written for text mode IO without encoding conversion.
(rb_io_extract_modeenc): binmode requirement changed.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19273 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
ECONV_ENCODER_MASK and ECONV_DECORATOR_MASK.
(ECONV_UNIVERSAL_NEWLINE_DECORATOR): renamed from
ECONV_UNIVERSAL_NEWLINE_DECODER.
(ECONV_CRLF_NEWLINE_DECORATOR): renamed from
ECONV_CRLF_NEWLINE_ENCODER.
(ECONV_CR_NEWLINE_DECORATOR): renamed from ECONV_CR_NEWLINE_ENCODER.
(ECONV_XML_TEXT_DECORATOR): renamed from ECONV_XML_TEXT_ENCODER.
(ECONV_XML_ATTR_CONTENT_DECORATOR): renamed from
ECONV_XML_ATTR_CONTENT_ENCODER.
(ECONV_STATEFUL_DECORATOR_MASK): renamed from
ECONV_STATEFUL_ENCODER_MASK.
(ECONV_XML_ATTR_QUOTE_DECORATOR): renamed from
ECONV_XML_ATTR_CONTENT_DECORATOR.
* io.c: follow the renaming.
* transcode.c: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19271 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Debian's openssl 0.9.8g-13 failed at assert(ssl.session_reused?),
when use default SSLContext. [ruby-dev:36167]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19268 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
(rb_econv_decorate_at_last): declared.
* transcode.c (rb_econv_open_by_transcoder_entries): initialize
replacement_enc. allocate outbuf for the last transcoder.
(rb_econv_open0): extracted from rb_econv_open.
(rb_econv_open): use rb_econv_open0 and decorate the result using
rb_econv_decorate_at_first and rb_econv_decorate_at_last.
(rb_econv_decorate_at): new function.
(rb_econv_decorate_at_first): ditto.
(rb_econv_decorate_at_last): ditto.
(rb_econv_binmode): fix iteration end condition.
(econv_init): don't set source_encoding_name and
destination_encoding_name because they are set in rb_econv_open0.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19262 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
from rb_econv_stateless_encoding to apply stateless ASCII
incompatible encodings such as UTF-16BE.
* io.c (make_writeconv): use rb_econv_asciicompat_encoding.
* transcode_data.h (rb_transcoder_asciicompat_type_t): renamed from
rb_transcoder_stateful_type_t.
(rb_transcoder): use rb_transcoder_asciicompat_type_t.
* transcode.c: follow the type change.
(asciicompat_encoding_i): renamed from stateless_encoding_i.
(rb_econv_asciicompat_encoding): renamed from
rb_econv_stateless_encoding.
(econv_s_asciicompat_encoding): method renamed.
* tool/transcode-tblgen.rb: follow the type change.
* enc/trans/utf_16_32.trans: follow the type change.
rb_from_UTF_16BE to UTF-8 is asciicompat_decoder.
rb_from_UTF_16LE to UTF-8 is asciicompat_decoder.
rb_from_UTF_32BE to UTF-8 is asciicompat_decoder.
rb_from_UTF_32LE to UTF-8 is asciicompat_decoder.
UTF-8 to rb_to_UTF_16BE is asciicompat_encoder.
UTF-8 to rb_to_UTF_16LE is asciicompat_encoder.
UTF-8 to rb_to_UTF_32BE is asciicompat_encoder.
UTF-8 to rb_to_UTF_32LE is asciicompat_encoder.
* enc/trans/newline.trans: follow the type change. universal newline
decoder is asciicompat_converter.
* enc/trans/escape.trans: follow the type change.
* enc/trans/iso2022.trans: ditto.
* enc/trans/japanese.trans: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19249 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
undefined conversion error between iso-2022-jp and the corresponding
stateless encoding.
* enc/emacs_mule.c: replicate emacs-mule as stateless-iso-2022-jp.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19220 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Encoding::Converter.stateless_encoding("html-attr-escaped") should be
nil.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19174 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
(fun_so_escape_html_attr): new function.
(escape_html_attr_finish): new function.
(rb_escape_html_attr): use them to quote the converted result.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19173 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
(econv_primitive_convert): accept a hash as 5th argument as well.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19131 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
it tend to cause security problem. If the behaviour is really
required, ECONV_INVALID_REPLACE with empty string can be used.
For example, CVE-2006-2313, CVE-2008-1036, [ruby-core:15645]
(ECONV_UNDEF_IGNORE): ditto.
* transcode.c (rb_econv_convert): follow the above change.
(econv_opts): ditto.
(Init_transcode): ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19123 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* transcode.c (rb_econv_t): new fields: replacement_str,
replacement_len, replacement_enc and replacement_allocated.
(get_replacement_character): make len as size_t.
(rb_econv_open_by_transcoder_entries): initialize the new fields.
(rb_econv_close): deallocate replacement_str if it allocated.
(make_replacement): new function.
(output_replacement_character): use make_replacement.
(rb_econv_set_replacemenet): defined.
(econv_get_replacement): new method.
(econv_set_replacement): new method.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19108 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Process.euid and egid instead of Etc.getlogin to identify current
user which may be different from a login one (e.g., su command).
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19100 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
eucJP-ms.
* enc/trans/japanese.trans (eucJP-ms): eucJP-ms is the correct
name of the encoding in Ruby.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19021 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
