1berry/ruby
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

* regenc.c (onigenc_minimum_property_name_to_ctype):

Browse Source 
\p{...} should be case insensitive. [ruby-core:33000]

* regenc.c (onigenc_property_list_add_property):
  ditto.

* enc/euc_jp.c (init_property_list, property_name_to_ctype):
  to lowercase property names.

* enc/shift_jis.c (init_property_list, property_name_to_ctype):
  ditto.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@29732 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
This commit is contained in:
naruse 2010-11-09 07:36:33 +00:00
parent 63b0601792
commit dbf7e6f9f2
5 changed files with 43 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
ChangeLog
View File

@ -1,3 +1,17 @@
Tue Nov 9 13:24:33 2010 NARUSE, Yui <naruse@ruby-lang.org>
* regenc.c (onigenc_minimum_property_name_to_ctype):
\p{...} should be case insensitive. [ruby-core:33000]
* regenc.c (onigenc_property_list_add_property):
ditto.
* enc/euc_jp.c (init_property_list, property_name_to_ctype):
to lowercase property names.
* enc/shift_jis.c (init_property_list, property_name_to_ctype):
ditto.
Tue Nov 9 13:29:36 2010 NAKAMURA Usaku <usa@ruby-lang.org> Tue Nov 9 13:29:36 2010 NAKAMURA Usaku <usa@ruby-lang.org>
* win32/win32.c (overlapped_socket_io): get rid of a warning of 64bit * win32/win32.c (overlapped_socket_io): get rid of a warning of 64bit

14
enc/euc_jp.c
View File

@ -274,8 +274,8 @@ init_property_list(void)
{ {
int r; int r;
PROPERTY_LIST_ADD_PROP("Hiragana", CR_Hiragana); PROPERTY_LIST_ADD_PROP("hiragana", CR_Hiragana);
PROPERTY_LIST_ADD_PROP("Katakana", CR_Katakana); PROPERTY_LIST_ADD_PROP("katakana", CR_Katakana);
PropertyInited = 1; PropertyInited = 1;
end: end:
@ -286,11 +286,17 @@ static int
property_name_to_ctype(OnigEncoding enc, UChar* p, UChar* end) property_name_to_ctype(OnigEncoding enc, UChar* p, UChar* end)
{ {
st_data_t ctype; st_data_t ctype;
UChar *s, *e;
PROPERTY_LIST_INIT_CHECK; PROPERTY_LIST_INIT_CHECK;
if (onig_st_lookup_strend(PropertyNameTable, p, end, &ctype) == 0) { s = e = ALLOC_N(UChar, end-p+1);
return onigenc_minimum_property_name_to_ctype(enc, p, end); for (; p < end; p++) {
*e++ = ONIGENC_ASCII_CODE_TO_LOWER_CASE(*p);
}
if (onig_st_lookup_strend(PropertyNameTable, s, e, &ctype) == 0) {
return onigenc_minimum_property_name_to_ctype(enc, s, e);
} }
return ctype; return ctype;

14
enc/shift_jis.c
View File

@ -283,8 +283,8 @@ init_property_list(void)
{ {
int r; int r;
PROPERTY_LIST_ADD_PROP("Hiragana", CR_Hiragana); PROPERTY_LIST_ADD_PROP("hiragana", CR_Hiragana);
PROPERTY_LIST_ADD_PROP("Katakana", CR_Katakana); PROPERTY_LIST_ADD_PROP("katakana", CR_Katakana);
PropertyInited = 1; PropertyInited = 1;
end: end:
@ -295,11 +295,17 @@ static int
property_name_to_ctype(OnigEncoding enc, UChar* p, UChar* end) property_name_to_ctype(OnigEncoding enc, UChar* p, UChar* end)
{ {
hash_data_type ctype; hash_data_type ctype;
UChar *s, *e;
PROPERTY_LIST_INIT_CHECK; PROPERTY_LIST_INIT_CHECK;
if (onig_st_lookup_strend(PropertyNameTable, p, end, &ctype) == 0) { s = e = ALLOC_N(UChar, end-p+1);
return onigenc_minimum_property_name_to_ctype(enc, p, end); for (; p < end; p++) {
*e++ = ONIGENC_ASCII_CODE_TO_LOWER_CASE(*p);
}
if (onig_st_lookup_strend(PropertyNameTable, s, e, &ctype) == 0) {
return onigenc_minimum_property_name_to_ctype(enc, s, e);
} }
return (int)ctype; return (int)ctype;

2
regenc.c
View File

@ -811,7 +811,7 @@ onigenc_minimum_property_name_to_ctype(OnigEncoding enc, UChar* p, UChar* end)
len = onigenc_strlen(enc, p, end); len = onigenc_strlen(enc, p, end);
for (pbe = (pb = PBS) + sizeof(PBS)/sizeof(PBS[0]); pb < pbe; ++pb) { for (pbe = (pb = PBS) + sizeof(PBS)/sizeof(PBS[0]); pb < pbe; ++pb) {
if (len == pb->len && if (len == pb->len &&
onigenc_with_ascii_strncmp(enc, p, end, pb->name, pb->len) == 0) STRNCASECMP(p, pb->name, len) == 0)
return pb->ctype; return pb->ctype;
} }

8
test/ruby/test_m17n.rb
View File

@ -538,6 +538,14 @@ class TestM17N < Test::Unit::TestCase
assert(r.fixed_encoding?) assert(r.fixed_encoding?)
assert_match(r, "\xa4\xa2".force_encoding("euc-jp")) assert_match(r, "\xa4\xa2".force_encoding("euc-jp"))
r = /\p{AsciI}/e
assert(r.fixed_encoding?)
assert_match(r, "a".force_encoding("euc-jp"))
r = /\p{hiraganA}/e
assert(r.fixed_encoding?)
assert_match(r, "\xa4\xa2".force_encoding("euc-jp"))
r = eval('/\u{3042}\p{Hiragana}/'.force_encoding("euc-jp")) r = eval('/\u{3042}\p{Hiragana}/'.force_encoding("euc-jp"))
assert(r.fixed_encoding?) assert(r.fixed_encoding?)
assert_equal(Encoding::UTF_8, r.encoding) assert_equal(Encoding::UTF_8, r.encoding)