diff options
Diffstat (limited to 'lib/xmerl')
-rw-r--r-- | lib/xmerl/doc/src/notes.xml | 42 | ||||
-rw-r--r-- | lib/xmerl/notes.html | 28 | ||||
-rw-r--r-- | lib/xmerl/src/xmerl_regexp.erl | 4 | ||||
-rw-r--r-- | lib/xmerl/src/xmerl_sax_old_dom.erl | 9 | ||||
-rw-r--r-- | lib/xmerl/src/xmerl_sax_parser.erl | 171 | ||||
-rw-r--r-- | lib/xmerl/src/xmerl_sax_parser_base.erlsrc | 293 | ||||
-rw-r--r-- | lib/xmerl/src/xmerl_sax_simple_dom.erl | 7 | ||||
-rw-r--r-- | lib/xmerl/src/xmerl_scan.erl | 4 | ||||
-rw-r--r-- | lib/xmerl/test/Makefile | 1 | ||||
-rw-r--r-- | lib/xmerl/test/xmerl_SUITE.erl | 9 | ||||
-rw-r--r-- | lib/xmerl/test/xmerl_sax_SUITE.erl | 76 | ||||
-rw-r--r-- | lib/xmerl/test/xmerl_sax_SUITE_data/test_data_1.xml | 4 | ||||
-rw-r--r-- | lib/xmerl/test/xmerl_sax_stream_SUITE.erl | 17 | ||||
-rw-r--r-- | lib/xmerl/vsn.mk | 2 | ||||
-rw-r--r-- | lib/xmerl/xmerl.pub | 19 |
15 files changed, 479 insertions, 207 deletions
diff --git a/lib/xmerl/doc/src/notes.xml b/lib/xmerl/doc/src/notes.xml index 652560f60c..1162561225 100644 --- a/lib/xmerl/doc/src/notes.xml +++ b/lib/xmerl/doc/src/notes.xml @@ -4,7 +4,7 @@ <chapter> <header> <copyright> - <year>2004</year><year>2016</year> + <year>2004</year><year>2017</year> <holder>Ericsson AB. All Rights Reserved.</holder> </copyright> <legalnotice> @@ -32,6 +32,46 @@ <p>This document describes the changes made to the Xmerl application.</p> +<section><title>Xmerl 1.3.15</title> + + <section><title>Fixed Bugs and Malfunctions</title> + <list> + <item> + <p> + Improves accumulator fun in xmerl_scan so that only one + #xmlText record is returned for strings which have + character references.</p> + <p> + (Thanks to Jimmy Zöger)</p> + <p> + Own Id: OTP-14377 Aux Id: PR-1369 </p> + </item> + </list> + </section> + +</section> + +<section><title>Xmerl 1.3.14</title> + + <section><title>Fixed Bugs and Malfunctions</title> + <list> + <item> + <p> A couple of bugs are fixed in the sax parser + (xmerl_sax_parser). </p> <list> <item>The continuation + function was not called correctly when the XML directive + was fragmented. </item> <item>When the event callback + modules (xmerl_sax_old_dom and xmerl_sax_simple) got an + endDocument event at certain conditions the parser + crashed.</item> <item>Replaced internal ets table with + map to avoid table leakage.</item> </list> + <p> + Own Id: OTP-14430</p> + </item> + </list> + </section> + +</section> + <section><title>Xmerl 1.3.13</title> <section><title>Fixed Bugs and Malfunctions</title> diff --git a/lib/xmerl/notes.html b/lib/xmerl/notes.html deleted file mode 100644 index 43610a37fa..0000000000 --- a/lib/xmerl/notes.html +++ /dev/null @@ -1,28 +0,0 @@ -<HTML> -<HEAD> - <TITLE>xmerl Release Notes</TITLE> - <style type="text/css"> -<!-- - body { background: white; margin: 3em } - - body { font-family: Verdana, Arial, Helvetica, sans-serif } - h1 h2 h3 h4 { font-family: Verdana, Arial, Helvetica, sans-serif } - h1 { font-size: 48 } - p li { font-family: Verdana, Arial, Helvetica, sans-serif } ---> - </style> -</HEAD> - -<BODY BGCOLOR="#FFFFFF"> - -<CENTER><H1>xmerl Release Notes</H1></CENTER> - -<h2>xmerl 1.0</h2> - - -<p> -There are also release notes for -<a href="notes_history.html">older versions</a>. - -</body> -</html> diff --git a/lib/xmerl/src/xmerl_regexp.erl b/lib/xmerl/src/xmerl_regexp.erl index fc89b80ff1..1bf8496673 100644 --- a/lib/xmerl/src/xmerl_regexp.erl +++ b/lib/xmerl/src/xmerl_regexp.erl @@ -1,7 +1,7 @@ %% %% %CopyrightBegin% %% -%% Copyright Ericsson AB 2006-2016. All Rights Reserved. +%% Copyright Ericsson AB 2006-2017. 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. @@ -1154,7 +1154,7 @@ comp_crs([], Last) -> [{Last,maxchar}]. %% build_dfa(NFA, NfaStartState) -> {DFA,DfaStartState}. %% Build a DFA from an NFA using "subset construction". The major %% difference from the book is that we keep the marked and unmarked -%% DFA states in seperate lists. New DFA states are added to the +%% DFA states in separate lists. New DFA states are added to the %% unmarked list and states are marked by moving them to the marked %% list. We assume that the NFA accepting state numbers are in %% ascending order for the rules and use ordsets to keep this order. diff --git a/lib/xmerl/src/xmerl_sax_old_dom.erl b/lib/xmerl/src/xmerl_sax_old_dom.erl index fefcf03fce..6d0d836487 100644 --- a/lib/xmerl/src/xmerl_sax_old_dom.erl +++ b/lib/xmerl/src/xmerl_sax_old_dom.erl @@ -2,7 +2,7 @@ %%-------------------------------------------------------------------- %% %CopyrightBegin% %% -%% Copyright Ericsson AB 2009-2016. All Rights Reserved. +%% Copyright Ericsson AB 2009-2017. 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. @@ -127,9 +127,10 @@ build_dom(endDocument, State#xmerl_sax_old_dom_state{dom=[Decl, Current#xmlElement{ content=lists:reverse(C) }]}; - _ -> - %%?dbg("~p\n", [D]), - ?error("we're not at end the document when endDocument event is encountered.") + _ -> + %% endDocument is also sent by the parser when a fault occur to tell + %% the event receiver that no more input will be sent + State end; %% Element diff --git a/lib/xmerl/src/xmerl_sax_parser.erl b/lib/xmerl/src/xmerl_sax_parser.erl index 1aef6c58c4..e383c4c349 100644 --- a/lib/xmerl/src/xmerl_sax_parser.erl +++ b/lib/xmerl/src/xmerl_sax_parser.erl @@ -64,7 +64,7 @@ %% Description: Parse file containing an XML document. %%---------------------------------------------------------------------- file(Name,Options) -> - case file:open(Name, [raw, read,binary]) of + case file:open(Name, [raw, read_ahead, read,binary]) of {error, Reason} -> {error,{Name, file:format_error(Reason)}}; {ok, FD} -> @@ -120,21 +120,22 @@ stream(Xml, Options, InputType) when is_binary(Xml), is_list(Options) -> normal -> parse end, - case detect_charset(Xml, State) of - {error, Reason} -> {fatal_error, - { - State#xmerl_sax_parser_state.current_location, - State#xmerl_sax_parser_state.entity, - 1 - }, - Reason, - [], - State#xmerl_sax_parser_state.event_state}; - {Xml1, State1} -> - parse_binary(Xml1, - State1#xmerl_sax_parser_state{input_type = InputType}, - ParseFunction) - end + try + {Xml1, State1} = detect_charset(Xml, State), + parse_binary(Xml1, + State1#xmerl_sax_parser_state{input_type = InputType}, + ParseFunction) + catch + throw:{fatal_error, {State2, Reason}} -> + {fatal_error, + { + State2#xmerl_sax_parser_state.current_location, + State2#xmerl_sax_parser_state.entity, + 1 + }, + Reason, [], + State2#xmerl_sax_parser_state.event_state} + end end. %%---------------------------------------------------------------------- @@ -156,8 +157,8 @@ parse_binary(Xml, #xmerl_sax_parser_state{encoding={utf16,big}}=State, F) -> xmerl_sax_parser_utf16be:F(Xml, State); parse_binary(Xml, #xmerl_sax_parser_state{encoding=latin1}=State, F) -> xmerl_sax_parser_latin1:F(Xml, State); -parse_binary(_, #xmerl_sax_parser_state{encoding=Enc}, _) -> - {error, lists:flatten(io_lib:format("Charcter set ~p not supported", [Enc]))}. +parse_binary(_, #xmerl_sax_parser_state{encoding=Enc}, State) -> + ?fatal_error(State, lists:flatten(io_lib:format("Charcter set ~p not supported", [Enc]))). %%---------------------------------------------------------------------- %% Function: initial_state/0 @@ -211,8 +212,7 @@ parse_options([{entity, Entity} |Options], State) -> parse_options([skip_external_dtd |Options], State) -> parse_options(Options, State#xmerl_sax_parser_state{skip_external_dtd = true}); parse_options([O |_], _State) -> - {error, - lists:flatten(io_lib:format("Option: ~p not supported", [O]))}. + {error, lists:flatten(io_lib:format("Option: ~p not supported", [O]))}. check_encoding_option(E) when E==utf8; E=={utf16,little}; E=={utf16,big}; @@ -230,16 +230,10 @@ check_encoding_option(E) -> %% Output: {utf8|utf16le|utf16be|iso8859, Xml, State} %% Description: Detects which character set is used in a binary stream. %%---------------------------------------------------------------------- -detect_charset(<<>>, #xmerl_sax_parser_state{continuation_fun = undefined} = _) -> - {error, "Can't detect character encoding due to no indata"}; -detect_charset(<<>>, #xmerl_sax_parser_state{continuation_fun = CFun, - continuation_state = CState} = State) -> - case CFun(CState) of - {<<>>, _} -> - {error, "Can't detect character encoding due to lack of indata"}; - {NewBytes, NewContState} -> - detect_charset(NewBytes, State#xmerl_sax_parser_state{continuation_state = NewContState}) - end; +detect_charset(<<>>, #xmerl_sax_parser_state{continuation_fun = undefined} = State) -> + ?fatal_error(State, "Can't detect character encoding due to lack of indata"); +detect_charset(<<>>, State) -> + cf(<<>>, State, fun detect_charset/2); detect_charset(Bytes, State) -> case unicode:bom_to_encoding(Bytes) of {latin1, 0} -> @@ -249,25 +243,47 @@ detect_charset(Bytes, State) -> {RealBytes, State#xmerl_sax_parser_state{encoding=Enc}} end. +detect_charset_1(<<16#00>> = Xml, State) -> + cf(Xml, State, fun detect_charset_1/2); +detect_charset_1(<<16#00, 16#3C>> = Xml, State) -> + cf(Xml, State, fun detect_charset_1/2); +detect_charset_1(<<16#00, 16#3C, 16#00>> = Xml, State) -> + cf(Xml, State, fun detect_charset_1/2); detect_charset_1(<<16#00, 16#3C, 16#00, 16#3F, _/binary>> = Xml, State) -> {Xml, State#xmerl_sax_parser_state{encoding={utf16, big}}}; +detect_charset_1(<<16#3C>> = Xml, State) -> + cf(Xml, State, fun detect_charset_1/2); +detect_charset_1(<<16#3C, 16#00>> = Xml, State) -> + cf(Xml, State, fun detect_charset_1/2); +detect_charset_1(<<16#3C, 16#00, 16#3F>> = Xml, State) -> + cf(Xml, State, fun detect_charset_1/2); detect_charset_1(<<16#3C, 16#00, 16#3F, 16#00, _/binary>> = Xml, State) -> {Xml, State#xmerl_sax_parser_state{encoding={utf16, little}}}; -detect_charset_1(<<16#3C, 16#3F, 16#78, 16#6D, 16#6C, Xml2/binary>> = Xml, State) -> - case parse_xml_directive(Xml2) of +detect_charset_1(<<16#3C>> = Xml, State) -> + cf(Xml, State, fun detect_charset_1/2); +detect_charset_1(<<16#3C, 16#3F>> = Xml, State) -> + cf(Xml, State, fun detect_charset_1/2); +detect_charset_1(<<16#3C, 16#3F, 16#78>> = Xml, State) -> + cf(Xml, State, fun detect_charset_1/2); +detect_charset_1(<<16#3C, 16#3F, 16#78, 16#6D>> = Xml, State) -> + cf(Xml, State, fun detect_charset_1/2); +detect_charset_1(<<16#3C, 16#3F, 16#78, 16#6D, 16#6C, Xml2/binary>>, State) -> + {Xml3, State1} = read_until_end_of_xml_directive(Xml2, State), + case parse_xml_directive(Xml3) of {error, Reason} -> - {error, Reason}; + ?fatal_error(State, Reason); AttrList -> case lists:keysearch("encoding", 1, AttrList) of {value, {_, E}} -> case convert_encoding(E) of {error, Reason} -> - {error, Reason}; + ?fatal_error(State, Reason); Enc -> - {Xml, State#xmerl_sax_parser_state{encoding=Enc}} + {<<16#3C, 16#3F, 16#78, 16#6D, 16#6C, Xml3/binary>>, + State1#xmerl_sax_parser_state{encoding=Enc}} end; _ -> - {Xml, State} + {<<16#3C, 16#3F, 16#78, 16#6D, 16#6C, Xml3/binary>>, State1} end end; detect_charset_1(Xml, State) -> @@ -377,7 +393,7 @@ parse_value_1(<<C, Rest/binary>>, Stop, Acc) -> parse_value_1(Rest, Stop, [C |Acc]). %%====================================================================== -%%Default functions +%% Default functions %%====================================================================== %%---------------------------------------------------------------------- %% Function: default_event_cb(Event, LineNo, State) -> Result @@ -393,7 +409,7 @@ default_event_cb(_Event, _LineNo, State) -> %%---------------------------------------------------------------------- %% Function: default_continuation_cb(IoDevice) -> Result %% IoDevice = iodevice() -%% Output: Result = {[char()], State} +%% Output: Result = {binary(), IoDevice} %% Description: Default continuation callback reading blocks. %%---------------------------------------------------------------------- default_continuation_cb(IoDevice) -> @@ -403,3 +419,82 @@ default_continuation_cb(IoDevice) -> {ok, FileBin} -> {FileBin, IoDevice} end. + +%%---------------------------------------------------------------------- +%% Function: read_until_end_of_xml_directive(Rest, State) -> Result +%% Rest = binary() +%% Output: Result = {binary(), State} +%% Description: Reads a utf8 or latin1 until it finds '?>' +%%---------------------------------------------------------------------- +read_until_end_of_xml_directive(Rest, State) -> + case binary:match(Rest, <<"?>">>) of + nomatch -> + case cf(Rest, State) of + {<<>>, _} -> + ?fatal_error(State, "Can't detect character encoding due to lack of indata"); + {NewBytes, NewState} -> + read_until_end_of_xml_directive(NewBytes, NewState) + end; + _ -> + {Rest, State} + end. + + +%%---------------------------------------------------------------------- +%% Function : cf(Rest, State) -> Result +%% Parameters: Rest = binary() +%% State = #xmerl_sax_parser_state{} +%% NextCall = fun() +%% Result : {Rest, State} +%% Description: Function that uses provided fun to read another chunk from +%% input stream and calls the fun in NextCall. +%%---------------------------------------------------------------------- +cf(_Rest, #xmerl_sax_parser_state{continuation_fun = undefined} = State) -> + ?fatal_error(State, "Continuation function undefined"); +cf(Rest, #xmerl_sax_parser_state{continuation_fun = CFun, continuation_state = CState} = State) -> + Result = + try + CFun(CState) + catch + throw:ErrorTerm -> + ?fatal_error(State, ErrorTerm); + exit:Reason -> + ?fatal_error(State, {'EXIT', Reason}) + end, + case Result of + {<<>>, _} -> + ?fatal_error(State, "Can't detect character encoding due to lack of indata"); + {NewBytes, NewContState} -> + {<<Rest/binary, NewBytes/binary>>, + State#xmerl_sax_parser_state{continuation_state = NewContState}} + end. + +%%---------------------------------------------------------------------- +%% Function : cf(Rest, State, NextCall) -> Result +%% Parameters: Rest = binary() +%% State = #xmerl_sax_parser_state{} +%% NextCall = fun() +%% Result : {Rest, State} +%% Description: Function that uses provided fun to read another chunk from +%% input stream and calls the fun in NextCall. +%%---------------------------------------------------------------------- +cf(_Rest, #xmerl_sax_parser_state{continuation_fun = undefined} = State, _) -> + ?fatal_error(State, "Continuation function undefined"); +cf(Rest, #xmerl_sax_parser_state{continuation_fun = CFun, continuation_state = CState} = State, + NextCall) -> + Result = + try + CFun(CState) + catch + throw:ErrorTerm -> + ?fatal_error(State, ErrorTerm); + exit:Reason -> + ?fatal_error(State, {'EXIT', Reason}) + end, + case Result of + {<<>>, _} -> + ?fatal_error(State, "Can't detect character encoding due to lack of indata"); + {NewBytes, NewContState} -> + NextCall(<<Rest/binary, NewBytes/binary>>, + State#xmerl_sax_parser_state{continuation_state = NewContState}) + end. diff --git a/lib/xmerl/src/xmerl_sax_parser_base.erlsrc b/lib/xmerl/src/xmerl_sax_parser_base.erlsrc index f3470b2809..1dca9608cb 100644 --- a/lib/xmerl/src/xmerl_sax_parser_base.erlsrc +++ b/lib/xmerl/src/xmerl_sax_parser_base.erlsrc @@ -64,38 +64,42 @@ %% Description: Parsing XML from input stream. %%---------------------------------------------------------------------- parse(Xml, State) -> - RefTable = ets:new(xmerl_sax_entity_refs, [private]), - - State1 = event_callback(startDocument, State), - - case catch parse_document(Xml, State1#xmerl_sax_parser_state{ref_table=RefTable}) of - {ok, Rest, State2} -> - State3 = event_callback(endDocument, State2), - ets:delete(RefTable), - case check_if_rest_ok(State3#xmerl_sax_parser_state.input_type, Rest) of - true -> - {ok, State3#xmerl_sax_parser_state.event_state, Rest}; - false -> - format_error(fatal_error, State3, "Input found after legal document") - end; - {fatal_error, {State2, Reason}} -> - State3 = event_callback(endDocument, State2), - ets:delete(RefTable), - format_error(fatal_error, State3, Reason); - {event_receiver_error, State2, {Tag, Reason}} -> - State3 = event_callback(endDocument, State2), - ets:delete(RefTable), - format_error(Tag, State3, Reason); - {endDocument, Rest, State2} -> - State3 = event_callback(endDocument, State2), - ets:delete(RefTable), - {ok, State3#xmerl_sax_parser_state.event_state, Rest}; - Other -> - _State2 = event_callback(endDocument, State1), - ets:delete(RefTable), - {fatal_error, Other} + RefTable = maps:new(), + + try + State1 = event_callback(startDocument, State), + Result = parse_document(Xml, State1#xmerl_sax_parser_state{ref_table=RefTable}), + handle_end_document(Result) + catch + throw:Exception -> + handle_end_document(Exception); + _:OtherError -> + handle_end_document({other, OtherError, State}) end. + % case catch parse_document(Xml, State1#xmerl_sax_parser_state{ref_table=RefTable}) of + % {ok, Rest, State2} -> + % State3 = event_callback(endDocument, State2), + % case check_if_rest_ok(State3#xmerl_sax_parser_state.input_type, Rest) of + % true -> + % {ok, State3#xmerl_sax_parser_state.event_state, Rest}; + % false -> + % format_error(fatal_error, State3, "Input found after legal document") + % end; + % {fatal_error, {State2, Reason}} -> + % State3 = event_callback(endDocument, State2), + % format_error(fatal_error, State3, Reason); + % {event_receiver_error, State2, {Tag, Reason}} -> + % State3 = event_callback(endDocument, State2), + % format_error(Tag, State3, Reason); + % {endDocument, Rest, State2} -> + % State3 = event_callback(endDocument, State2), + % {ok, State3#xmerl_sax_parser_state.event_state, Rest}; + % Other -> + % _State2 = event_callback(endDocument, State1), + % {fatal_error, Other} + % end. + %%---------------------------------------------------------------------- %% Function: parse_dtd(Xml, State) -> Result %% Input: Xml = string() | binary() @@ -105,38 +109,120 @@ parse(Xml, State) -> %% Description: Parsing XML DTD from input stream. %%---------------------------------------------------------------------- parse_dtd(Xml, State) -> - RefTable = ets:new(xmerl_sax_entity_refs, [private]), - - State1 = event_callback(startDocument, State), - - case catch parse_external_entity_1(Xml, State1#xmerl_sax_parser_state{ref_table=RefTable}) of - {fatal_error, {State2, Reason}} -> - State3 = event_callback(endDocument, State2), - ets:delete(RefTable), - format_error(fatal_error, State3, Reason); - {event_receiver_error, State2, {Tag, Reason}} -> - State3 = event_callback(endDocument, State2), - format_error(Tag, State3, Reason); - {Rest, State2} when is_record(State2, xmerl_sax_parser_state) -> - State3 = event_callback(endDocument, State2), - ets:delete(RefTable), - {ok, State3#xmerl_sax_parser_state.event_state, Rest}; - {endDocument, Rest, State2} when is_record(State2, xmerl_sax_parser_state) -> - State3 = event_callback(endDocument, State2), - ets:delete(RefTable), - {ok, State3#xmerl_sax_parser_state.event_state, Rest}; - Other -> - _State2 = event_callback(endDocument, State1), - ets:delete(RefTable), - {fatal_error, Other} + RefTable = maps:new(), + + try + State1 = event_callback(startDocument, State), + Result = parse_external_entity_1(Xml, State1#xmerl_sax_parser_state{ref_table=RefTable}), + handle_end_document(Result) + catch + throw:Exception -> + handle_end_document(Exception); + _:OtherError -> + handle_end_document({other, OtherError, State}) end. + % case catch parse_external_entity_1(Xml, State1#xmerl_sax_parser_state{ref_table=RefTable}) of + % {fatal_error, {State2, Reason}} -> + % State3 = event_callback(endDocument, State2), + % format_error(fatal_error, State3, Reason); + % {event_receiver_error, State2, {Tag, Reason}} -> + % State3 = event_callback(endDocument, State2), + % format_error(Tag, State3, Reason); + % {Rest, State2} when is_record(State2, xmerl_sax_parser_state) -> + % State3 = event_callback(endDocument, State2), + % {ok, State3#xmerl_sax_parser_state.event_state, Rest}; + % {endDocument, Rest, State2} when is_record(State2, xmerl_sax_parser_state) -> + % State3 = event_callback(endDocument, State2), + % {ok, State3#xmerl_sax_parser_state.event_state, Rest}; + % Other -> + % _State2 = event_callback(endDocument, State1), + % {fatal_error, Other} + % end. + + %%====================================================================== %% Internal functions %%====================================================================== %%---------------------------------------------------------------------- +%% Function: handle_end_document(ParserResult) -> Result +%% Input: ParseResult = term() +%% Output: Result = {ok, Rest, EventState} | +%% EventState = term() +%% Description: Ends the parsing and formats output +%%---------------------------------------------------------------------- +handle_end_document({ok, Rest, State}) -> + %%ok case from parse + try + State1 = event_callback(endDocument, State), + case check_if_rest_ok(State1#xmerl_sax_parser_state.input_type, Rest) of + true -> + {ok, State1#xmerl_sax_parser_state.event_state, Rest}; + false -> + format_error(fatal_error, State1, "Input found after legal document") + end + catch + throw:{event_receiver_error, State2, {Tag, Reason}} -> + format_error(Tag, State2, Reason); + _:Other -> + {fatal_error, Other} + end; +handle_end_document({endDocument, Rest, State}) -> + %% ok case from parse and parse_dtd + try + State1 = event_callback(endDocument, State), + {ok, State1#xmerl_sax_parser_state.event_state, Rest} + catch + throw:{event_receiver_error, State2, {Tag, Reason}} -> + format_error(Tag, State2, Reason); + _:Other -> + {fatal_error, Other} + end; +handle_end_document({fatal_error, {State, Reason}}) -> + try + State1 = event_callback(endDocument, State), + format_error(fatal_error, State1, Reason) + catch + throw:{event_receiver_error, State2, {Tag, Reason}} -> + format_error(Tag, State2, Reason); + _:Other -> + {fatal_error, Other} + end; +handle_end_document({event_receiver_error, State, {Tag, Reason}}) -> + try + State1 = event_callback(endDocument, State), + format_error(Tag, State1, Reason) + catch + throw:{event_receiver_error, State2, {Tag, Reason}} -> + format_error(Tag, State2, Reason); + _:Other -> + {fatal_error, Other} + end; +handle_end_document({Rest, State}) when is_record(State, xmerl_sax_parser_state) -> + %%ok case from parse_dtd + try + State1 = event_callback(endDocument, State), + {ok, State1#xmerl_sax_parser_state.event_state, Rest} + catch + throw:{event_receiver_error, State2, {Tag, Reason}} -> + format_error(Tag, State2, Reason); + _:Other -> + {fatal_error, Other} + end; +handle_end_document({other, Error, State}) -> + try + _State1 = event_callback(endDocument, State), + {fatal_error, Error} + catch + throw:{event_receiver_error, State2, {Tag, Reason}} -> + format_error(Tag, State2, Reason); + _:Other -> + {fatal_error, Other} + end. + +%%---------------------------------------------------------------------- %% Function: parse_document(Rest, State) -> Result %% Input: Rest = string() | binary() %% State = #xmerl_sax_parser_state{} @@ -439,6 +525,7 @@ check_if_rest_ok(stream, _) -> check_if_rest_ok(_, _) -> false. + %%---------------------------------------------------------------------- %% Function: parse_pi_1(Rest, State) -> Result %% Input: Rest = string() | binary() @@ -883,11 +970,11 @@ send_end_prefix_mapping_event([{Prefix, _Uri} |Ns], State) -> parse_eq(?STRING_EMPTY, State) -> cf(?STRING_EMPTY, State, fun parse_eq/2); parse_eq(?STRING_REST("=", Rest), State) -> - {Rest, State}; + {Rest, State}; parse_eq(?STRING_UNBOUND_REST(C, _) = Bytes, State) when ?is_whitespace(C) -> - {_WS, Rest, State1} = - whitespace(Bytes, State, []), - parse_eq(Rest, State1); + {_WS, Rest, State1} = + whitespace(Bytes, State, []), + parse_eq(Rest, State1); parse_eq(Bytes, State) -> unicode_incomplete_check([Bytes, State, fun parse_eq/2], "expecting = or whitespace"). @@ -905,11 +992,11 @@ parse_eq(Bytes, State) -> parse_att_value(?STRING_EMPTY, State) -> cf(?STRING_EMPTY, State, fun parse_att_value/2); parse_att_value(?STRING_UNBOUND_REST(C, Rest), State) when C == $'; C == $" -> - parse_att_value(Rest, State, C, []); + parse_att_value(Rest, State, C, []); parse_att_value(?STRING_UNBOUND_REST(C, _) = Bytes, State) when ?is_whitespace(C) -> - {_WS, Rest, State1} = - whitespace(Bytes, State, []), - parse_att_value(Rest, State1); + {_WS, Rest, State1} = + whitespace(Bytes, State, []), + parse_att_value(Rest, State1); parse_att_value(Bytes, State) -> unicode_incomplete_check([Bytes, State, fun parse_att_value/2], "\', \" or whitespace expected"). @@ -1360,17 +1447,20 @@ parse_pe_reference_1(Bytes, State, Name) -> %%---------------------------------------------------------------------- -%% Function: insert_reference(Reference, State) -> Result -%% Parameters: Reference = string() +%% Function: insert_reference(Name, Ref, State) -> Result +%% Parameters: Name = string() +%% Ref = {Type, Value} +%% Type = atom() +%% Value = term() %% State = #xmerl_sax_parser_state{} %% Result : %%---------------------------------------------------------------------- -insert_reference({Name, Type, Value}, Table) -> - case ets:lookup(Table, Name) of - [{Name, _, _}] -> - ok; +insert_reference(Name, Value, #xmerl_sax_parser_state{ref_table = Map} = State) -> + case maps:find(Name, Map) of + error -> + State#xmerl_sax_parser_state{ref_table = maps:put(Name, Value, Map)}; _ -> - ets:insert(Table, {Name, Type, Value}) + State end. @@ -1391,8 +1481,8 @@ look_up_reference("apos", _, _) -> look_up_reference("quot", _, _) -> {internal_general, "quot", "\""}; look_up_reference(Name, HaveToExist, State) -> - case ets:lookup(State#xmerl_sax_parser_state.ref_table, Name) of - [{Name, Type, Value}] -> + case maps:find(Name, State#xmerl_sax_parser_state.ref_table) of + {ok, {Type, Value}} -> {Type, Name, Value}; _ -> case HaveToExist of @@ -1474,7 +1564,7 @@ parse_system_litteral(?STRING_EMPTY, State, Stop, Acc) -> parse_system_litteral(?STRING_UNBOUND_REST(Stop, Rest), State, Stop, Acc) -> {lists:reverse(Acc), Rest, State}; parse_system_litteral(?STRING_UNBOUND_REST(C, Rest), State, Stop, Acc) -> - parse_system_litteral(Rest, State, Stop, [C |Acc]); + parse_system_litteral(Rest, State, Stop, [C |Acc]); parse_system_litteral(Bytes, State, Stop, Acc) -> unicode_incomplete_check([Bytes, State, Stop, Acc, fun parse_system_litteral/4], undefined). @@ -1646,9 +1736,11 @@ parse_external_entity(State, _PubId, SysId) -> end_tags = []}, - EventState = handle_external_entity(ExtRef, State1), + {EventState, RefTable} = handle_external_entity(ExtRef, State1), - NewState = event_callback({endEntity, SysId}, SaveState#xmerl_sax_parser_state{event_state=EventState}), + NewState = event_callback({endEntity, SysId}, + SaveState#xmerl_sax_parser_state{event_state=EventState, + ref_table=RefTable}), NewState#xmerl_sax_parser_state{file_type=normal}. @@ -1675,7 +1767,8 @@ handle_external_entity({file, FileToOpen}, State) -> entity=filename:basename(FileToOpen), input_type=file}), ok = file:close(FD), - EntityState#xmerl_sax_parser_state.event_state + {EntityState#xmerl_sax_parser_state.event_state, + EntityState#xmerl_sax_parser_state.ref_table} end; handle_external_entity({http, Url}, State) -> @@ -1695,7 +1788,9 @@ handle_external_entity({http, Url}, State) -> input_type=file}), ok = file:close(FD), ok = file:delete(TmpFile), - EntityState#xmerl_sax_parser_state.event_state + {EntityState#xmerl_sax_parser_state.event_state, + EntityState#xmerl_sax_parser_state.ref_table} + end catch throw:{error, Error} -> @@ -1716,7 +1811,7 @@ handle_external_entity({Tag, _Url}, State) -> parse_external_entity_1(?STRING_EMPTY, #xmerl_sax_parser_state{file_type=Type} = State) -> case catch cf(?STRING_EMPTY, State, fun parse_external_entity_1/2) of {Rest, State1} when is_record(State1, xmerl_sax_parser_state) -> - {Rest, State}; + {Rest, State1}; {fatal_error, {State1, "No more bytes"}} when Type == dtd; Type == entity -> {?STRING_EMPTY, State1}; Other -> @@ -2442,24 +2537,24 @@ parse_entity_def(?STRING_EMPTY, State, Name) -> cf(?STRING_EMPTY, State, Name, fun parse_entity_def/3); parse_entity_def(?STRING_UNBOUND_REST(C, Rest), State, Name) when C == $'; C == $" -> {Value, Rest1, State1} = parse_entity_value(Rest, State, C, []), - insert_reference({Name, internal_general, Value}, State1#xmerl_sax_parser_state.ref_table), - State2 = event_callback({internalEntityDecl, Name, Value}, State1), - {_WS, Rest2, State3} = whitespace(Rest1, State2, []), - parse_def_end(Rest2, State3); + State2 = insert_reference(Name, {internal_general, Value}, State1), + State3 = event_callback({internalEntityDecl, Name, Value}, State2), + {_WS, Rest2, State4} = whitespace(Rest1, State3, []), + parse_def_end(Rest2, State4); parse_entity_def(?STRING_UNBOUND_REST(C, _) = Rest, State, Name) when C == $S; C == $P -> {PubId, SysId, Rest1, State1} = parse_external_id(Rest, State, false), {Ndata, Rest2, State2} = parse_ndata(Rest1, State1), case Ndata of undefined -> - insert_reference({Name, external_general, {PubId, SysId}}, - State2#xmerl_sax_parser_state.ref_table), - State3 = event_callback({externalEntityDecl, Name, PubId, SysId}, State2), - {Rest2, State3}; + State3 = insert_reference(Name, {external_general, {PubId, SysId}}, + State2), + State4 = event_callback({externalEntityDecl, Name, PubId, SysId}, State3), + {Rest2, State4}; _ -> - insert_reference({Name, unparsed, {PubId, SysId, Ndata}}, - State2#xmerl_sax_parser_state.ref_table), - State3 = event_callback({unparsedEntityDecl, Name, PubId, SysId, Ndata}, State2), - {Rest2, State3} + State3 = insert_reference(Name, {unparsed, {PubId, SysId, Ndata}}, + State2), + State4 = event_callback({unparsedEntityDecl, Name, PubId, SysId, Ndata}, State3), + {Rest2, State4} end; parse_entity_def(Bytes, State, Name) -> unicode_incomplete_check([Bytes, State, Name, fun parse_entity_def/3], @@ -2636,19 +2731,19 @@ parse_pe_def(?STRING_EMPTY, State, Name) -> parse_pe_def(?STRING_UNBOUND_REST(C, Rest), State, Name) when C == $'; C == $" -> {Value, Rest1, State1} = parse_entity_value(Rest, State, C, []), Name1 = "%" ++ Name, - insert_reference({Name1, internal_parameter, Value}, - State1#xmerl_sax_parser_state.ref_table), - State2 = event_callback({internalEntityDecl, Name1, Value}, State1), - {_WS, Rest2, State3} = whitespace(Rest1, State2, []), - parse_def_end(Rest2, State3); + State2 = insert_reference(Name1, {internal_parameter, Value}, + State1), + State3 = event_callback({internalEntityDecl, Name1, Value}, State2), + {_WS, Rest2, State4} = whitespace(Rest1, State3, []), + parse_def_end(Rest2, State4); parse_pe_def(?STRING_UNBOUND_REST(C, _) = Bytes, State, Name) when C == $S; C == $P -> {PubId, SysId, Rest1, State1} = parse_external_id(Bytes, State, false), Name1 = "%" ++ Name, - insert_reference({Name1, external_parameter, {PubId, SysId}}, - State1#xmerl_sax_parser_state.ref_table), - State2 = event_callback({externalEntityDecl, Name1, PubId, SysId}, State1), - {_WS, Rest2, State3} = whitespace(Rest1, State2, []), - parse_def_end(Rest2, State3); + State2 = insert_reference(Name1, {external_parameter, {PubId, SysId}}, + State1), + State3 = event_callback({externalEntityDecl, Name1, PubId, SysId}, State2), + {_WS, Rest2, State4} = whitespace(Rest1, State3, []), + parse_def_end(Rest2, State4); parse_pe_def(Bytes, State, Name) -> unicode_incomplete_check([Bytes, State, Name, fun parse_pe_def/3], "\", \', SYSTEM or PUBLIC expected"). diff --git a/lib/xmerl/src/xmerl_sax_simple_dom.erl b/lib/xmerl/src/xmerl_sax_simple_dom.erl index 7eb3afd499..7b15cd92dc 100644 --- a/lib/xmerl/src/xmerl_sax_simple_dom.erl +++ b/lib/xmerl/src/xmerl_sax_simple_dom.erl @@ -2,7 +2,7 @@ %%-------------------------------------------------------------------- %% %CopyrightBegin% %% -%% Copyright Ericsson AB 2009-2016. All Rights Reserved. +%% Copyright Ericsson AB 2009-2017. 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. @@ -129,8 +129,9 @@ build_dom(endDocument, State#xmerl_sax_simple_dom_state{dom=[Decl, {Tag, Attributes, lists:reverse(Content)}]}; _ -> - ?dbg("~p\n", [D]), - ?error("we're not at end the document when endDocument event is encountered.") + %% endDocument is also sent by the parser when a fault occur to tell + %% the event receiver that no more input will be sent + State end; %% Element diff --git a/lib/xmerl/src/xmerl_scan.erl b/lib/xmerl/src/xmerl_scan.erl index 3dd9504f87..a1f6ad4e2c 100644 --- a/lib/xmerl/src/xmerl_scan.erl +++ b/lib/xmerl/src/xmerl_scan.erl @@ -474,8 +474,8 @@ event(_X, S) -> %% into multiple objects (in which case {Acc',Pos',S'} should be returned.) %% If {Acc',S'} is returned, Pos will be incremented by 1 by default. %% Below is an example of an acceptable operation -acc(X = #xmlText{value = Text}, Acc, S) -> - {[X#xmlText{value = Text}|Acc], S}; +acc(#xmlText{value = Text}, [X = #xmlText{value = AccText}], S) -> + {[X#xmlText{value = AccText ++ Text}], S}; acc(X, Acc, S) -> {[X|Acc], S}. diff --git a/lib/xmerl/test/Makefile b/lib/xmerl/test/Makefile index e5c89f84b9..3204f081ba 100644 --- a/lib/xmerl/test/Makefile +++ b/lib/xmerl/test/Makefile @@ -126,5 +126,6 @@ release_tests_spec: opt @tar cfh - xmerl_xsd_MS2002-01-16_SUITE_data | (cd "$(RELSYSDIR)"; tar xf -) @tar cfh - xmerl_xsd_NIST2002-01-16_SUITE_data | (cd "$(RELSYSDIR)"; tar xf -) @tar cfh - xmerl_xsd_Sun2002-01-16_SUITE_data | (cd "$(RELSYSDIR)"; tar xf -) + @tar cfh - xmerl_sax_SUITE_data | (cd "$(RELSYSDIR)"; tar xf -) @tar cfh - xmerl_sax_stream_SUITE_data | (cd "$(RELSYSDIR)"; tar xf -) chmod -R u+w "$(RELSYSDIR)" diff --git a/lib/xmerl/test/xmerl_SUITE.erl b/lib/xmerl/test/xmerl_SUITE.erl index e7e3b3c2b1..66f04f6c25 100644 --- a/lib/xmerl/test/xmerl_SUITE.erl +++ b/lib/xmerl/test/xmerl_SUITE.erl @@ -55,7 +55,7 @@ groups() -> {misc, [], [latin1_alias, syntax_bug1, syntax_bug2, syntax_bug3, pe_ref1, copyright, testXSEIF, export_simple1, export, - default_attrs_bug, xml_ns]}, + default_attrs_bug, xml_ns, scan_splits_string_bug]}, {eventp_tests, [], [sax_parse_and_export]}, {ticket_tests, [], [ticket_5998, ticket_7211, ticket_7214, ticket_7430, @@ -268,6 +268,10 @@ xml_ns(Config) -> } = xmerl_scan:string(Doc2, [{namespace_conformant, true}]), ok. +scan_splits_string_bug(_Config) -> + {#xmlElement{ content = [#xmlText{ value = "Jimmy Zöger" }] }, []} + = xmerl_scan:string("<name>Jimmy Zöger</name>"). + pe_ref1(Config) -> file:set_cwd(datadir(Config)), {#xmlElement{},[]} = xmerl_scan:file(datadir_join(Config,[misc,"PE_ref1.xml"]),[{validation,true}]). @@ -533,8 +537,7 @@ ticket_7430(Config) -> {xmlElement,a,a,[], {xmlNamespace,[],[]}, [],1,[], - [{xmlText,[{a,1}],1,[],"é",text}, - {xmlText,[{a,1}],2,[],"\né",text}], + [{xmlText,[{a,1}],1,[],"é\né",text}], [],_,undeclared} -> ok; _ -> diff --git a/lib/xmerl/test/xmerl_sax_SUITE.erl b/lib/xmerl/test/xmerl_sax_SUITE.erl index eb9cefe0df..68b9bcc4a2 100644 --- a/lib/xmerl/test/xmerl_sax_SUITE.erl +++ b/lib/xmerl/test/xmerl_sax_SUITE.erl @@ -33,7 +33,6 @@ %%====================================================================== %% External functions %%====================================================================== - %%---------------------------------------------------------------------- %% Initializations %%---------------------------------------------------------------------- @@ -42,12 +41,15 @@ all() -> [{group, bugs}]. groups() -> - [{bugs, [], [ticket_8213, ticket_8214, ticket_11551]}]. + [{bugs, [], [ticket_8213, ticket_8214, ticket_11551, + fragmented_xml_directive, + old_dom_event_fun_endDocument_bug, + event_fun_endDocument_error_test, + event_fun_startDocument_error_test]}]. -%%---------------------------------------------------------------------- +%%====================================================================== %% Tests -%%---------------------------------------------------------------------- - +%%====================================================================== %%---------------------------------------------------------------------- %% Test Case %% ID: ticket_8213 @@ -56,7 +58,6 @@ ticket_8213(_Config) -> {ok,ok,[]} = xmerl_sax_parser:stream("<elem/>", [{event_fun, fun (_E,_,_) -> ok end}]), ok. - %%---------------------------------------------------------------------- %% Test Case %% ID: ticket_8214 @@ -99,3 +100,66 @@ ticket_11551(_Config) -> <a>hej</a>">>, {ok, undefined, <<"\n\n<?xml", _/binary>>} = xmerl_sax_parser:stream(Stream3, []), ok. + +%%---------------------------------------------------------------------- +%% Test Case +%% ID: fragmented_xml_directive +%% Test of fragmented xml directive by reading one byte per continuation ca +fragmented_xml_directive(Config) -> + DataDir = proplists:get_value(data_dir, Config), + Name = filename:join(DataDir, "test_data_1.xml"), + {ok, Fd} = file:open(Name, [raw, read,binary]), + Cf = fun cf_fragmented_xml_directive/1, + {ok, undefined, _} = xmerl_sax_parser:stream(<<>>, + [{continuation_fun, Cf}, + {continuation_state, Fd}]), + ok. + +%%---------------------------------------------------------------------- +%% Test Case +%% ID: old_dom_event_fun_endDocument_bug +%% The old_dom backend previous generateded an uncatched exception +%% instead of the correct fatal_error from the parser. +old_dom_event_fun_endDocument_bug(_Config) -> + %% Stream contains bad characters, + {fatal_error, _, _, _, _} = + xmerl_sax_parser:stream([60,63,120,109,108,32,118,101,114,115,105,111,110,61,39,49,46,48,39,32,101,110,99,111,100,105,110,103,61,39,117,116,102,45,56,39,63,62,60, + 99,111,109,109,97,110,100,62,60,104,101,97,100,101,114,62,60,116,114,97,110,115,97,99,116,105,111,110,73,100,62,49,60,47,116,114,97,110, + 115,97,99,116,105,111,110,73,100,62,60,47,104,101,97,100,101,114,62,60,98,111,100,121,62,95,226,130,172,59,60,60,47,98,111,100,121,62,60, + 47,99,111,109,109,97,110,100,62,60,47,120,49,95,49,62], + [{event_fun,fun xmerl_sax_old_dom:event/3}, + {event_state,xmerl_sax_old_dom:initial_state()}]), + ok. + +%%---------------------------------------------------------------------- +%% Test Case +%% ID: event_fun_endDocument_error_test +event_fun_endDocument_error_test(_Config) -> + Stream = <<"<?xml version=\"1.0\" encoding=\"utf-8\"?><a>hej</a>">>, + Ef = fun(endDocument, _ , _) -> throw({event_error, "endDocument error"}); + (_, _, S) -> S + end, + {event_error, _, _, _, _} = xmerl_sax_parser:stream(Stream, [{event_fun, Ef}]), + ok. + +%%---------------------------------------------------------------------- +%% Test Case +%% ID: event_fun_startDocument_error_test +event_fun_startDocument_error_test(_Config) -> + Stream = <<"<?xml version=\"1.0\" encoding=\"utf-8\"?><a>hej</a>">>, + Ef = fun(startDocument, _ , _) -> throw({event_error, "endDocument error"}); + (_, _, S) -> S + end, + {event_error, _, _, _, _} = xmerl_sax_parser:stream(Stream, [{event_fun, Ef}]), + ok. + +%%====================================================================== +%% Internal functions +%%====================================================================== +cf_fragmented_xml_directive(IoDevice) -> + case file:read(IoDevice, 1) of + eof -> + {<<>>, IoDevice}; + {ok, FileBin} -> + {FileBin, IoDevice} + end. diff --git a/lib/xmerl/test/xmerl_sax_SUITE_data/test_data_1.xml b/lib/xmerl/test/xmerl_sax_SUITE_data/test_data_1.xml new file mode 100644 index 0000000000..efbaee6b81 --- /dev/null +++ b/lib/xmerl/test/xmerl_sax_SUITE_data/test_data_1.xml @@ -0,0 +1,4 @@ +<?xml version="1.0" encoding="utf-8"?> +<a> +Hej +</a> diff --git a/lib/xmerl/test/xmerl_sax_stream_SUITE.erl b/lib/xmerl/test/xmerl_sax_stream_SUITE.erl index a306eb66a2..7315f67374 100644 --- a/lib/xmerl/test/xmerl_sax_stream_SUITE.erl +++ b/lib/xmerl/test/xmerl_sax_stream_SUITE.erl @@ -40,7 +40,8 @@ all() -> [ one_document, two_documents, - one_document_and_junk + one_document_and_junk, + end_of_stream ]. %%---------------------------------------------------------------------- @@ -62,6 +63,9 @@ end_per_testcase(_Func, _Config) -> %%---------------------------------------------------------------------- %% Tests %%---------------------------------------------------------------------- + +%%---------------------------------------------------------------------- +%% Send One doc over stream one_document(Config) -> Port = 11111, @@ -107,6 +111,8 @@ one_document(Config) -> end, ok. +%%---------------------------------------------------------------------- +%% Send Two doc over stream two_documents(Config) -> Port = 11111, @@ -152,6 +158,8 @@ two_documents(Config) -> end, ok. +%%---------------------------------------------------------------------- +%% Send one doc and then junk on stream one_document_and_junk(Config) -> Port = 11111, @@ -196,6 +204,13 @@ one_document_and_junk(Config) -> ct:fail("Timeout") end, ok. + +%%---------------------------------------------------------------------- +%% Test of continuation when end of stream +end_of_stream(Config) -> + Stream = <<"<?xml version=\"1.0\" encoding=\"utf-8\"?><a>hej</a>">>, + {ok, undefined, <<>>} = xmerl_sax_parser:stream(Stream, []), + ok. %%---------------------------------------------------------------------- %% Utility functions diff --git a/lib/xmerl/vsn.mk b/lib/xmerl/vsn.mk index 1515a4e37d..2e9c9061d9 100644 --- a/lib/xmerl/vsn.mk +++ b/lib/xmerl/vsn.mk @@ -1 +1 @@ -XMERL_VSN = 1.3.13 +XMERL_VSN = 1.3.15 diff --git a/lib/xmerl/xmerl.pub b/lib/xmerl/xmerl.pub deleted file mode 100644 index 29a81bbde2..0000000000 --- a/lib/xmerl/xmerl.pub +++ /dev/null @@ -1,19 +0,0 @@ -{name, "xmerl"}. -{vsn, {0,18}}. -{summary, "XML processing tools"}. -{author, "Ulf Wiger, Johan Blom, Richard Carlsson, Mickael Remond", "[email protected]", "20020307"}. -{keywords, ["xml", "html"]}. -{needs, [{compiler, "3.0"}]}. -{abstract, - "Implements a set of tools for processing XML documents, as well as " - "working with XML-like structures in Erlang. The main attraction so far " - "is a single-pass, highly customizable XML processor. Other components are " - "an export/translation facility and an XPATH query engine. " - "This version fixes a few bugs in the scanner, and improves HTML export. " - "The latest version can be found at http://sowap.sourceforge.net/ " - "Note that this is still very much a beta product."}. - - - - - |