diff options
author | duerst <duerst@b2dd03c8-39d4-4d8f-98ff-823fe69b080e> | 2016-07-06 09:59:23 +0000 |
---|---|---|
committer | duerst <duerst@b2dd03c8-39d4-4d8f-98ff-823fe69b080e> | 2016-07-06 09:59:23 +0000 |
commit | 6fb97b3f32b491d3f15108f23f90f16f6034248e (patch) | |
tree | c2a4f7078348d08cab90c9d0fa9e29f7127e04cb /enc/iso_8859_15.c | |
parent | 79b1758b65563815abb5d626a21b597e39fcfea4 (diff) | |
download | ruby-6fb97b3f32b491d3f15108f23f90f16f6034248e.tar.gz |
* enc/iso_8859_15.c, test/ruby/enc/test_case_comprehensive.rb:
Implement non-ASCII case conversion for ISO-8859-15, by Maho Harada.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@55591 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Diffstat (limited to 'enc/iso_8859_15.c')
-rw-r--r-- | enc/iso_8859_15.c | 65 |
1 files changed, 64 insertions, 1 deletions
diff --git a/enc/iso_8859_15.c b/enc/iso_8859_15.c index 75adcafbd2..4f33e13f61 100644 --- a/enc/iso_8859_15.c +++ b/enc/iso_8859_15.c @@ -220,6 +220,69 @@ get_case_fold_codes_by_str(OnigCaseFoldType flag, 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 (code==0xAA || code==0xBA) ; + else if ((EncISO_8859_15_CtypeTable[code] & BIT_CTYPE_UPPER) + && (flags & (ONIGENC_CASE_DOWNCASE|ONIGENC_CASE_FOLD))) { + flags |= ONIGENC_CASE_MODIFIED; + if (code==0xA6) + code += 2; + else if (code==0xB4) + code += 4; + else if (code==0xBC) + code += 1; + else if (code==0xBE) + code += 0x41; + else + code += 0x20; + } + else if ((EncISO_8859_15_CtypeTable[code]&BIT_CTYPE_LOWER) + && (flags&ONIGENC_CASE_UPCASE)) { + flags |= ONIGENC_CASE_MODIFIED; + if (code==0xA8) + code -= 2; + else if (code==0xB8) + code -= 4; + else if (code==0xBD) + code -= 1; + else if (code==0xFF) + code -= 0x41; + 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_15, ISO_8859_15) = { onigenc_single_byte_mbc_enc_len, "ISO-8859-15", /* name */ @@ -240,7 +303,7 @@ OnigEncodingDefine(iso_8859_15, ISO_8859_15) = { 0, ONIGENC_FLAG_NONE, #ifdef ONIG_CASE_MAPPING - onigenc_single_byte_ascii_only_case_map, + case_map, #endif /* ONIG_CASE_MAPPING */ }; ENC_ALIAS("ISO8859-15", "ISO-8859-15") |