Age | Commit message (Collapse) | Author |
|
When reading text files with file:read_file/1, assume that the file
has the default encding (epp:default_encoding()) and covert the binary
to string by unicode:characters_to_list(Bin,DefaultEncoding) instead
of binary_to_list(Bin) as before.
|
|
|
|
If the encoding of the source file was utf-8, then the link target
could be faulty encoded in the resulting HTML file. This has been
corrected.
|
|
* Use UTF-8 encoding for all HTML files, except the HTML version of
the test suite generated with erl2html2:convert, which will have the
same encoding as the original test suite (.erl) file.
* Encode link targets in HTML files, allowing both special characters
like "/", "&", "?" etc, and latin1 or unicode characters.
* Use unicode modifier 't' with ~s and ~p when appropriate.
* Use unicode:characters_to_list and unicode:characters_to_binary for
conversion between binaries and strings instead of binary_to_list
and list_to_binary.
|
|
|
|
OTP-10702
Arguments to functions started with the -s option to erl were
converted to atoms with list_to_atom(binary_to_list(Bin)) and thus
were always seen as latin1 encoded.
For other arguments to erl the conversion (to string) would take into
account the file name translation mode (set with +fnl, +fnu or +fna
switch to erl) and try to convert using
unicode:characters_to_list/2. The reason behind this is that it must
be possible to give paths and filenames on the command line and get
the expected encoding.
With the same reasoning, this commit changes the behaviour for
arguments to functions started with the -s option to also take
into account the file name translation mode. The arguments are now
converted to atoms by
list_to_atom(unicode:characters_to_list(Bin,file:native_name_encoding())
|
|
* siri/cuddle-with-tests:
[common_test] Clean logdir for ct_surefire_SUITE
[common_test] Fix log dirs on windows
[test_server] Fix copyright in erl2html2_SUITE.erl
Fix release_handler_SUITE:otp_9395_check_old_code
|
|
ct_surefire_SUITE:logdir/1 creates a directory named "specific_logdir"
which was never cleaned. This commit makes sure the directory exists
and is empty before the test case runs.
|
|
ct_surefire_SUITE erronously assumed that logs would always be located
under priv_dir, but this is not true on windows where logs are located
in temp dir. This has been corrected.
|
|
|
|
This test often fails dues to too high expectations. Don't expect the
test to be more than 1000 times faster with no old code - it just
doesn't happen!
|
|
|
|
* nox/enable-silent-rules-doc:
Document output.mk.in
Fix a typo in erts/lib_src/Makefile.in
|
|
* hb/stdlib/fix_qlc_suite/OTP-10771:
[stdlib] Fix minor issue with qlc_SUITE
|
|
|
|
OTP-10742
OTP-10745
OTP-10749
* hb/unicode/OTP-10302:
[stdlib] Remove documentation of ~tp
Remove one use of iolist_size/1 in io_lib_pretty.erl
Add a new function proc_lib:format/2 which takes encoding
Export the type erl_scan:token()
Make adjustments for Unicode
[stdlib] Change default of erl_scan's unicode option
Update preloaded
Correct recently introduced Unicode related type errors
[stdlib] Introduce new functions epp:read_encoding_from_binary/1,2
Extend char() to Unicode characters
[kernel] Correct bugs in the old shell (user.erl)
[parsetools] Change the encoding of test suites to UTF-8
[stdlib] Fix a contract bug
[stdlib] Update the Unicode examples in STDLIB User's Guide
[stdlib] Remove documentation of Unicode functions in io_lib
|
|
There seems to be a need for a more elaborate solution when it comes
to recognizing printable Unicode characters.
The ~tp syntax is experimental, and may be removed in R16B.
|
|
In order to simplify the testing of Unicode atoms a minor adjustment
has been made in io_lib_pretty. Unicode atoms will not be introduced
until R18 but it may come in handy to be able to pretty-print such
atoms in R16. This update affects ~p only; ~w already works.
|
|
|
|
The type erl_parse:token() used to be a two-tuple, but it can also be
a three-tuple.
|
|
|
|
The value of the undocumented unicode option is very limited. The option
will most likely be removed completely soon.
|
|
|
|
|
|
|
|
The code related to the introduction of unicode_string() and
unicode_char() has been removed. The types char() and string() have
been extended to include Unicode characters.
In fact char() was changed some time ago; this commit is about
cleaning up the documentation and introduce better names for some
functions.
|
|
The inline attribute has been simplified due to problems with
compiling to native code. The exact cause it beyond the scope
of this commit.
|
|
Should have been done in OTP-10630.
|
|
Also minor corrections of the docs of io(3)
|
|
|
|
The documentation of some Unicode functions in io_lib was committed by
mistake.
|
|
* bjorn/asn1/further-cleanup/OTP-10588: (28 commits)
Don't export encode_disp/2 and decode_disp/2 in generated modules
Remove vestiges of support for the {TypeName,Value} notation
Simplify the functions for decoding lengths
per,uper: Optimize decoding of the remaining data types
per,uper: Optimize decoding of the remaining string types
Share all code for dec_gen_prim/3 between per/uper back-ends
per,uper: Optimize decoding of the string data types
testPrimStrings: Test some constraints
By default, encode BIT STRING to bitstrings
Teach encode functions to accept a bitstring term for a BIT STRING
Fix EXTERNAL 1990/1994 conversion information loss
uper: Look up some SizeConstraints at compile-time
Enumeration decoding: Don't emit a default clause if it cannot match
Slightly optimize per encoding of large INTEGERs with constraints
BER run-time: Refactor decoding of string data types
Refactor decoding of BIT STRINGs
Optimize encoding of ENUMERATED in per and uper
Remove the unused run-time modules
eldap: Remove calls to undocumented asn1rt* functions
BER: Correct bug in 'undec_rest'
...
|
|
* anders/diameter/R16A_release/OTP-10608:
vsn -> 1.4
Update appup
Spec fix
Test makefile tweak
Warn about applications/capabilities mismatches in doc
Minor documentation fixes
Remove upgrade code not needed at a major release
Comment fix
Add patch target to makefile
|
|
* ia/ssl/test-case-timeout:
ssl: Too short timeout in test case
|
|
* ia/ssl/dialyzer:
ssl: Enhance dialyzer specs
ssl: Do not return random values
ssl: M-x erlang-indent-buffer
ssl: Remove unnecessary construction of a return value
|
|
* hsv/http_redirect_on_post_303/OTP-10765:
inets: fix autoredirect for POST requests responding 303
inets - httpc_SUITE: add redirect tests for response 303
|
|
* lukas/tools/emacs_add_error_bif/OTP-10774:
Update all bif highlighting for emacs
|
|
* mh/fix-highlighting-dollar-backslash-singlequote/OTP-10766:
Fix syntax highlighting of $\' in Emacs mode
|
|
|
|
* sa/dialyzer-unknown-arity-funs/OTP-10772:
Update one more unused funs result as per ffe582b
Remove pubsub and comm_layer tests from testsuite
Properly support functions with arbitrary arity in type specs.
Conflicts:
lib/hipe/cerl/erl_types.erl
|
|
* ia/public_key/PKCS-10-usage-of-PKCS-9:
public_key: Include some PKCS-9 support needed by PKCS-10
|
|
The single quote in $\' does not begin an atom. Change the regexp to
treat it just like $\". Previously this code would be incorrectly
highlighted:
foo() ->
Quote = $\',
%% This gets coloured as an atom, not as a comment
...
|
|
* kp/odbc-oracle-autocommit-fix/OTP-10735:
Fix aotocommit for Oracle ODBC driver in Linux.
|
|
* bjorn/kernel/undefined-function-handler/OTP-10617:
Add a warning to discourage abuse of $handle_undefined_function/2
|
|
* bjorn/warnings-zero-tolerance:
Turn warnings to errors on selected applications
runtime_tools_sup: Eliminate warning
inet_parse: Eliminate a compiler warning
|
|
Those functions were exported for no good reason.
|
|
Support in BER was removed in 3d1279f3cebfdd2483c3afea9f225613fe45cd00.
|
|
* rickard/r16/port-optimizations/OTP-10336:
Bump fewer vreds per task callback
|
|
|
|
|