervisor_1, start_child, []}, temporary, infinity, + supervisor, []}, + {ok, SupPid} = start_link({ok, {{simple_one_for_one, 2, 3600}, [Child]}}), + [supervisor:start_child(sup_test, []) || _Ignore <- lists:seq(1,1000)], + + Children = supervisor:which_children(sup_test), + ChildCount = get_child_counts(sup_test), + + [supervisor:start_child(sup_test, []) || _Ignore2 <- lists:seq(1,1000)], + + ChildCount2 = get_child_counts(sup_test), + Children2 = supervisor:which_children(sup_test), + + ChildCount3 = get_child_counts(sup_test), + Children3 = supervisor:which_children(sup_test), + + 1000 = length(Children), + [1,1000,1000,0] = ChildCount, + 2000 = length(Children2), + [1,2000,2000,0] = ChildCount2, + Children3 = Children2, + ChildCount3 = ChildCount2, + + [terminate(SupPid, Pid, child, kill) || {undefined, Pid, supervisor, _Modules} <- Children3], + [1,0,0,0] = get_child_counts(sup_test). + +%%------------------------------------------------------------------------- %% Test count_children when some children are restarting count_restarting_children(Config) when is_list(Config) -> process_flag(trap_exit, true), @@ -1577,11 +1696,11 @@ dont_save_start_parameters_for_temporary_children(simple_one_for_one = Type) -> start_children(Sup2, [LargeList], 100), start_children(Sup3, [LargeList], 100), - [{memory,Mem1}] = process_info(Sup1, [memory]), - [{memory,Mem2}] = process_info(Sup2, [memory]), - [{memory,Mem3}] = process_info(Sup3, [memory]), + Size1 = erts_debug:flat_size(sys:get_status(Sup1)), + Size2 = erts_debug:flat_size(sys:get_status(Sup2)), + Size3 = erts_debug:flat_size(sys:get_status(Sup3)), - true = (Mem3 < Mem1) and (Mem3 < Mem2), + true = (Size3 < Size1) and (Size3 < Size2), terminate(Sup1, shutdown), terminate(Sup2, shutdown), @@ -1605,11 +1724,11 @@ dont_save_start_parameters_for_temporary_children(Type) -> start_children(Sup2, Transient, 100), start_children(Sup3, Temporary, 100), - [{memory,Mem1}] = process_info(Sup1, [memory]), - [{memory,Mem2}] = process_info(Sup2, [memory]), - [{memory,Mem3}] = process_info(Sup3, [memory]), + Size1 = erts_debug:flat_size(sys:get_status(Sup1)), + Size2 = erts_debug:flat_size(sys:get_status(Sup2)), + Size3 = erts_debug:flat_size(sys:get_status(Sup3)), - true = (Mem3 < Mem1) and (Mem3 < Mem2), + true = (Size3 < Size1) and (Size3 < Size2), terminate(Sup1, shutdown), terminate(Sup2, shutdown), @@ -1847,6 +1966,61 @@ hanging_restart_loop(Config) when is_list(Config) -> undefined = whereis(sup_test), ok. +hanging_restart_loop_rest_for_one(Config) when is_list(Config) -> + process_flag(trap_exit, true), + {ok, Pid} = start_link({ok, {{rest_for_one, 8, 10}, []}}), + Child1 = {child1, {supervisor_1, start_child, []}, + permanent, brutal_kill, worker, []}, + Child2 = {child2, {supervisor_deadlock, start_child, []}, + permanent, brutal_kill, worker, []}, + Child3 = {child3, {supervisor_1, start_child, []}, + permanent, brutal_kill, worker, []}, + + %% Ets table with state read by supervisor_deadlock.erl + ets:new(supervisor_deadlock,[set,named_table,public]), + ets:insert(supervisor_deadlock,{fail_start,false}), + + {ok, CPid1} = supervisor:start_child(sup_test, Child1), + {ok, CPid2} = supervisor:start_child(sup_test, Child2), + link(CPid2), + {ok, _CPid3} = supervisor:start_child(sup_test, Child3), + + ets:insert(supervisor_deadlock,{fail_start,true}), + supervisor_deadlock:restart_child(), + timer:sleep(2000), % allow restart to happen before proceeding + + {error, already_present} = supervisor:start_child(sup_test, Child2), + {error, restarting} = supervisor:restart_child(sup_test, child2), + {error, restarting} = supervisor:delete_child(sup_test, child2), + [{child3,undefined,worker,[]}, + {child2,restarting,worker,[]}, + {child1,CPid1,worker,[]}] = supervisor:which_children(sup_test), + [3,1,0,3] = get_child_counts(sup_test), + + ok = supervisor:terminate_child(sup_test, child2), + check_exit_reason(CPid2, error), + [{child3,undefined,worker,[]}, + {child2,undefined,worker,[]}, + {child1,CPid1,worker,[]}] = supervisor:which_children(sup_test), + + ets:insert(supervisor_deadlock,{fail_start,false}), + {ok, CPid22} = supervisor:restart_child(sup_test, child2), + link(CPid22), + + ets:insert(supervisor_deadlock,{fail_start,true}), + supervisor_deadlock:restart_child(), + timer:sleep(2000), % allow restart to happen before proceeding + + %% Terminating supervisor. + %% OTP-9549 fixes so this does not give a timetrap timeout - + %% i.e. that supervisor does not hang in restart loop. + terminate(Pid,shutdown), + + %% Check that child died with reason from 'restart' request above + check_exit_reason(CPid22, error), + undefined = whereis(sup_test), + ok. + %%------------------------------------------------------------------------- %% Test that child and supervisor can be shutdown while hanging in %% restart loop, simple_one_for_one. @@ -2022,11 +2196,11 @@ code_change_simple(_Config) -> SimpleChild2 = {child2,{supervisor_1, start_child, []}, permanent, brutal_kill, worker, []}, - {error, {error, {ok,[_,_]}}} = + {error, {error, {ok,{[_,_],_}}}} = fake_upgrade(SimplePid,{ok,{SimpleFlags,[SimpleChild1,SimpleChild2]}}), %% Attempt to remove child - {error, {error, {ok,[]}}} = fake_upgrade(SimplePid,{ok,{SimpleFlags,[]}}), + {error, {error, {ok,{[],_}}}} = fake_upgrade(SimplePid,{ok,{SimpleFlags,[]}}), terminate(SimplePid,shutdown), ok. @@ -2047,11 +2221,11 @@ code_change_simple_map(_Config) -> %% Attempt to add child SimpleChild2 = #{id=>child2, start=>{supervisor_1, start_child, []}}, - {error, {error, {ok, [_,_]}}} = + {error, {error, {ok, {[_,_],_}}}} = fake_upgrade(SimplePid,{ok,{SimpleFlags,[SimpleChild1,SimpleChild2]}}), %% Attempt to remove child - {error, {error, {ok, []}}} = + {error, {error, {ok, {[],_}}}} = fake_upgrade(SimplePid,{ok,{SimpleFlags,[]}}), terminate(SimplePid,shutdown), @@ -2075,6 +2249,148 @@ fake_upgrade(Pid,NewInitReturn) -> ok = sys:resume(Pid), R. +%% Test that children are started in the order they are given, and +%% terminated in the opposite order +order_of_children(_Config) -> + process_flag(trap_exit, true), + %% Use child ids that are not alphabetically storted + Id1 = ch7, + Id2 = ch3, + Id3 = ch10, + Id4 = ch2, + Id5 = ch5, + Children = + [{Id, {supervisor_1, start_child, []}, permanent, 1000, worker, []} || + Id <- [Id1,Id2,Id3,Id4,Id5]], + + {ok, SupPid} = start_link({ok, {{rest_for_one, 2, 3600}, Children}}), + + + %% Check start order (pids are growing) + Which1 = supervisor:which_children(sup_test), + IsPid = fun({_,P,_,_}) when is_pid(P) -> true; (_) -> false end, + true = lists:all(IsPid,Which1), + SortedOnPid1 = lists:keysort(2,Which1), + [{Id1,Pid1,_,_}, + {Id2,Pid2,_,_}, + {Id3,Pid3,_,_}, + {Id4,Pid4,_,_}, + {Id5,Pid5,_,_}] = SortedOnPid1, + + TPid = self(), + TraceHandler = fun({trace,P,exit,_},{Last,Ps}) when P=:=Last -> + TPid ! {exited,lists:reverse([P|Ps])}, + {Last,Ps}; + ({trace,P,exit,_},{Last,Ps}) -> + {Last,[P|Ps]}; + (_T,Acc) -> + Acc + end, + + %% Terminate Pid3 and check that Pid4 and Pid5 are terminated in + %% expected order. + Expected1 = [Pid5,Pid4], + {ok,_} = dbg:tracer(process,{TraceHandler,{Pid4,[]}}), + [{ok,[_]} = dbg:p(P,procs) || P <- Expected1], + terminate(Pid3, abnormal), + receive {exited,ExitedPids1} -> + dbg:stop_clear(), + case ExitedPids1 of + Expected1 -> ok; + _ -> ct:fail({faulty_termination_order, + {expected,Expected1}, + {got,ExitedPids1}}) + end + after 3000 -> + dbg:stop_clear(), + ct:fail({shutdown_fail,timeout}) + end, + + %% Then check that Id3-5 are started again in correct order + Which2 = supervisor:which_children(sup_test), + true = lists:all(IsPid,Which2), + SortedOnPid2 = lists:keysort(2,Which2), + [{Id1,Pid1,_,_}, + {Id2,Pid2,_,_}, + {Id3,Pid32,_,_}, + {Id4,Pid42,_,_}, + {Id5,Pid52,_,_}] = SortedOnPid2, + + %% Terminate supervisor and check that all children are terminated + %% in opposite start order + Expected2 = [Pid52,Pid42,Pid32,Pid2,Pid1], + {ok,_} = dbg:tracer(process,{TraceHandler,{Pid1,[]}}), + [{ok,[_]} = dbg:p(P,procs) || P <- Expected2], + exit(SupPid,shutdown), + receive {exited,ExitedPids2} -> + dbg:stop_clear(), + case ExitedPids2 of + Expected2 -> ok; + _ -> ct:fail({faulty_termination_order, + {expected,Expected2}, + {got,ExitedPids2}}) + end + after 3000 -> + dbg:stop_clear(), + ct:fail({shutdown_fail,timeout}) + end, + ok. + +%% Test that a non-simple supervisor scales well for starting and +%% stopping many children. +scale_start_stop_many_children(_Config) -> + process_flag(trap_exit, true), + {ok, _Pid} = start_link({ok, {{one_for_one, 2, 3600}, []}}), + N1 = 1000, + N2 = 100000, + Ids1 = lists:seq(1,N1), + Ids2 = lists:seq(1,N2), + Children1 = [{Id,{supervisor_1,start_child,[]},permanent,1000,worker,[]} || + Id <- Ids1], + Children2 = [{Id,{supervisor_1,start_child,[]},permanent,1000,worker,[]} || + Id <- Ids2], + + {StartT1,_} = + timer:tc(fun() -> + [supervisor:start_child(sup_test,C) || C <- Children1] + end), + {StopT1,_} = + timer:tc(fun() -> + [supervisor:terminate_child(sup_test,I) || I <- Ids1] + end), + ct:log("~w children, start time: ~w ms, stop time: ~w ms", + [N1, StartT1 div 1000, StopT1 div 1000]), + + {StartT2,_} = + timer:tc(fun() -> + [supervisor:start_child(sup_test,C) || C <- Children2] + end), + {StopT2,_} = + timer:tc(fun() -> + [supervisor:terminate_child(sup_test,I) || I <- Ids2] + end), + ct:log("~w children, start time: ~w ms, stop time: ~w ms", + [N2, StartT2 div 1000, StopT2 div 1000]), + + %% Scaling should be more or less linear, but allowing a bit more + %% to avoid false alarms + ScaleLimit = (N2 div N1) * 10, + StartScale = StartT2 div StartT1, + StopScale = StopT2 div StopT1, + + ct:log("Scale limit: ~w~nStart scale: ~w~nStop scale: ~w", + [ScaleLimit, StartScale, StopScale]), + + if StartScale > ScaleLimit -> + ct:fail({bad_start_scale,StartScale}); + StopScale > ScaleLimit -> + ct:fail({bad_stop_scale,StopScale}); + true -> + ok + end, + + ok. + %%------------------------------------------------------------------------- terminate(Pid, Reason) when Reason =/= supervisor -> terminate(dummy, Pid, dummy, Reason). diff --git a/lib/stdlib/test/supervisor_deadlock.erl b/lib/stdlib/test/supervisor_deadlock.erl index 8d3d1c6f30..f51aecccb2 100644 --- a/lib/stdlib/test/supervisor_deadlock.erl +++ b/lib/stdlib/test/supervisor_deadlock.erl @@ -1,5 +1,5 @@ -module(supervisor_deadlock). --compile(export_all). +-compile([export_all,nowarn_export_all]). %%%----------------------------------------------------------------- diff --git a/lib/stdlib/test/unicode_util_SUITE.erl b/lib/stdlib/test/unicode_util_SUITE.erl index 7dba0a2fd0..632d9ae6e6 100644 --- a/lib/stdlib/test/unicode_util_SUITE.erl +++ b/lib/stdlib/test/unicode_util_SUITE.erl @@ -312,12 +312,23 @@ get(_) -> add_get_tests. count(Config) -> + Parent = self(), + Exec = fun() -> + do_measure(Config), + Parent ! {test_done, self()} + end, ct:timetrap({minutes,5}), case ct:get_timetrap_info() of - {_,{_,Scale}} -> + {_,{_,Scale}} when Scale > 1 -> {skip,{measurments_skipped_debug,Scale}}; - _ -> % No scaling - do_measure(Config) + _ -> % No scaling, run at most 2 min + Tester = spawn(Exec), + receive {test_done, Tester} -> ok + after 120000 -> + io:format("Timelimit reached stopping~n",[]), + exit(Tester, die) + end, + ok end. do_measure(Config) -> diff --git a/lib/stdlib/test/uri_string_SUITE.erl b/lib/stdlib/test/uri_string_SUITE.erl new file mode 100644 index 0000000000..fef356355c --- /dev/null +++ b/lib/stdlib/test/uri_string_SUITE.erl @@ -0,0 +1,926 @@ +%% +%% %CopyrightBegin% +%% +%% Copyright Ericsson AB 2008-2017. All Rights Reserved. +%% +%% Licensed under the Apache License, Version 2.0 (the "License"); +%% you may not use this file except in compliance with the License. +%% You may obtain a copy of the License at +%% +%% http://www.apache.org/licenses/LICENSE-2.0 +%% +%% Unless required by applicable law or agreed to in writing, software +%% distributed under the License is distributed on an "AS IS" BASIS, +%% WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +%% See the License for the specific language governing permissions and +%% limitations under the License. +%% +%% %CopyrightEnd% +%% +-module(uri_string_SUITE). + +-include_lib("common_test/include/ct.hrl"). + +-export([all/0, suite/0,groups/0, + normalize/1, + parse_binary_fragment/1, parse_binary_host/1, parse_binary_host_ipv4/1, + parse_binary_host_ipv6/1, + parse_binary_path/1, parse_binary_pct_encoded_fragment/1, parse_binary_pct_encoded_query/1, + parse_binary_pct_encoded_userinfo/1, parse_binary_port/1, + parse_binary_query/1, parse_binary_scheme/1, parse_binary_userinfo/1, + parse_fragment/1, parse_host/1, parse_host_ipv4/1, parse_host_ipv6/1, + parse_path/1, parse_pct_encoded_fragment/1, parse_pct_encoded_query/1, + parse_pct_encoded_userinfo/1, parse_port/1, + parse_query/1, parse_scheme/1, parse_userinfo/1, + parse_list/1, parse_binary/1, parse_mixed/1, parse_relative/1, + parse_special/1, parse_special2/1, parse_negative/1, + recompose_fragment/1, recompose_parse_fragment/1, + recompose_query/1, recompose_parse_query/1, + recompose_path/1, recompose_parse_path/1, + recompose_autogen/1, parse_recompose_autogen/1, + transcode_basic/1, transcode_options/1, transcode_mixed/1, transcode_negative/1, + compose_query/1, compose_query_latin1/1, compose_query_negative/1, + dissect_query/1, dissect_query_negative/1, + interop_query_latin1/1, interop_query_utf8/1 + ]). + + +-define(SCHEME, "foo"). +-define(USERINFO, "åsa"). +-define(USERINFO_ENC, "%C3%A5sa"). +-define(HOST, "älvsjö"). +-define(HOST_ENC, "%C3%A4lvsj%C3%B6"). +-define(IPV6, "::127.0.0.1"). +-define(IPV6_ENC, "[::127.0.0.1]"). +-define(PORT, 8042). +-define(PORT_ENC, ":8042"). +-define(PATH, "/där"). +-define(PATH_ENC, "/d%C3%A4r"). +-define(QUERY, "name=örn"). +-define(QUERY_ENC, "?name=%C3%B6rn"). +-define(FRAGMENT, "näsa"). +-define(FRAGMENT_ENC, "#n%C3%A4sa"). + + +suite() -> + [{timetrap,{minutes,1}}]. + +all() -> + [ + normalize, + parse_binary_scheme, + parse_binary_userinfo, + parse_binary_pct_encoded_userinfo, + parse_binary_host, + parse_binary_host_ipv4, + parse_binary_host_ipv6, + parse_binary_port, + parse_binary_path, + parse_binary_query, + parse_binary_pct_encoded_query, + parse_binary_fragment, + parse_binary_pct_encoded_fragment, + parse_scheme, + parse_userinfo, + parse_pct_encoded_userinfo, + parse_host, + parse_host_ipv4, + parse_host_ipv6, + parse_port, + parse_path, + parse_query, + parse_pct_encoded_query, + parse_fragment, + parse_pct_encoded_fragment, + parse_list, + parse_binary, + parse_mixed, + parse_relative, + parse_special, + parse_special2, + parse_negative, + recompose_fragment, + recompose_parse_fragment, + recompose_query, + recompose_parse_query, + recompose_path, + recompose_parse_path, + recompose_autogen, + parse_recompose_autogen, + transcode_basic, + transcode_options, + transcode_mixed, + transcode_negative, + compose_query, + compose_query_latin1, + compose_query_negative, + dissect_query, + dissect_query_negative, + interop_query_latin1, + interop_query_utf8 + ]. + +groups() -> + []. + + +%%------------------------------------------------------------------------- +%% Helper functions +%%------------------------------------------------------------------------- +uri_combinations() -> + [[Sch,Usr,Hst,Prt,Pat,Qry,Frg] || + Sch <- [fun update_scheme/1, fun update_scheme_binary/1, none], + Usr <- [fun update_userinfo/1, fun update_userinfo_binary/1, none], + Hst <- [fun update_host/1, fun update_host_binary/1, + fun update_ipv6/1, fun update_ipv6_binary/1, none], + Prt <- [fun update_port/1, none], + Pat <- [fun update_path/1, fun update_path_binary/1], + Qry <- [fun update_query/1,fun update_query_binary/1, none], + Frg <- [fun update_fragment/1, fun update_fragment_binary/1, none], + not (Usr =:= none andalso Hst =:= none andalso Prt =/= none), + not (Usr =/= none andalso Hst =:= none andalso Prt =:= none), + not (Usr =/= none andalso Hst =:= none andalso Prt =/= none)]. + + +generate_test_vector(Comb) -> + Fun = fun (F, {Map, URI}) when is_function(F) -> F({Map, URI}); + (_, Map) -> Map + end, + lists:foldl(Fun, {#{}, empty}, Comb). + +generate_test_vectors(L) -> + lists:map(fun generate_test_vector/1, L). + +update_fragment({In, empty}) -> + {In#{fragment => ?FRAGMENT}, ?FRAGMENT_ENC}; +update_fragment({In, Out}) when is_list(Out) -> + {In#{fragment => ?FRAGMENT}, Out ++ ?FRAGMENT_ENC}; +update_fragment({In, Out}) when is_binary(Out) -> + {In#{fragment => ?FRAGMENT}, binary_to_list(Out) ++ ?FRAGMENT_ENC}. + +update_fragment_binary({In, empty}) -> + {In#{fragment => <<?FRAGMENT/utf8>>}, <<?FRAGMENT_ENC>>}; +update_fragment_binary({In, Out}) when is_list(Out) -> + {In#{fragment => <<?FRAGMENT/utf8>>}, Out ++ ?FRAGMENT_ENC}; +update_fragment_binary({In, Out}) when is_binary(Out) -> + {In#{fragment => <<?FRAGMENT/utf8>>}, <<Out/binary,?FRAGMENT_ENC>>}. + + +update_query({In, empty}) -> + {In#{query => ?QUERY}, ?QUERY_ENC}; +update_query({In, Out}) when is_list(Out) -> + {In#{query => ?QUERY}, Out ++ ?QUERY_ENC}; +update_query({In, Out}) when is_binary(Out) -> + {In#{query => ?QUERY}, binary_to_list(Out) ++ ?QUERY_ENC}. + +update_query_binary({In, empty}) -> + {In#{query => <<?QUERY/utf8>>}, <<?QUERY_ENC>>}; +update_query_binary({In, Out}) when is_list(Out) -> + {In#{query => <<?QUERY/utf8>>}, Out ++ ?QUERY_ENC}; +update_query_binary({In, Out}) when is_binary(Out) -> + {In#{query => <<?QUERY/utf8>>}, <<Out/binary,?QUERY_ENC>>}. + +update_path({In, empty}) -> + {In#{path => ?PATH}, ?PATH_ENC}; +update_path({In, Out}) when is_list(Out) -> + {In#{path => ?PATH}, Out ++ ?PATH_ENC}; +update_path({In, Out}) when is_binary(Out) -> + {In#{path => ?PATH}, binary_to_list(Out) ++ ?PATH_ENC}. + +update_path_binary({In, empty}) -> + {In#{path => <<?PATH/utf8>>}, <<?PATH_ENC>>}; +update_path_binary({In, Out}) when is_list(Out) -> + {In#{path => <<?PATH/utf8>>}, Out ++ ?PATH_ENC}; +update_path_binary({In, Out}) when is_binary(Out) -> + {In#{path => <<?PATH/utf8>>}, <<Out/binary,?PATH_ENC>>}. + +update_port({In, Out}) when is_list(Out) -> + {In#{port => ?PORT}, Out ++ ?PORT_ENC}; +update_port({In, Out}) when is_binary(Out) -> + {In#{port => ?PORT}, <<Out/binary,?PORT_ENC>>}. + +update_host({In, empty}) -> + {In#{host => ?HOST}, "//" ++ ?HOST_ENC}; +update_host({In, Out}) when is_list(Out) -> + case maps:is_key(userinfo, In) of + true -> {In#{host => ?HOST}, Out ++ [$@|?HOST_ENC]}; + false -> {In#{host => ?HOST}, Out ++ [$/,$/|?HOST_ENC]} + end; +update_host({In, Out}) when is_binary(Out) -> + case maps:is_key(userinfo, In) of + true -> {In#{host => ?HOST}, binary_to_list(Out) ++ [$@|?HOST_ENC]}; + false -> {In#{host => ?HOST}, binary_to_list(Out) ++ [$/,$/|?HOST_ENC]} + end. + +update_host_binary({In, empty}) -> + {In#{host => <<?HOST/utf8>>}, <<"//",?HOST_ENC>>}; +update_host_binary({In, Out}) when is_list(Out) -> + case maps:is_key(userinfo, In) of + true -> {In#{host => <<?HOST/utf8>>}, Out ++ [$@|?HOST_ENC]}; + false -> {In#{host => <<?HOST/utf8>>}, Out ++ [$/,$/|?HOST_ENC]} + end; +update_host_binary({In, Out}) when is_binary(Out) -> + case maps:is_key(userinfo, In) of + true -> {In#{host => <<?HOST/utf8>>}, <<Out/binary,$@,?HOST_ENC>>}; + false-> {In#{host => <<?HOST/utf8>>}, <<Out/binary,"//",?HOST_ENC>>} + end. + +update_ipv6({In, empty}) -> + {In#{host => ?IPV6}, "//" ++ ?IPV6_ENC}; +update_ipv6({In, Out}) when is_list(Out) -> + case maps:is_key(userinfo, In) of + true -> {In#{host => ?IPV6}, Out ++ [$@|?IPV6_ENC]}; + false -> {In#{host => ?IPV6}, Out ++ [$/,$/|?IPV6_ENC]} + end; +update_ipv6({In, Out}) when is_binary(Out) -> + case maps:is_key(userinfo, In) of + true -> {In#{host => ?IPV6}, binary_to_list(Out) ++ [$@|?IPV6_ENC]}; + false -> {In#{host => ?IPV6}, binary_to_list(Out) ++ [$/,$/|?IPV6_ENC]} + end. + +update_ipv6_binary({In, empty}) -> + {In#{host => <<?IPV6/utf8>>}, <<"//",?IPV6_ENC>>}; +update_ipv6_binary({In, Out}) when is_list(Out) -> + case maps:is_key(userinfo, In) of + true -> {In#{host => <<?IPV6/utf8>>}, Out ++ [$@|?IPV6_ENC]}; + false -> {In#{host => <<?IPV6/utf8>>}, Out ++ [$/,$/|?IPV6_ENC]} + end; +update_ipv6_binary({In, Out}) when is_binary(Out) -> + case maps:is_key(userinfo, In) of + true -> {In#{host => <<?IPV6/utf8>>}, <<Out/binary,$@,?IPV6_ENC>>}; + false-> {In#{host => <<?IPV6/utf8>>}, <<Out/binary,"//",?IPV6_ENC>>} + end. + +update_userinfo({In, empty}) -> + {In#{userinfo => ?USERINFO}, "//" ++ ?USERINFO_ENC}; +update_userinfo({In, Out}) when is_list(Out) -> + {In#{userinfo => ?USERINFO}, Out ++ "//" ++ ?USERINFO_ENC}; +update_userinfo({In, Out}) when is_binary(Out) -> + {In#{userinfo => ?USERINFO}, binary_to_list(Out) ++ "//" ++ ?USERINFO_ENC}. + +update_userinfo_binary({In, empty}) -> + {In#{userinfo => <<?USERINFO/utf8>>}, <<"//",?USERINFO_ENC>>}; +update_userinfo_binary({In, Out}) when is_list(Out) -> + {In#{userinfo => <<?USERINFO/utf8>>}, Out ++ "//" ++ ?USERINFO_ENC}; +update_userinfo_binary({In, Out}) when is_binary(Out) -> + {In#{userinfo => <<?USERINFO/utf8>>}, <<Out/binary,"//",?USERINFO_ENC>>}. + +update_scheme({In, empty}) -> + {In#{scheme => ?SCHEME}, ?SCHEME ++ ":"}. + +update_scheme_binary({In, empty}) -> + {In#{scheme => <<?SCHEME/utf8>>}, <<?SCHEME,$:>>}. + + +%% Test recompose on a generated test vector +run_test_recompose({#{}, empty}) -> + try "" = uri_string:recompose(#{}) of + _ -> ok + catch + _:_ -> error({test_failed, #{}, ""}) + end; +run_test_recompose({Map, URI}) -> + try URI = uri_string:recompose(Map) of + URI -> ok + catch + _:_ -> error({test_failed, Map, URI}) + end. + +%% Test parse - recompose on a generated test vector +run_test_parse_recompose({#{}, empty}) -> + try "" = uri_string:recompose(uri_string:parse("")) of + _ -> ok + catch + _:_ -> error({test_failed, #{}, ""}) + end; +run_test_parse_recompose({Map, URI}) -> + try URI = uri_string:recompose(uri_string:parse(URI)) of + URI -> ok + catch + _:_ -> error({test_failed, Map, URI}) + end. + + +%%------------------------------------------------------------------------- +%% Parse tests +%%------------------------------------------------------------------------- + +parse_binary_scheme(_Config) -> + #{} = uri_string:parse(<<>>), + #{path := <<"foo">>} = uri_string:parse(<<"foo">>), + #{scheme := <<"foo">>} = uri_string:parse(<<"foo:">>), + #{scheme := <<"foo">>, path := <<"bar:nisse">>} = uri_string:parse(<<"foo:bar:nisse">>), + #{scheme := <<"foo">>, host := <<"">>} = uri_string:parse(<<"foo://">>), + #{scheme := <<"foo">>, host := <<"">>, path := <<"/">>} = uri_string:parse(<<"foo:///">>), + #{scheme := <<"foo">>, host := <<"">>, path := <<"//">>} = uri_string:parse(<<"foo:////">>), + + #{path := <<"/">>} = uri_string:parse(<<"/">>), + #{host := <<>>} = uri_string:parse(<<"//">>), + #{host := <<>>, path := <<"/">>} = uri_string:parse(<<"///">>). + +parse_binary_userinfo(_Config) -> + #{scheme := <<"user">>, path := <<"password@localhost">>} = + uri_string:parse(<<"user:password@localhost">>), + #{path := <<"user@">>} = uri_string:parse(<<"user@">>), + #{path := <<"/user@">>} = uri_string:parse(<<"/user@">>), + #{path := <<"user@localhost">>} = uri_string:parse(<<"user@localhost">>), + #{userinfo := <<"user">>, host := <<"localhost">>} = uri_string:parse(<<"//user@localhost">>), + #{userinfo := <<"user:password">>, host := <<"localhost">>} = + uri_string:parse(<<"//user:password@localhost">>), + #{scheme := <<"foo">>, path := <<"/user@">>} = + uri_string:parse(<<"foo:/user@">>), + #{scheme := <<"foo">>, userinfo := <<"user">>, host := <<"localhost">>} = + uri_string:parse(<<"foo://user@localhost">>), + #{scheme := <<"foo">>, userinfo := <<"user:password">>, host := <<"localhost">>} = + uri_string:parse(<<"foo://user:password@localhost">>). + +parse_binary_pct_encoded_userinfo(_Config) -> + #{scheme := <<"user">>, path := <<"合@気道"/utf8>>} = + uri_string:parse(<<"user:%E5%90%88@%E6%B0%97%E9%81%93">>), + #{path := <<"合気道@"/utf8>>} = uri_string:parse(<<"%E5%90%88%E6%B0%97%E9%81%93@">>), + #{path := <<"/合気道@"/utf8>>} = uri_string:parse(<<"/%E5%90%88%E6%B0%97%E9%81%93@">>), + #{path := <<"合@気道"/utf8>>} = uri_string:parse(<<"%E5%90%88@%E6%B0%97%E9%81%93">>), + #{userinfo := <<"合"/utf8>>, host := <<"気道"/utf8>>} = + uri_string:parse(<<"//%E5%90%88@%E6%B0%97%E9%81%93">>), + #{userinfo := <<"合:気"/utf8>>, host := <<"道"/utf8>>} = + uri_string:parse(<<"//%E5%90%88:%E6%B0%97@%E9%81%93">>), + #{scheme := <<"foo">>, path := <<"/合気道@"/utf8>>} = + uri_string:parse(<<"foo:/%E5%90%88%E6%B0%97%E9%81%93@">>), + #{scheme := <<"foo">>, userinfo := <<"合"/utf8>>, host := <<"気道"/utf8>>} = + uri_string:parse(<<"foo://%E5%90%88@%E6%B0%97%E9%81%93">>), + #{scheme := <<"foo">>, userinfo := <<"合:気"/utf8>>, host := <<"道"/utf8>>} = + uri_string:parse(<<"foo://%E5%90%88:%E6%B0%97@%E9%81%93">>), + {error,invalid_uri,"@"} = uri_string:parse(<<"//%E5%90%88@%E6%B0%97%E9%81%93@">>), + {error,invalid_uri,":"} = uri_string:parse(<<"foo://%E5%90%88@%E6%B0%97%E9%81%93@">>). + +parse_binary_host(_Config) -> + #{host := <<"hostname">>} = uri_string:parse(<<"//hostname">>), + #{host := <<"hostname">>,scheme := <<"foo">>} = uri_string:parse(<<"foo://hostname">>), + #{host := <<"hostname">>,scheme := <<"foo">>, userinfo := <<"user">>} = + uri_string:parse(<<"foo://user@hostname">>). + +parse_binary_host_ipv4(_Config) -> + #{host := <<"127.0.0.1">>} = uri_string:parse(<<"//127.0.0.1">>), + #{host := <<"127.0.0.1">>, path := <<"/over/there">>} = + uri_string:parse(<<"//127.0.0.1/over/there">>), + #{host := <<"127.0.0.1">>, query := <<"name=ferret">>} = + uri_string:parse(<<"//127.0.0.1?name=ferret">>), + #{host := <<"127.0.0.1">>, fragment := <<"nose">>} = uri_string:parse(<<"//127.0.0.1#nose">>), + {error,invalid_uri,"x"} = uri_string:parse(<<"//127.0.0.x">>), + {error,invalid_uri,"1227.0.0.1"} = uri_string:parse(<<"//1227.0.0.1">>). + +parse_binary_host_ipv6(_Config) -> + #{host := <<"::127.0.0.1">>} = uri_string:parse(<<"//[::127.0.0.1]">>), + #{host := <<"2001:0db8:0000:0000:0000:0000:1428:07ab">>} = + uri_string:parse(<<"//[2001:0db8:0000:0000:0000:0000:1428:07ab]">>), + #{host := <<"::127.0.0.1">>, path := <<"/over/there">>} = + uri_string:parse(<<"//[::127.0.0.1]/over/there">>), + #{host := <<"::127.0.0.1">>, query := <<"name=ferret">>} = + uri_string:parse(<<"//[::127.0.0.1]?name=ferret">>), + #{host := <<"::127.0.0.1">>, fragment := <<"nose">>} = + uri_string:parse(<<"//[::127.0.0.1]#nose">>), + {error,invalid_uri,"x"} = uri_string:parse(<<"//[::127.0.0.x]">>), + {error,invalid_uri,"::1227.0.0.1"} = uri_string:parse(<<"//[::1227.0.0.1]">>), + {error,invalid_uri,"G"} = uri_string:parse(<<"//[2001:0db8:0000:0000:0000:0000:1428:G7ab]">>). + +parse_binary_port(_Config) -> + #{path:= <<"/:8042">>} = + uri_string:parse(<<"/:8042">>), + #{host:= <<>>, port := 8042} = + uri_string:parse(<<"//:8042">>), + #{host := <<"example.com">>, port:= 8042} = + uri_string:parse(<<"//example.com:8042">>), + #{scheme := <<"foo">>, path := <<"/:8042">>} = + uri_string:parse(<<"foo:/:8042">>), + #{scheme := <<"foo">>, host := <<>>, port := 8042} = + uri_string:parse(<<"foo://:8042">>), + #{scheme := <<"foo">>, host := <<"example.com">>, port := 8042} = + uri_string:parse(<<"foo://example.com:8042">>), + {error,invalid_uri,":"} = uri_string:parse(":600"), + {error,invalid_uri,"x"} = uri_string:parse("//:8042x"). + +parse_binary_path(_Config) -> + #{path := <<"over/there">>} = uri_string:parse(<<"over/there">>), + #{path := <<"/over/there">>} = uri_string:parse(<<"/over/there">>), + #{scheme := <<"foo">>, path := <<"/over/there">>} = + uri_string:parse(<<"foo:/over/there">>), + #{scheme := <<"foo">>, host := <<"example.com">>, path := <<"/over/there">>} = + uri_string:parse(<<"foo://example.com/over/there">>), + #{scheme := <<"foo">>, host := <<"example.com">>, path := <<"/over/there">>, port := 8042} = + uri_string:parse(<<"foo://example.com:8042/over/there">>). + +parse_binary_query(_Config) -> + #{scheme := <<"foo">>, query := <<"name=ferret">>} = + uri_string:parse(<<"foo:?name=ferret">>), + #{scheme := <<"foo">>, path:= <<"over/there">>, query := <<"name=ferret">>} = + uri_string:parse(<<"foo:over/there?name=ferret">>), + #{scheme := <<"foo">>, path:= <<"/over/there">>, query := <<"name=ferret">>} = + uri_string:parse(<<"foo:/over/there?name=ferret">>), + #{scheme := <<"foo">>, host := <<"example.com">>, query := <<"name=ferret">>} = + uri_string:parse(<<"foo://example.com?name=ferret">>), + #{scheme := <<"foo">>, host := <<"example.com">>, path := <<"/">>, query := <<"name=ferret">>} = + uri_string:parse(<<"foo://example.com/?name=ferret">>), + + #{path := <<>>, query := <<"name=ferret">>} = + uri_string:parse(<<"?name=ferret">>), + #{path := <<"over/there">>, query := <<"name=ferret">>} = + uri_string:parse(<<"over/there?name=ferret">>), + #{path := <<"/">>, query := <<"name=ferret">>} = + uri_string:parse(<<"/?name=ferret">>), + #{path := <<"/over/there">>, query := <<"name=ferret">>} = + uri_string:parse(<<"/over/there?name=ferret">>), + #{host := <<"example.com">>, query := <<"name=ferret">>} = + uri_string:parse(<<"//example.com?name=ferret">>), + #{host := <<"example.com">>, path := <<"/">>, query := <<"name=ferret">>} = + uri_string:parse(<<"//example.com/?name=ferret">>). + +parse_binary_pct_encoded_query(_Config) -> + #{scheme := <<"foo">>, host := <<"example.com">>, path := <<"/">>, + query := <<"name=合気道"/utf8>>} = + uri_string:parse(<<"foo://example.com/?name=%E5%90%88%E6%B0%97%E9%81%93">>), + #{host := <<"example.com">>, path := <<"/">>, query := <<"name=合気道"/utf8>>} = + uri_string:parse(<<"//example.com/?name=%E5%90%88%E6%B0%97%E9%81%93">>). + +parse_binary_fragment(_Config) -> + #{scheme := <<"foo">>, fragment := <<"nose">>} = + uri_string:parse(<<"foo:#nose">>), + #{scheme := <<"foo">>, path:= <<"over/there">>, fragment := <<"nose">>} = + uri_string:parse(<<"foo:over/there#nose">>), + #{scheme := <<"foo">>, path:= <<"/over/there">>, fragment := <<"nose">>} = + uri_string:parse(<<"foo:/over/there#nose">>), + #{scheme := <<"foo">>, host := <<"example.com">>, fragment := <<"nose">>} = + uri_string:parse(<<"foo://example.com#nose">>), + #{scheme := <<"foo">>, host := <<"example.com">>, path := <<"/">>, fragment := <<"nose">>} = + uri_string:parse(<<"foo://example.com/#nose">>), + #{scheme := <<"foo">>, host := <<"example.com">>, fragment := <<"nose">>} = + uri_string:parse(<<"foo://example.com#nose">>), + + #{fragment := <<"nose">>} = + uri_string:parse(<<"#nose">>), + #{path := <<"over/there">>, fragment := <<"nose">>} = + uri_string:parse(<<"over/there#nose">>), + #{path := <<"/">>, fragment := <<"nose">>} = + uri_string:parse(<<"/#nose">>), + #{path := <<"/over/there">>, fragment := <<"nose">>} = + uri_string:parse(<<"/over/there#nose">>), + #{host := <<"example.com">>, fragment := <<"nose">>} = + uri_string:parse(<<"//example.com#nose">>), + #{host := <<"example.com">>, path := <<"/">>, fragment := <<"nose">>} = + uri_string:parse(<<"//example.com/#nose">>). + +parse_binary_pct_encoded_fragment(_Config) -> + #{scheme := <<"foo">>, host := <<"example.com">>, fragment := <<"合気道"/utf8>>} = + uri_string:parse(<<"foo://example.com#%E5%90%88%E6%B0%97%E9%81%93">>), + #{host := <<"example.com">>, path := <<"/">>, fragment := <<"合気道"/utf8>>} = + uri_string:parse(<<"//example.com/#%E5%90%88%E6%B0%97%E9%81%93">>). + +parse_scheme(_Config) -> + #{} = uri_string:parse(""), + #{path := "foo"} = uri_string:parse("foo"), + #{scheme := "foo"} = uri_string:parse("foo:"), + #{scheme := "foo", path := "bar:nisse"} = uri_string:parse("foo:bar:nisse"), + #{scheme := "foo", host := ""} = uri_string:parse("foo://"), + #{scheme := "foo", host := "", path := "/"} = uri_string:parse("foo:///"), + #{scheme := "foo", host := "", path := "//"} = uri_string:parse("foo:////"), + + #{path := "/"} = uri_string:parse("/"), + #{host := ""} = uri_string:parse("//"), + #{host := "", path := "/"} = uri_string:parse("///"). + +parse_userinfo(_Config) -> + #{scheme := "user", path := "password@localhost"} = uri_string:parse("user:password@localhost"), + #{path := "user@"} = uri_string:parse("user@"), + #{path := "/user@"} = uri_string:parse("/user@"), + #{path := "user@localhost"} = uri_string:parse("user@localhost"), + #{userinfo := "user", host := "localhost"} = uri_string:parse("//user@localhost"), + #{userinfo := "user:password", host := "localhost"} = + uri_string:parse("//user:password@localhost"), + #{scheme := "foo", path := "/user@"} = + uri_string:parse("foo:/user@"), + #{scheme := "foo", userinfo := "user", host := "localhost"} = + uri_string:parse("foo://user@localhost"), + #{scheme := "foo", userinfo := "user:password", host := "localhost"} = + uri_string:parse("foo://user:password@localhost"). + +parse_pct_encoded_userinfo(_Config) -> + #{scheme := "user", path := "合@気道"} = + uri_string:parse("user:%E5%90%88@%E6%B0%97%E9%81%93"), + #{path := "合気道@"} = uri_string:parse("%E5%90%88%E6%B0%97%E9%81%93@"), + #{path := "/合気道@"} = uri_string:parse("/%E5%90%88%E6%B0%97%E9%81%93@"), + #{path := "合@気道"} = uri_string:parse("%E5%90%88@%E6%B0%97%E9%81%93"), + #{userinfo := "合", host := "気道"} = + uri_string:parse("//%E5%90%88@%E6%B0%97%E9%81%93"), + #{userinfo := "合:気", host := "道"} = + uri_string:parse("//%E5%90%88:%E6%B0%97@%E9%81%93"), + #{scheme := "foo", path := "/合気道@"} = + uri_string:parse("foo:/%E5%90%88%E6%B0%97%E9%81%93@"), + #{scheme := "foo", userinfo := "合", host := "気道"} = + uri_string:parse("foo://%E5%90%88@%E6%B0%97%E9%81%93"), + #{scheme := "foo", userinfo := "合:気", host := "道"} = + uri_string:parse("foo://%E5%90%88:%E6%B0%97@%E9%81%93"), + {error,invalid_uri,"@"} = uri_string:parse("//%E5%90%88@%E6%B0%97%E9%81%93@"), + {error,invalid_uri,":"} = uri_string:parse("foo://%E5%90%88@%E6%B0%97%E9%81%93@"). + + +parse_host(_Config) -> + #{host := "hostname"} = uri_string:parse("//hostname"), + #{host := "hostname",scheme := "foo"} = uri_string:parse("foo://hostname"), + #{host := "hostname",scheme := "foo", userinfo := "user"} = + uri_string:parse("foo://user@hostname"). + +parse_host_ipv4(_Config) -> + #{host := "127.0.0.1"} = uri_string:parse("//127.0.0.1"), + #{host := "2001:0db8:0000:0000:0000:0000:1428:07ab"} = + uri_string:parse("//[2001:0db8:0000:0000:0000:0000:1428:07ab]"), + #{host := "127.0.0.1", path := "/over/there"} = uri_string:parse("//127.0.0.1/over/there"), + #{host := "127.0.0.1", query := "name=ferret"} = uri_string:parse("//127.0.0.1?name=ferret"), + #{host := "127.0.0.1", fragment := "nose"} = uri_string:parse("//127.0.0.1#nose"), + {error,invalid_uri,"x"} = uri_string:parse("//127.0.0.x"), + {error,invalid_uri,"1227.0.0.1"} = uri_string:parse("//1227.0.0.1"). + +parse_host_ipv6(_Config) -> + #{host := "::127.0.0.1"} = uri_string:parse("//[::127.0.0.1]"), + #{host := "::127.0.0.1", path := "/over/there"} = uri_string:parse("//[::127.0.0.1]/over/there"), + #{host := "::127.0.0.1", query := "name=ferret"} = + uri_string:parse("//[::127.0.0.1]?name=ferret"), + #{host := "::127.0.0.1", fragment := "nose"} = uri_string:parse("//[::127.0.0.1]#nose"), + {error,invalid_uri,"x"} = uri_string:parse("//[::127.0.0.x]"), + {error,invalid_uri,"::1227.0.0.1"} = uri_string:parse("//[::1227.0.0.1]"), + {error,invalid_uri,"G"} = uri_string:parse("//[2001:0db8:0000:0000:0000:0000:1428:G7ab]"). + +parse_port(_Config) -> + #{path:= "/:8042"} = + uri_string:parse("/:8042"), + #{host:= "", port := 8042} = + uri_string:parse("//:8042"), + #{host := "example.com", port:= 8042} = + uri_string:parse("//example.com:8042"), + #{scheme := "foo", path := "/:8042"} = + uri_string:parse("foo:/:8042"), + #{scheme := "foo", host := "", port := 8042} = + uri_string:parse("foo://:8042"), + #{scheme := "foo", host := "example.com", port := 8042} = + uri_string:parse("foo://example.com:8042"). + +parse_path(_Config) -> + #{path := "over/there"} = uri_string:parse("over/there"), + #{path := "/over/there"} = uri_string:parse("/over/there"), + #{scheme := "foo", path := "/over/there"} = + uri_string:parse("foo:/over/there"), + #{scheme := "foo", host := "example.com", path := "/over/there"} = + uri_string:parse("foo://example.com/over/there"), + #{scheme := "foo", host := "example.com", path := "/over/there", port := 8042} = + uri_string:parse("foo://example.com:8042/over/there"). + +parse_query(_Config) -> + #{scheme := "foo", query := "name=ferret"} = + uri_string:parse("foo:?name=ferret"), + #{scheme := "foo", path:= "over/there", query := "name=ferret"} = + uri_string:parse("foo:over/there?name=ferret"), + #{scheme := "foo", path:= "/over/there", query := "name=ferret"} = + uri_string:parse("foo:/over/there?name=ferret"), + #{scheme := "foo", host := "example.com", query := "name=ferret"} = + uri_string:parse("foo://example.com?name=ferret"), + #{scheme := "foo", host := "example.com", path := "/", query := "name=ferret"} = + uri_string:parse("foo://example.com/?name=ferret"), + + #{path := "", query := "name=ferret"} = + uri_string:parse("?name=ferret"), + #{path := "over/there", query := "name=ferret"} = + uri_string:parse("over/there?name=ferret"), + #{path := "/", query := "name=ferret"} = + uri_string:parse("/?name=ferret"), + #{path := "/over/there", query := "name=ferret"} = + uri_string:parse("/over/there?name=ferret"), + #{host := "example.com", query := "name=ferret"} = + uri_string:parse("//example.com?name=ferret"), + #{host := "example.com", path := "/", query := "name=ferret"} = + uri_string:parse("//example.com/?name=ferret"). + +parse_pct_encoded_query(_Config) -> + #{scheme := "foo", host := "example.com", path := "/", + query := "name=合気道"} = + uri_string:parse("foo://example.com/?name=%E5%90%88%E6%B0%97%E9%81%93"), + #{host := "example.com", path := "/", query := "name=合気道"} = + uri_string:parse("//example.com/?name=%E5%90%88%E6%B0%97%E9%81%93"). + +parse_fragment(_Config) -> + #{scheme := "foo", fragment := "nose"} = + uri_string:parse("foo:#nose"), + #{scheme := "foo", path:= "over/there", fragment := "nose"} = + uri_string:parse("foo:over/there#nose"), + #{scheme := "foo", path:= "/over/there", fragment := "nose"} = + uri_string:parse("foo:/over/there#nose"), + #{scheme := "foo", host := "example.com", fragment := "nose"} = + uri_string:parse("foo://example.com#nose"), + #{scheme := "foo", host := "example.com", path := "/", fragment := "nose"} = + uri_string:parse("foo://example.com/#nose"), + #{scheme := "foo", host := "example.com", fragment := "nose"} = + uri_string:parse("foo://example.com#nose"), + + #{fragment := "nose"} = + uri_string:parse("#nose"), + #{path := "over/there", fragment := "nose"} = + uri_string:parse("over/there#nose"), + #{path := "/", fragment := "nose"} = + uri_string:parse("/#nose"), + #{path := "/over/there", fragment := "nose"} = + uri_string:parse("/over/there#nose"), + #{host := "example.com", fragment := "nose"} = + uri_string:parse("//example.com#nose"), + #{host := "example.com", path := "/", fragment := "nose"} = + uri_string:parse("//example.com/#nose"). + +parse_pct_encoded_fragment(_Config) -> + #{scheme := "foo", host := "example.com", fragment := "合気道"} = + uri_string:parse("foo://example.com#%E5%90%88%E6%B0%97%E9%81%93"), + #{host := "example.com", path := "/", fragment := "合気道"} = + uri_string:parse("//example.com/#%E5%90%88%E6%B0%97%E9%81%93"). + +parse_list(_Config) -> + #{scheme := "foo", path := "bar:nisse"} = uri_string:parse("foo:bar:nisse"), + #{scheme := "foo", host := "example.com", port := 8042, + path := "/over/there", query := "name=ferret", fragment := "nose"} = + uri_string:parse("foo://example.com:8042/over/there?name=ferret#nose"), + #{scheme := "foo", userinfo := "admin:admin", host := "example.com", port := 8042, + path := "/over/there", query := "name=ferret", fragment := "nose"} = + uri_string:parse("foo://admin:[email protected]:8042/over/there?name=ferret#nose"). + +parse_binary(_Config) -> + #{scheme := <<"foo">>, path := <<"bar:nisse">>} = uri_string:parse(<<"foo:bar:nisse">>), + #{scheme := <<"foo">>, host := <<"example.com">>, port := 8042, + path := <<"/over/there">>, query := <<"name=ferret">>, fragment := <<"nose">>} = + uri_string:parse(<<"foo://example.com:8042/over/there?name=ferret#nose">>), + #{scheme := <<"foo">>, userinfo := <<"admin:admin">>, host := <<"example.com">>, port := 8042, + path := <<"/over/there">>, query := <<"name=ferret">>, fragment := <<"nose">>} = + uri_string:parse(<<"foo://admin:[email protected]:8042/over/there?name=ferret#nose">>). + + +parse_mixed(_Config) -> + #{scheme := "foo", path := "bar"} = + uri_string:parse(lists:append("fo",<<"o:bar">>)), + #{scheme := "foo", path := "bar"} = + uri_string:parse(lists:append("foo:b",<<"ar">>)), + #{scheme := "foo", path := "bar:bar"} = + uri_string:parse([[102],[111,111],<<":bar">>,58,98,97,114]). + +parse_relative(_Config) -> + #{path := "/path"} = + uri_string:parse(lists:append("/pa",<<"th">>)), + #{path := "foo"} = + uri_string:parse(lists:append("fo",<<"o">>)). + +parse_special(_Config) -> + #{host := [],query := []} = uri_string:parse("//?"), + #{fragment := [],host := []} = uri_string:parse("//#"), + #{host := [],query := [],scheme := "foo"} = uri_string:parse("foo://?"), + #{fragment := [],host := [],scheme := "foo"} = uri_string:parse("foo://#"), + #{host := <<>>, path := <<"/">>} = uri_string:parse(<<"///">>), + #{host := <<"hostname">>} = uri_string:parse(<<"//hostname">>), + #{host := <<>>, path := <<"/hostname">>} = uri_string:parse(<<"///hostname">>), + #{host := [],path := "/",query := []} = uri_string:parse("///?"), + #{fragment := [],host := [],path := "/"} = uri_string:parse("///#"), + #{host := "foo",query := []} = uri_string:parse("//foo?"), + #{fragment := [],host := "foo"} = uri_string:parse("//foo#"), + #{host := "foo",path := "/"} = uri_string:parse("//foo/"), + #{host := "foo",query := [],scheme := "http"} = uri_string:parse("http://foo?"), + #{fragment := [],host := "foo",scheme := "http"} = uri_string:parse("http://foo#"), + #{host := "foo",path := "/",scheme := "http"} = uri_string:parse("http://foo/"), + #{fragment := [],host := "host",port := 80,scheme := "http"} = uri_string:parse("http://host:80#"), + #{host := "host",port := 80,query := [],scheme := "http"} = uri_string:parse("http://host:80?"), + #{path := [],query := []} = uri_string:parse("?"), + #{path := [],query := "?"} = uri_string:parse("??"), + #{path := [],query := "??"} = uri_string:parse("???"). + +parse_special2(_Config) -> + #{host := [],path := "/",port := 1,scheme := "a"} = uri_string:parse("a://:1/"), + #{path := "/a/",scheme := "a"} = uri_string:parse("a:/a/"), + #{host := [],path := [],userinfo := []} = uri_string:parse("//@"), + #{host := [],path := [],scheme := "foo",userinfo := []} = uri_string:parse("foo://@"), + #{host := [],path := "/",userinfo := []} = uri_string:parse("//@/"), + #{host := [],path := "/",scheme := "foo",userinfo := []} = uri_string:parse("foo://@/"), + #{host := "localhost",path := "/",port := undefined} = uri_string:parse("//localhost:/"), + #{host := [],path := [],port := undefined} = uri_string:parse("//:"). + +parse_negative(_Config) -> + {error,invalid_uri,"å"} = uri_string:parse("å"), + {error,invalid_uri,"å"} = uri_string:parse("aå:/foo"), + {error,invalid_uri,":"} = uri_string:parse("foo://usär@host"), + {error,invalid_uri,"ö"} = uri_string:parse("//host/path?foö=bar"), + {error,invalid_uri,"ö"} = uri_string:parse("//host/path#foö"), + {error,invalid_uri,"127.256.0.1"} = uri_string:parse("//127.256.0.1"), + {error,invalid_uri,":::127.0.0.1"} = uri_string:parse("//[:::127.0.0.1]"), + {error,invalid_utf8,<<0,0,0,246>>} = uri_string:parse("//%00%00%00%F6"), + {error,invalid_uri,"A"} = uri_string:parse("//localhost:A8"). + + +%%------------------------------------------------------------------------- +%% Recompose tests +%%------------------------------------------------------------------------- +recompose_fragment(_Config) -> + <<?FRAGMENT_ENC>> = uri_string:recompose(#{fragment => <<?FRAGMENT/utf8>>, path => <<>>}), + ?FRAGMENT_ENC = uri_string:recompose(#{fragment => ?FRAGMENT, path => ""}). + +recompose_parse_fragment(_Config) -> + <<?FRAGMENT_ENC>> = uri_string:recompose(uri_string:parse(<<?FRAGMENT_ENC>>)), + ?FRAGMENT_ENC = uri_string:recompose(uri_string:parse(?FRAGMENT_ENC)). + +recompose_query(_Config) -> + <<?QUERY_ENC>> = + uri_string:recompose(#{query => <<?QUERY/utf8>>, path => <<>>}), + <<?QUERY_ENC?FRAGMENT_ENC>> = + uri_string:recompose(#{query => <<?QUERY/utf8>>, + fragment => <<?FRAGMENT/utf8>>, + path => <<>>}), + "?name=%C3%B6rn" = + uri_string:recompose(#{query => "name=örn", path => ""}), + "?name=%C3%B6rn#n%C3%A4sa" = + uri_string:recompose(#{query => "name=örn", + fragment => "näsa", + path => ""}). + +recompose_parse_query(_Config) -> + <<"?name=%C3%B6rn">> = uri_string:recompose(uri_string:parse(<<"?name=%C3%B6rn">>)), + <<"?name=%C3%B6rn#n%C3%A4sa">> = + uri_string:recompose(uri_string:parse(<<"?name=%C3%B6rn#n%C3%A4sa">>)), + "?name=%C3%B6rn" = uri_string:recompose(uri_string:parse("?name=%C3%B6rn")), + "?name=%C3%B6rn#n%C3%A4sa" = uri_string:recompose(uri_string:parse("?name=%C3%B6rn#n%C3%A4sa")). + +recompose_path(_Config) -> + <<"/d%C3%A4r">> = + uri_string:recompose(#{path => <<"/där"/utf8>>}), + <<"/d%C3%A4r#n%C3%A4sa">> = + uri_string:recompose(#{path => <<"/där"/utf8>>, + fragment => <<"näsa"/utf8>>}), + <<"/d%C3%A4r?name=%C3%B6rn">> = + uri_string:recompose(#{path => <<"/där"/utf8>>, + query => <<"name=örn"/utf8>>}), + <<"/d%C3%A4r?name=%C3%B6rn#n%C3%A4sa">> = + uri_string:recompose(#{path => <<"/där"/utf8>>, + query => <<"name=örn"/utf8>>, + fragment => <<"näsa"/utf8>>}), + + + "/d%C3%A4r" = + uri_string:recompose(#{path => "/där"}), + "/d%C3%A4r#n%C3%A4sa" = + uri_string:recompose(#{path => "/där", + fragment => "näsa"}), + "/d%C3%A4r?name=%C3%B6rn" = + uri_string:recompose(#{path => "/där", + query => "name=örn"}), + "/d%C3%A4r?name=%C3%B6rn#n%C3%A4sa" = + uri_string:recompose(#{path => "/där", + query => "name=örn", + fragment => "näsa"}). + + +recompose_parse_path(_Config) -> + <<"/d%C3%A4r">> = + uri_string:recompose(uri_string:parse(<<"/d%C3%A4r">>)), + <<"/d%C3%A4r#n%C3%A4sa">> = + uri_string:recompose(uri_string:parse(<<"/d%C3%A4r#n%C3%A4sa">>)), + <<"/d%C3%A4r?name=%C3%B6rn">> = + uri_string:recompose(uri_string:parse(<<"/d%C3%A4r?name=%C3%B6rn">>)), + + "/d%C3%A4r" = + uri_string:recompose(uri_string:parse("/d%C3%A4r")), + "/d%C3%A4r#n%C3%A4sa" = + uri_string:recompose(uri_string:parse("/d%C3%A4r#n%C3%A4sa")), + "/d%C3%A4r?name=%C3%B6rn" = + uri_string:recompose(uri_string:parse("/d%C3%A4r?name=%C3%B6rn")). + + +recompose_autogen(_Config) -> + Tests = generate_test_vectors(uri_combinations()), + lists:map(fun run_test_recompose/1, Tests). + +parse_recompose_autogen(_Config) -> + Tests = generate_test_vectors(uri_combinations()), + lists:map(fun run_test_parse_recompose/1, Tests). + +transcode_basic(_Config) -> + <<"foo%C3%B6bar"/utf8>> = + uri_string:transcode(<<"foo%00%00%00%F6bar"/utf32>>, [{in_encoding, utf32},{out_encoding, utf8}]), + "foo%C3%B6bar" = + uri_string:transcode("foo%00%00%00%F6bar", [{in_encoding, utf32},{out_encoding, utf8}]), + <<"foo%00%00%00%F6bar"/utf32>> = + uri_string:transcode(<<"foo%C3%B6bar"/utf8>>, [{in_encoding, utf8},{out_encoding, utf32}]), + "foo%00%00%00%F6bar" = + uri_string:transcode("foo%C3%B6bar", [{in_encoding, utf8},{out_encoding, utf32}]), + "foo%C3%B6bar" = + uri_string:transcode("foo%F6bar", [{in_encoding, latin1},{out_encoding, utf8}]). + +transcode_options(_Config) -> + <<"foo%C3%B6bar"/utf8>> = + uri_string:transcode(<<"foo%C3%B6bar"/utf8>>, []), + <<"foo%C3%B6bar"/utf8>> = + uri_string:transcode(<<"foo%00%00%00%F6bar"/utf32>>, [{in_encoding, utf32}]), + <<"foo%00%00%00%F6bar"/utf32>> = + uri_string:transcode(<<"foo%C3%B6bar"/utf8>>, [{out_encoding, utf32}]). + +transcode_mixed(_Config) -> + "foo%00%00%00%F6bar" = + uri_string:transcode(["foo",<<"%C3%B6"/utf8>>,<<"ba"/utf8>>,"r"], [{out_encoding, utf32}]), + "foo%00%00%00%F6bar" = + uri_string:transcode(["foo",<<"%C3%"/utf8>>,<<"B6ba"/utf8>>,"r"], [{out_encoding, utf32}]), + "foo%C3%B6bar" = + uri_string:transcode(["foo%00", <<"%00%0"/utf32>>,<<"0%F"/utf32>>,"6bar"], [{in_encoding, utf32},{out_encoding, utf8}]). + +transcode_negative(_Config) -> + {error,invalid_percent_encoding,"%BXbar"} = + uri_string:transcode(<<"foo%C3%BXbar"/utf8>>, [{in_encoding, utf8},{out_encoding, utf32}]), + {error,invalid_input,<<"ö">>} = + uri_string:transcode("foo%F6bar", [{in_encoding, utf8},{out_encoding, utf8}]). + +compose_query(_Config) -> + [] = uri_string:compose_query([]), + "foo=1&bar=2" = uri_string:compose_query([{<<"foo">>,"1"}, {"bar", "2"}]), + "foo=1&b%C3%A4r=2" = uri_string:compose_query([{"foo","1"}, {"bär", "2"}],[{encoding,utf8}]), + "foo=1&b%C3%A4r=2" = uri_string:compose_query([{"foo","1"}, {"bär", "2"}],[{encoding,unicode}]), + "foo=1&b%E4r=2" = uri_string:compose_query([{"foo","1"}, {"bär", "2"}],[{encoding,latin1}]), + "foo+bar=1&%E5%90%88=2" = uri_string:compose_query([{"foo bar","1"}, {"合", "2"}]), + "foo+bar=1&%26%2321512%3B=2" = + uri_string:compose_query([{"foo bar","1"}, {"合", "2"}],[{encoding,latin1}]), + "foo+bar=1&%C3%B6=2" = uri_string:compose_query([{<<"foo bar">>,<<"1">>}, {"ö", <<"2">>}]), + <<"foo+bar=1&%C3%B6=2">> = + uri_string:compose_query([{<<"foo bar">>,<<"1">>}, {<<"ö"/utf8>>, <<"2">>}]). + +compose_query_latin1(_Config) -> + Q = uri_string:compose_query([{"合foö bar","1"}, {"合", "合"}],[{encoding,latin1}]), + Q1 = uri_string:transcode(Q, [{in_encoding, latin1}]), + [{"合foö bar","1"}, {"合", "合"}] = uri_string:dissect_query(Q1), + Q2 = uri_string:compose_query([{<<"合foö bar"/utf8>>,<<"1">>}, {<<"合"/utf8>>, <<"合"/utf8>>}], + [{encoding,latin1}]), + Q3 = uri_string:transcode(Q2, [{in_encoding, latin1}]), + [{<<"合foö bar"/utf8>>,<<"1">>}, {<<"合"/utf8>>, <<"合"/utf8>>}] = + uri_string:dissect_query(Q3). + +compose_query_negative(_Config) -> + {error,invalid_input,4} = uri_string:compose_query([{"",4}]), + {error,invalid_input,5} = uri_string:compose_query([{5,""}]), + {error,invalid_encoding,utf16} = + uri_string:compose_query([{"foo bar","1"}, {<<"ö">>, "2"}],[{encoding,utf16}]). + +dissect_query(_Config) -> + [] = uri_string:dissect_query(""), + [{"foo","1"}, {"amp;bar", "2"}] = uri_string:dissect_query("foo=1&bar=2"), + [{"foo","1"}, {"bar", "2"}] = uri_string:dissect_query("foo=1&bar=2"), + [{"foo","1;bar=2"}] = uri_string:dissect_query("foo=1;bar=2"), + [{"foo","1"}, {"bar", "222"}] = uri_string:dissect_query([<<"foo=1&bar=2">>,"22"]), + [{"foo","ö"}, {"bar", "2"}] = uri_string:dissect_query("foo=%C3%B6&bar=2"), + [{<<"foo">>,<<"ö"/utf8>>}, {<<"bar">>, <<"2">>}] = + uri_string:dissect_query(<<"foo=%C3%B6&bar=2">>), + [{"foo bar","1"},{"ö","2"}] = + uri_string:dissect_query([<<"foo+bar=1&">>,<<"%C3%B6=2">>]), + [{"foo bar","1"},{[21512],"2"}] = + uri_string:dissect_query("foo+bar=1&%26%2321512%3B=2"), + [{<<"foo bar">>,<<"1">>},{<<"合"/utf8>>,<<"2">>}] = + uri_string:dissect_query(<<"foo+bar=1&%26%2321512%3B=2">>), + [{"föo bar","1"},{"ö","2"}] = + uri_string:dissect_query("föo+bar=1&%C3%B6=2"), + [{<<"föo bar"/utf8>>,<<"1">>},{<<"ö"/utf8>>,<<"2">>}] = + uri_string:dissect_query(<<"föo+bar=1&%C3%B6=2"/utf8>>). + +dissect_query_negative(_Config) -> + {error,missing_value,"&"} = + uri_string:dissect_query("foo1&bar=2"), + {error,invalid_percent_encoding,"%XX%B6"} = uri_string:dissect_query("foo=%XX%B6&bar=2"), + {error,invalid_input,[153]} = + uri_string:dissect_query("foo=%99%B6&bar=2"), + {error,invalid_character,"ö"} = uri_string:dissect_query(<<"föo+bar=1&%C3%B6=2">>), + {error,invalid_input,<<"ö">>} = + uri_string:dissect_query([<<"foo+bar=1&">>,<<"%C3%B6=2ö">>]). + +normalize(_Config) -> + "/a/g" = uri_string:normalize("/a/b/c/./../../g"), + <<"mid/6">> = uri_string:normalize(<<"mid/content=5/../6">>), + "http://localhost-%C3%B6rebro/a/g" = + uri_string:normalize("http://localhos%74-%c3%b6rebro:80/a/b/c/./../../g"), + <<"http://localhost-%C3%B6rebro/a/g">> = + uri_string:normalize(<<"http://localhos%74-%c3%b6rebro:80/a/b/c/./../../g">>), + <<"https://localhost/">> = + uri_string:normalize(<<"https://localhost:443">>), + <<"https://localhost:445/">> = + uri_string:normalize(<<"https://localhost:445">>), + <<"ftp://localhost">> = + uri_string:normalize(<<"ftp://localhost:21">>), + <<"ssh://localhost">> = + uri_string:normalize(<<"ssh://localhost:22">>), + <<"sftp://localhost">> = + uri_string:normalize(<<"sftp://localhost:22">>), + <<"tftp://localhost">> = + uri_string:normalize(<<"tftp://localhost:69">>). + +interop_query_utf8(_Config) -> + Q = uri_string:compose_query([{"foo bar","1"}, {"合", "2"}]), + Uri = uri_string:recompose(#{path => "/", query => Q}), + #{query := Q1} = uri_string:parse(Uri), + [{"foo bar","1"}, {"合", "2"}] = uri_string:dissect_query(Q1). + +interop_query_latin1(_Config) -> + Q = uri_string:compose_query([{"foo bar","1"}, {"合", "2"}], [{encoding,latin1}]), + Uri = uri_string:recompose(#{path => "/", query => Q}), + Uri1 = uri_string:transcode(Uri, [{in_encoding, latin1}]), + #{query := Q1} = uri_string:parse(Uri1), + [{"foo bar","1"}, {"合", "2"}] = uri_string:dissect_query(Q1). diff --git a/lib/stdlib/test/uri_string_property_test_SUITE.erl b/lib/stdlib/test/uri_string_property_test_SUITE.erl new file mode 100644 index 0000000000..ae2c61c7aa --- /dev/null +++ b/lib/stdlib/test/uri_string_property_test_SUITE.erl @@ -0,0 +1,39 @@ +%% +%% %CopyrightBegin% +%% +%% Copyright Ericsson AB 2008-2017. All Rights Reserved. +%% +%% Licensed under the Apache License, Version 2.0 (the "License"); +%% you may not use this file except in compliance with the License. +%% You may obtain a copy of the License at +%% +%% http://www.apache.org/licenses/LICENSE-2.0 +%% +%% Unless required by applicable law or agreed to in writing, software +%% distributed under the License is distributed on an "AS IS" BASIS, +%% WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +%% See the License for the specific language governing permissions and +%% limitations under the License. +%% +%% %CopyrightEnd% +%% +-module(uri_string_property_test_SUITE). + +-include_lib("common_test/include/ct.hrl"). +-compile(export_all). + +all() -> [recompose]. + +init_per_suite(Config) -> + ct_property_test:init_per_suite(Config). + +end_per_suite(Config) -> + Config. + +%%%======================================================================== +%%% Test suites +%%%======================================================================== +recompose(Config) -> + ct_property_test:quickcheck( + uri_string_recompose:prop_recompose(), + Config). diff --git a/lib/stdlib/uc_spec/gen_unicode_mod.escript b/lib/stdlib/uc_spec/gen_unicode_mod.escript index 674e5a0628..fe5a860d45 100755 --- a/lib/stdlib/uc_spec/gen_unicode_mod.escript +++ b/lib/stdlib/uc_spec/gen_unicode_mod.escript @@ -170,7 +170,7 @@ gen_header(Fd) -> io:put_chars(Fd, "-export([spec_version/0, lookup/1, get_case/1]).\n"), io:put_chars(Fd, "-inline([class/1]).\n"), io:put_chars(Fd, "-compile(nowarn_unused_vars).\n"), - io:put_chars(Fd, "-dialyzer({no_improper_lists, [cp/1, gc_prepend/2, gc_e_cont/2]}).\n"), + io:put_chars(Fd, "-dialyzer({no_improper_lists, [cp/1, gc/1, gc_prepend/2, gc_e_cont/2]}).\n"), io:put_chars(Fd, "-type gc() :: char()|[char()].\n\n\n"), ok. @@ -240,7 +240,7 @@ gen_norm(Fd) -> "-spec nfd(unicode:chardata()) -> maybe_improper_list(gc(),unicode:chardata()) | {error, unicode:chardata()}.\n" "nfd(Str0) ->\n" " case gc(Str0) of\n" - " [GC|R] when GC < 127 -> [GC|R];\n" + " [GC|R] when GC < 128 -> [GC|R];\n" " [GC|Str] -> [decompose(GC)|Str];\n" " [] -> [];\n" " {error,_}=Error -> Error\n end.\n\n" @@ -250,7 +250,7 @@ gen_norm(Fd) -> "-spec nfkd(unicode:chardata()) -> maybe_improper_list(gc(),unicode:chardata()) | {error, unicode:chardata()}.\n" "nfkd(Str0) ->\n" " case gc(Str0) of\n" - " [GC|R] when GC < 127 -> [GC|R];\n" + " [GC|R] when GC < 128 -> [GC|R];\n" " [GC|Str] -> [decompose_compat(GC)|Str];\n" " [] -> [];\n" " {error,_}=Error -> Error\n end.\n\n" @@ -260,7 +260,7 @@ gen_norm(Fd) -> "-spec nfc(unicode:chardata()) -> maybe_improper_list(gc(),unicode:chardata()) | {error, unicode:chardata()}.\n" "nfc(Str0) ->\n" " case gc(Str0) of\n" - " [GC|R] when GC < 255 -> [GC|R];\n" + " [GC|R] when GC < 256 -> [GC|R];\n" " [GC|Str] -> [compose(decompose(GC))|Str];\n" " [] -> [];\n" " {error,_}=Error -> Error\n end.\n\n" @@ -270,7 +270,7 @@ gen_norm(Fd) -> "-spec nfkc(unicode:chardata()) -> maybe_improper_list(gc(),unicode:chardata()) | {error, unicode:chardata()}.\n" "nfkc(Str0) ->\n" " case gc(Str0) of\n" - " [GC|R] when GC < 127 -> [GC|R];\n" + " [GC|R] when GC < 128 -> [GC|R];\n" " [GC|Str] -> [compose_compat_0(decompose_compat(GC))|Str];\n" " [] -> [];\n" " {error,_}=Error -> Error\n end.\n\n" @@ -476,13 +476,30 @@ gen_gc(Fd, GBP) -> "-spec gc(String::unicode:chardata()) ->" " maybe_improper_list() | {error, unicode:chardata()}.\n"), io:put_chars(Fd, + "gc([CP1, CP2|_]=T)\n" + " when CP1 < 256, CP2 < 256, CP1 =/= $\r -> %% Ascii Fast path\n" + " T;\n" + "gc(<<CP1/utf8, Rest/binary>>) ->\n" + " if CP1 < 256, CP1 =/= $\r ->\n" + " case Rest of\n" + " <<CP2/utf8, _/binary>> when CP2 < 256 -> %% Ascii Fast path\n" + " [CP1|Rest];\n" + " _ -> gc_1([CP1|Rest])\n" + " end;\n" + " true -> gc_1([CP1|Rest])\n" + " end;\n" "gc(Str) ->\n" " gc_1(cp(Str)).\n\n" "gc_1([$\\r|R0] = R) ->\n" " case cp(R0) of % Don't break CRLF\n" " [$\\n|R1] -> [[$\\r,$\\n]|R1];\n" " _ -> R\n" - " end;\n"), + " end;\n" + %% "gc_1([CP1, CP2|_]=T) when CP1 < 256, CP2 < 256 ->\n" + %% " T; %% Fast path\n" + %% "gc_1([CP1|<<CP2/utf8, _/binary>>]=T) when CP1 < 256, CP2 < 256 ->\n" + %% " T; %% Fast path\n" + ), io:put_chars(Fd, "%% Handle control\n"), GenControl = fun(Range) -> io:format(Fd, "gc_1~s R0;\n", [gen_clause(Range)]) end, @@ -490,7 +507,7 @@ gen_gc(Fd, GBP) -> [R1,R2,R3|Crs] = CRs0, [GenControl(CP) || CP <- merge_ranges([R1,R2,R3], split), CP =/= {$\r, undefined}], %%GenControl(R1),GenControl(R2),GenControl(R3), - io:format(Fd, "gc_1([CP|R]) when CP < 255 -> gc_extend(R,CP);\n", []), + io:format(Fd, "gc_1([CP|R]) when CP < 256 -> gc_extend(R,CP);\n", []), [GenControl(CP) || CP <- Crs], %% One clause per CP %% CRs0 = merge_ranges(maps:get(cr, GBP) ++ maps:get(lf, GBP) ++ maps:get(control, GBP)), diff --git a/lib/stdlib/vsn.mk b/lib/stdlib/vsn.mk index 48db5dc900..69d258c2f0 100644 --- a/lib/stdlib/vsn.mk +++ b/lib/stdlib/vsn.mk @@ -1 +1 @@ -STDLIB_VSN = 3.4.2 +STDLIB_VSN = 3.4.3 |
