ruby-changes:43578
From: duerst <ko1@a...>
Date: Wed, 13 Jul 2016 10:50:22 +0900 (JST)
Subject: [ruby-changes:43578] duerst:r55651 (trunk): * enc/iso_8859_13.c, test/ruby/enc/test_case_comprehensive.rb:
duerst 2016-07-13 10:50:17 +0900 (Wed, 13 Jul 2016) New Revision: 55651 https://svn.ruby-lang.org/cgi-bin/viewvc.cgi?view=revision&revision=55651 Log: * enc/iso_8859_13.c, test/ruby/enc/test_case_comprehensive.rb: Implement non-ASCII case conversion for ISO-8859-13, by Kanon Shindo. Modified files: trunk/ChangeLog trunk/enc/iso_8859_13.c trunk/test/ruby/enc/test_case_comprehensive.rb Index: ChangeLog =================================================================== --- ChangeLog (revision 55650) +++ ChangeLog (revision 55651) @@ -1,3 +1,8 @@ https://github.com/ruby/ruby/blob/trunk/ChangeLog#L1 +Wed Jul 13 10:50:12 2016 Martin Duerst <duerst@i...> + + * enc/iso_8859_13.c, test/ruby/enc/test_case_comprehensive.rb: + Implement non-ASCII case conversion for ISO-8859-13, by Kanon Shindo. + Wed Jul 13 10:31:39 2016 Martin Duerst <duerst@i...> * enc/iso_8859_13.c, enc/windows_1257.c: Split Windows-1257 from Index: enc/iso_8859_13.c =================================================================== --- enc/iso_8859_13.c (revision 55650) +++ enc/iso_8859_13.c (revision 55651) @@ -217,6 +217,55 @@ get_case_fold_codes_by_str(OnigCaseFoldT https://github.com/ruby/ruby/blob/trunk/enc/iso_8859_13.c#L217 flag, p, end, items); } +#ifdef ONIG_CASE_MAPPING +static int +case_map (OnigCaseFoldType* flagP, const OnigUChar** pp, + const OnigUChar* end, OnigUChar* to, OnigUChar* to_end, + const struct OnigEncodingTypeST* enc) +{ + OnigCodePoint code; + OnigUChar *to_start = to; + OnigCaseFoldType flags = *flagP; + + while (*pp<end && to<to_end) { + code = *(*pp)++; + if (code==SHARP_s) { + if (flags&ONIGENC_CASE_UPCASE) { + flags |= ONIGENC_CASE_MODIFIED; + *to++ = 'S'; + code = (flags&ONIGENC_CASE_TITLECASE) ? 's' : 'S'; + } + else if (flags&ONIGENC_CASE_FOLD) { + flags |= ONIGENC_CASE_MODIFIED; + *to++ = 's'; + code = 's'; + } + } + else if ((EncISO_8859_13_CtypeTable[code] & BIT_CTYPE_UPPER) + && (flags & (ONIGENC_CASE_DOWNCASE|ONIGENC_CASE_FOLD))) { + flags |= ONIGENC_CASE_MODIFIED; + code = ENC_ISO_8859_13_TO_LOWER_CASE(code); + } + else if (code==0xB5) ; + else if ((EncISO_8859_13_CtypeTable[code]&BIT_CTYPE_LOWER) + && (flags&ONIGENC_CASE_UPCASE)) { + flags |= ONIGENC_CASE_MODIFIED; + if (code==0xB8 || code==0xBA || code==0xBF) { + code -= 0x10; + } + else { + code -= 0x20; + } + } + *to++ = code; + if (flags&ONIGENC_CASE_TITLECASE) /* switch from titlecase to lowercase for capitalize */ + flags ^= (ONIGENC_CASE_UPCASE|ONIGENC_CASE_DOWNCASE|ONIGENC_CASE_TITLECASE); + } + *flagP = flags; + return (int)(to-to_start); +} +#endif /* ONIG_CASE_MAPPING */ + OnigEncodingDefine(iso_8859_13, ISO_8859_13) = { onigenc_single_byte_mbc_enc_len, "ISO-8859-13", /* name */ @@ -237,7 +286,7 @@ OnigEncodingDefine(iso_8859_13, ISO_8859 https://github.com/ruby/ruby/blob/trunk/enc/iso_8859_13.c#L286 0, ONIGENC_FLAG_NONE, #ifdef ONIG_CASE_MAPPING - onigenc_single_byte_ascii_only_case_map, + case_map, #endif /* ONIG_CASE_MAPPING */ }; ENC_ALIAS("ISO8859-13", "ISO-8859-13") Index: test/ruby/enc/test_case_comprehensive.rb =================================================================== --- test/ruby/enc/test_case_comprehensive.rb (revision 55650) +++ test/ruby/enc/test_case_comprehensive.rb (revision 55651) @@ -257,7 +257,7 @@ class TestComprehensiveCaseFold https://github.com/ruby/ruby/blob/trunk/test/ruby/enc/test_case_comprehensive.rb#L257 generate_ascii_only_case_mapping_tests 'ISO-8859-7' generate_ascii_only_case_mapping_tests 'ISO-8859-9' generate_case_mapping_tests 'ISO-8859-10' - generate_ascii_only_case_mapping_tests 'ISO-8859-13' + generate_case_mapping_tests 'ISO-8859-13' generate_case_mapping_tests 'ISO-8859-14' generate_case_mapping_tests 'ISO-8859-15' generate_case_mapping_tests 'ISO-8859-16' -- ML: ruby-changes@q... Info: http://www.atdot.net/~ko1/quickml/