u 0.5.0 → 1.0.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/build/ext/u/data/attributes.rb +39 -0
- data/build/ext/u/data/bidi-mirroring.rb +27 -0
- data/build/ext/u/data/canonical-combining-class.rb +15 -0
- data/build/ext/u/data/case-folding.rb +39 -0
- data/build/ext/u/data/cased.rb +19 -0
- data/build/ext/u/data/compose.rb +304 -0
- data/build/ext/u/data/constants.rb +31 -0
- data/build/ext/u/data/decompose.rb +85 -0
- data/build/ext/u/data/general-category.rb +61 -0
- data/build/ext/u/data/grapheme-word-break.rb +15 -0
- data/build/ext/u/data/marshalled.rb +5 -0
- data/build/ext/u/data/script.rb +91 -0
- data/build/ext/u/data/soft-dotted.rb +17 -0
- data/build/ext/u/data/title-table.rb +30 -0
- data/build/ext/u/data/wide.rb +17 -0
- data/build/lib/u/build.rb +8 -0
- data/build/lib/u/build/data.rb +16 -0
- data/build/lib/u/build/data/bidimirroring.rb +26 -0
- data/build/lib/u/build/data/break.rb +14 -0
- data/build/lib/u/build/data/casefolding.rb +77 -0
- data/build/lib/u/build/data/compositionexclusions.rb +14 -0
- data/build/lib/u/build/data/derivedeastasianwidth.rb +15 -0
- data/build/lib/u/build/data/file.rb +88 -0
- data/build/lib/u/build/data/linebreak.rb +14 -0
- data/build/lib/u/build/data/proplist.rb +18 -0
- data/build/lib/u/build/data/scripts.rb +22 -0
- data/build/lib/u/build/data/specialcasing.rb +106 -0
- data/build/lib/u/build/data/unicode.rb +41 -0
- data/build/lib/u/build/data/unicode/entry.rb +27 -0
- data/build/lib/u/build/data/unicode/entry/decomposition.rb +29 -0
- data/build/lib/u/build/data/unicode/points.rb +32 -0
- data/build/lib/u/build/header.rb +11 -0
- data/build/lib/u/build/header/table.rb +19 -0
- data/build/lib/u/build/header/table/row.rb +64 -0
- data/build/lib/u/build/header/tables.rb +6 -0
- data/build/lib/u/build/header/tables/intervals.rb +50 -0
- data/build/lib/u/build/header/tables/split.rb +20 -0
- data/build/lib/u/build/header/tables/split/data.rb +29 -0
- data/build/lib/u/build/header/tables/split/part1.rb +28 -0
- data/build/lib/u/build/header/tables/split/part2.rb +13 -0
- data/build/lib/u/build/header/tables/split/row.rb +34 -0
- data/build/lib/u/build/header/tables/split/rows.rb +22 -0
- data/build/test/unit/break.rb +45 -0
- data/build/test/unit/case.rb +178 -0
- data/build/test/unit/foldcase.rb +44 -0
- data/build/test/unit/normalize.rb +81 -0
- data/ext/u/attributes.c +62 -0
- data/ext/u/attributes.h +5 -0
- data/ext/u/case.h +41 -0
- data/ext/u/data/attributes.h +3070 -0
- data/ext/u/data/bidi-mirroring.h +373 -0
- data/ext/u/data/canonical-combining-class.h +2157 -0
- data/ext/u/data/case-folding.h +171 -0
- data/ext/u/data/cased.h +42 -0
- data/ext/u/data/compose.h +1714 -0
- data/ext/u/data/constants.h +17 -0
- data/ext/u/data/decompose.h +9356 -0
- data/ext/u/data/general-category.h +28959 -0
- data/ext/u/data/grapheme-break.h +13201 -0
- data/ext/u/data/line-break.h +26501 -0
- data/ext/u/data/normalization-quick-check.h +3002 -0
- data/ext/u/data/script.h +2928 -0
- data/ext/u/data/soft-dotted.h +55 -0
- data/ext/u/data/title-table.h +41 -0
- data/ext/u/data/types.h +11117 -0
- data/ext/u/data/wide-cjk.h +197 -0
- data/ext/u/data/wide.h +59 -0
- data/ext/u/data/word-break.h +10001 -0
- data/ext/u/depend +281 -0
- data/ext/u/extconf.rb +158 -0
- data/ext/u/output.h +51 -0
- data/ext/{encoding/character/utf-8 → u}/private.c +11 -15
- data/ext/u/private.h +58 -0
- data/ext/u/rb_includes.h +10 -0
- data/ext/u/rb_private.c +98 -0
- data/ext/u/rb_private.h +67 -0
- data/ext/u/rb_u.c +251 -0
- data/ext/u/rb_u_buffer.c +443 -0
- data/ext/u/rb_u_buffer.h +24 -0
- data/ext/u/rb_u_re.c +43 -0
- data/ext/u/rb_u_re.h +15 -0
- data/ext/u/rb_u_string.c +478 -0
- data/ext/u/rb_u_string.h +173 -0
- data/ext/u/rb_u_string_alnum.c +10 -0
- data/ext/u/rb_u_string_alpha.c +10 -0
- data/ext/u/rb_u_string_aref.c +142 -0
- data/ext/u/rb_u_string_ascii_only.c +13 -0
- data/ext/u/rb_u_string_assigned.c +10 -0
- data/ext/u/rb_u_string_b.c +18 -0
- data/ext/u/rb_u_string_bytesize.c +10 -0
- data/ext/u/rb_u_string_byteslice.c +103 -0
- data/ext/u/rb_u_string_canonical_combining_class.c +33 -0
- data/ext/u/rb_u_string_case_ignorable.c +25 -0
- data/ext/u/rb_u_string_casecmp.c +61 -0
- data/ext/u/rb_u_string_cased.c +17 -0
- data/ext/u/rb_u_string_chomp.c +107 -0
- data/ext/u/rb_u_string_chop.c +33 -0
- data/ext/u/rb_u_string_chr.c +9 -0
- data/ext/u/rb_u_string_cntrl.c +10 -0
- data/ext/u/rb_u_string_collate.c +46 -0
- data/ext/u/rb_u_string_collation_key.c +18 -0
- data/ext/u/rb_u_string_count.c +38 -0
- data/ext/u/rb_u_string_defined.c +10 -0
- data/ext/u/rb_u_string_delete.c +62 -0
- data/ext/u/rb_u_string_digit.c +10 -0
- data/ext/u/rb_u_string_downcase.c +13 -0
- data/ext/u/rb_u_string_dump.c +153 -0
- data/ext/u/rb_u_string_each_byte.c +46 -0
- data/ext/u/rb_u_string_each_char.c +49 -0
- data/ext/u/rb_u_string_each_codepoint.c +45 -0
- data/ext/u/rb_u_string_each_grapheme_cluster.c +36 -0
- data/ext/u/rb_u_string_each_line.c +142 -0
- data/ext/u/rb_u_string_each_word.c +34 -0
- data/ext/u/rb_u_string_empty.c +11 -0
- data/ext/u/rb_u_string_end_with.c +31 -0
- data/ext/u/rb_u_string_eql.c +30 -0
- data/ext/u/rb_u_string_equal.c +33 -0
- data/ext/u/rb_u_string_foldcase.c +12 -0
- data/ext/u/rb_u_string_folded.c +13 -0
- data/ext/u/rb_u_string_format.c +1745 -0
- data/ext/u/rb_u_string_general_category.c +109 -0
- data/ext/u/rb_u_string_getbyte.c +21 -0
- data/ext/u/rb_u_string_graph.c +21 -0
- data/ext/u/rb_u_string_grapheme_break.c +61 -0
- data/ext/u/rb_u_string_gsub.c +164 -0
- data/ext/u/rb_u_string_hash.c +10 -0
- data/ext/u/rb_u_string_hex.c +9 -0
- data/ext/u/rb_u_string_include.c +10 -0
- data/ext/u/rb_u_string_index.c +110 -0
- data/ext/u/rb_u_string_inspect.c +189 -0
- data/ext/u/rb_u_string_internal_tr.c +148 -0
- data/ext/u/rb_u_string_internal_tr.h +29 -0
- data/ext/u/rb_u_string_justify.c +169 -0
- data/ext/u/rb_u_string_length.c +10 -0
- data/ext/u/rb_u_string_line_break.c +115 -0
- data/ext/u/rb_u_string_lower.c +13 -0
- data/ext/u/rb_u_string_lstrip.c +24 -0
- data/ext/u/rb_u_string_match.c +65 -0
- data/ext/u/rb_u_string_mirror.c +16 -0
- data/ext/u/rb_u_string_newline.c +21 -0
- data/ext/u/rb_u_string_normalize.c +70 -0
- data/ext/u/rb_u_string_normalized.c +28 -0
- data/ext/u/rb_u_string_oct.c +11 -0
- data/ext/u/rb_u_string_ord.c +14 -0
- data/ext/u/rb_u_string_partition.c +80 -0
- data/ext/u/rb_u_string_plus.c +33 -0
- data/ext/u/rb_u_string_print.c +10 -0
- data/ext/u/rb_u_string_punct.c +10 -0
- data/ext/u/rb_u_string_reverse.c +13 -0
- data/ext/u/rb_u_string_rindex.c +104 -0
- data/ext/u/rb_u_string_rpartition.c +81 -0
- data/ext/u/rb_u_string_rstrip.c +29 -0
- data/ext/u/rb_u_string_scan.c +109 -0
- data/ext/u/rb_u_string_script.c +253 -0
- data/ext/u/rb_u_string_soft_dotted.c +13 -0
- data/ext/u/rb_u_string_space.c +24 -0
- data/ext/u/rb_u_string_split.c +245 -0
- data/ext/u/rb_u_string_squeeze.c +75 -0
- data/ext/u/rb_u_string_start_with.c +31 -0
- data/ext/u/rb_u_string_strip.c +36 -0
- data/ext/u/rb_u_string_sub.c +147 -0
- data/ext/u/rb_u_string_times.c +35 -0
- data/ext/u/rb_u_string_title.c +10 -0
- data/ext/u/rb_u_string_titlecase.c +13 -0
- data/ext/u/rb_u_string_to_i.c +45 -0
- data/ext/u/rb_u_string_to_inum.c +364 -0
- data/ext/u/rb_u_string_to_inum.h +1 -0
- data/ext/u/rb_u_string_to_str.c +17 -0
- data/ext/u/rb_u_string_to_sym.c +12 -0
- data/ext/u/rb_u_string_tr.c +290 -0
- data/ext/u/rb_u_string_upcase.c +12 -0
- data/ext/u/rb_u_string_upper.c +13 -0
- data/ext/u/rb_u_string_valid.c +10 -0
- data/ext/u/rb_u_string_valid_encoding.c +12 -0
- data/ext/u/rb_u_string_wide.c +21 -0
- data/ext/u/rb_u_string_wide_cjk.c +21 -0
- data/ext/u/rb_u_string_width.c +19 -0
- data/ext/u/rb_u_string_word_break.c +63 -0
- data/ext/u/rb_u_string_xdigit.c +22 -0
- data/ext/u/rb_u_string_zero_width.c +16 -0
- data/ext/u/titled.c +55 -0
- data/ext/u/titled.h +1 -0
- data/ext/u/u.c +23 -0
- data/ext/u/u.h +458 -0
- data/ext/u/u_char_canonical_combining_class.c +31 -0
- data/ext/u/u_char_digit_value.c +21 -0
- data/ext/u/u_char_downcase.c +27 -0
- data/ext/u/u_char_general_category.c +31 -0
- data/ext/u/u_char_grapheme_break.c +28 -0
- data/ext/u/u_char_isalnum.c +24 -0
- data/ext/u/u_char_isalpha.c +21 -0
- data/ext/u/u_char_isassigned.c +16 -0
- data/ext/u/u_char_iscased.c +22 -0
- data/ext/u/u_char_iscaseignorable.c +29 -0
- data/ext/u/u_char_iscntrl.c +17 -0
- data/ext/u/u_char_isdefined.c +15 -0
- data/ext/u/u_char_isdigit.c +16 -0
- data/ext/u/u_char_isgraph.c +22 -0
- data/ext/u/u_char_islower.c +16 -0
- data/ext/u/u_char_isnewline.c +24 -0
- data/ext/u/u_char_isprint.c +21 -0
- data/ext/u/u_char_ispunct.c +27 -0
- data/ext/u/u_char_issoftdotted.c +18 -0
- data/ext/u/u_char_isspace.c +28 -0
- data/ext/u/u_char_isupper.c +16 -0
- data/ext/u/u_char_isvalid.c +18 -0
- data/ext/u/u_char_iswide.c +18 -0
- data/ext/u/u_char_iswide_cjk.c +22 -0
- data/ext/u/u_char_isxdigit.c +27 -0
- data/ext/u/u_char_iszerowidth.c +29 -0
- data/ext/u/u_char_line_break.c +29 -0
- data/ext/u/u_char_mirror.c +16 -0
- data/ext/u/u_char_normalized.c +23 -0
- data/ext/u/u_char_script.c +41 -0
- data/ext/u/u_char_to_u.c +48 -0
- data/ext/u/u_char_upcase.c +24 -0
- data/ext/u/u_char_width.c +12 -0
- data/ext/u/u_char_word_break.c +28 -0
- data/ext/u/u_char_xdigit_value.c +31 -0
- data/ext/u/u_collate.c +83 -0
- data/ext/u/u_collation_key.c +132 -0
- data/ext/u/u_decode.c +156 -0
- data/ext/u/u_downcase.c +201 -0
- data/ext/u/u_foldcase.c +68 -0
- data/ext/u/u_grapheme_clusters.c +57 -0
- data/ext/u/u_has_prefix.c +27 -0
- data/ext/u/u_index.c +93 -0
- data/ext/u/u_is_ascii_only.c +33 -0
- data/ext/u/u_locale.c +40 -0
- data/ext/u/u_locale.h +14 -0
- data/ext/u/u_mirror.c +20 -0
- data/ext/u/u_n_bytes.c +16 -0
- data/ext/u/u_n_chars.c +43 -0
- data/ext/u/u_normalize.c +232 -0
- data/ext/u/u_normalized.c +28 -0
- data/ext/u/u_offset_to_pointer.c +62 -0
- data/ext/u/u_pointer_to_offset.c +23 -0
- data/ext/u/u_recode.c +73 -0
- data/ext/u/u_reverse.c +21 -0
- data/ext/u/u_rindex.c +132 -0
- data/ext/u/u_titlecase.c +68 -0
- data/ext/u/u_upcase.c +89 -0
- data/ext/u/u_width.c +35 -0
- data/ext/u/u_words.c +82 -0
- data/ext/u/yield.h +27 -0
- data/lib/u-1.0.rb +20 -0
- data/lib/u-1.0/buffer.rb +10 -0
- data/lib/u-1.0/string.rb +9 -0
- data/lib/u-1.0/version.rb +287 -0
- data/test/unit/case.rb +2080 -0
- data/test/unit/foldcase.rb +1136 -0
- data/test/unit/graphemebreak.rb +407 -0
- data/test/unit/normalize.rb +367545 -0
- data/test/unit/u-1.0.rb +10 -0
- data/test/unit/u-1.0/buffer.rb +52 -0
- data/test/unit/u-1.0/string.rb +1439 -0
- data/test/unit/{u.rb → u-1.0/version.rb} +0 -1
- data/test/unit/wordbreak.rb +1083 -0
- metadata +603 -148
- data/README +0 -38
- data/Rakefile +0 -64
- data/ext/encoding/character/utf-8/break.c +0 -25
- data/ext/encoding/character/utf-8/data/break.h +0 -22931
- data/ext/encoding/character/utf-8/data/character-tables.h +0 -14358
- data/ext/encoding/character/utf-8/data/compose.h +0 -1607
- data/ext/encoding/character/utf-8/data/decompose.h +0 -10926
- data/ext/encoding/character/utf-8/data/generate-unicode-data.rb +0 -1070
- data/ext/encoding/character/utf-8/decompose.c +0 -444
- data/ext/encoding/character/utf-8/depend +0 -65
- data/ext/encoding/character/utf-8/extconf.rb +0 -67
- data/ext/encoding/character/utf-8/private.h +0 -51
- data/ext/encoding/character/utf-8/properties.c +0 -1056
- data/ext/encoding/character/utf-8/rb_includes.h +0 -19
- data/ext/encoding/character/utf-8/rb_methods.h +0 -49
- data/ext/encoding/character/utf-8/rb_private.h +0 -52
- data/ext/encoding/character/utf-8/rb_utf_aref.c +0 -111
- data/ext/encoding/character/utf-8/rb_utf_aset.c +0 -105
- data/ext/encoding/character/utf-8/rb_utf_casecmp.c +0 -24
- data/ext/encoding/character/utf-8/rb_utf_chomp.c +0 -114
- data/ext/encoding/character/utf-8/rb_utf_chop.c +0 -44
- data/ext/encoding/character/utf-8/rb_utf_collate.c +0 -13
- data/ext/encoding/character/utf-8/rb_utf_count.c +0 -30
- data/ext/encoding/character/utf-8/rb_utf_delete.c +0 -60
- data/ext/encoding/character/utf-8/rb_utf_downcase.c +0 -13
- data/ext/encoding/character/utf-8/rb_utf_each_char.c +0 -27
- data/ext/encoding/character/utf-8/rb_utf_foldcase.c +0 -13
- data/ext/encoding/character/utf-8/rb_utf_hex.c +0 -14
- data/ext/encoding/character/utf-8/rb_utf_index.c +0 -50
- data/ext/encoding/character/utf-8/rb_utf_insert.c +0 -48
- data/ext/encoding/character/utf-8/rb_utf_internal_bignum.c +0 -332
- data/ext/encoding/character/utf-8/rb_utf_internal_bignum.h +0 -12
- data/ext/encoding/character/utf-8/rb_utf_internal_tr.c +0 -142
- data/ext/encoding/character/utf-8/rb_utf_internal_tr.h +0 -41
- data/ext/encoding/character/utf-8/rb_utf_justify.c +0 -96
- data/ext/encoding/character/utf-8/rb_utf_length.c +0 -14
- data/ext/encoding/character/utf-8/rb_utf_lstrip.c +0 -41
- data/ext/encoding/character/utf-8/rb_utf_normalize.c +0 -51
- data/ext/encoding/character/utf-8/rb_utf_oct.c +0 -14
- data/ext/encoding/character/utf-8/rb_utf_reverse.c +0 -13
- data/ext/encoding/character/utf-8/rb_utf_rindex.c +0 -88
- data/ext/encoding/character/utf-8/rb_utf_rstrip.c +0 -51
- data/ext/encoding/character/utf-8/rb_utf_squeeze.c +0 -70
- data/ext/encoding/character/utf-8/rb_utf_strip.c +0 -27
- data/ext/encoding/character/utf-8/rb_utf_to_i.c +0 -25
- data/ext/encoding/character/utf-8/rb_utf_tr.c +0 -250
- data/ext/encoding/character/utf-8/rb_utf_upcase.c +0 -13
- data/ext/encoding/character/utf-8/tables.h +0 -38
- data/ext/encoding/character/utf-8/unicode.c +0 -319
- data/ext/encoding/character/utf-8/unicode.h +0 -216
- data/ext/encoding/character/utf-8/utf.c +0 -1334
- data/lib/encoding/character/utf-8.rb +0 -201
- data/lib/u.rb +0 -16
- data/lib/u/string.rb +0 -185
- data/lib/u/version.rb +0 -5
- data/test/unit/u/string.rb +0 -91
@@ -0,0 +1,132 @@
|
|
1
|
+
#include "extconf.h"
|
2
|
+
#include <assert.h>
|
3
|
+
#include <errno.h>
|
4
|
+
#define __USE_XOPEN2K8 1
|
5
|
+
#include <locale.h>
|
6
|
+
#ifdef HAVE_XLOCALE_H
|
7
|
+
# include <xlocale.h>
|
8
|
+
#endif
|
9
|
+
#define __USE_XOPEN2K 1
|
10
|
+
#include <langinfo.h>
|
11
|
+
#include <stdbool.h>
|
12
|
+
#include <stdint.h>
|
13
|
+
#include <stdlib.h>
|
14
|
+
#include <string.h>
|
15
|
+
#ifndef HAVE_STRXFRM_L
|
16
|
+
static inline size_t
|
17
|
+
strxfrm_l(char *restrict s1, const char *restrict s2, size_t n,
|
18
|
+
UNUSED(locale_t loc))
|
19
|
+
{
|
20
|
+
return strxfrm(s1, s2, n);
|
21
|
+
}
|
22
|
+
#endif
|
23
|
+
#ifndef HAVE_NL_LANGINFO_L
|
24
|
+
static inline char *
|
25
|
+
nl_langinfo_l(nl_item item, UNUSED(locale_t loc))
|
26
|
+
{
|
27
|
+
return nl_langinfo(item);
|
28
|
+
}
|
29
|
+
#endif
|
30
|
+
|
31
|
+
#include "u.h"
|
32
|
+
#include "private.h"
|
33
|
+
|
34
|
+
static inline const char *
|
35
|
+
codeset(locale_t locale)
|
36
|
+
{
|
37
|
+
return locale == NULL ?
|
38
|
+
nl_langinfo(CODESET) :
|
39
|
+
nl_langinfo_l(CODESET, locale);
|
40
|
+
}
|
41
|
+
|
42
|
+
static inline size_t
|
43
|
+
transform(char *result, const char *string, size_t n, locale_t locale)
|
44
|
+
{
|
45
|
+
return locale == NULL ?
|
46
|
+
strxfrm(result, string, n) :
|
47
|
+
strxfrm_l(result, string, n, locale);
|
48
|
+
}
|
49
|
+
|
50
|
+
static size_t
|
51
|
+
ckey(char *result, size_t m, const char *string, size_t n, locale_t locale)
|
52
|
+
{
|
53
|
+
char saved_sentinel = string[n];
|
54
|
+
((char *)string)[n] = '\0';
|
55
|
+
size_t l = 0;
|
56
|
+
const char *p = string;
|
57
|
+
const char *end = string + n + 1;
|
58
|
+
while (true) {
|
59
|
+
errno = 0;
|
60
|
+
size_t k = m > l ?
|
61
|
+
transform(result + l, p, m - l, locale) :
|
62
|
+
transform(NULL, p, 0, locale);
|
63
|
+
if (errno != 0)
|
64
|
+
break;
|
65
|
+
l += k;
|
66
|
+
p += strlen(p) + 1;
|
67
|
+
if (p == end)
|
68
|
+
break;
|
69
|
+
if (m > l)
|
70
|
+
result[l] = '\0';
|
71
|
+
l++;
|
72
|
+
}
|
73
|
+
((char *)string)[n] = saved_sentinel;
|
74
|
+
return l;
|
75
|
+
}
|
76
|
+
|
77
|
+
static size_t
|
78
|
+
recode_ckey(char *result, size_t m, const char *string, size_t n,
|
79
|
+
locale_t locale, const char *cs)
|
80
|
+
{
|
81
|
+
char buf[2048];
|
82
|
+
errno = 0;
|
83
|
+
size_t n_recoded = u_recode(buf, sizeof(buf), string, n, cs);
|
84
|
+
if (errno != 0)
|
85
|
+
return 0;
|
86
|
+
if (n_recoded < sizeof(buf))
|
87
|
+
return ckey(result, m, buf, n_recoded, locale);
|
88
|
+
char *recoded = malloc(n_recoded + 1);
|
89
|
+
if (recoded == NULL)
|
90
|
+
return 0;
|
91
|
+
u_recode(recoded, n_recoded + 1, string, n, cs);
|
92
|
+
size_t n_key = ckey(result, m, recoded, n_recoded, locale);
|
93
|
+
free(recoded);
|
94
|
+
return n_key;
|
95
|
+
}
|
96
|
+
|
97
|
+
size_t
|
98
|
+
u_collation_key(char *result, size_t m, const char *string, size_t n,
|
99
|
+
const char *locale)
|
100
|
+
{
|
101
|
+
assert(string != NULL);
|
102
|
+
assert(result != NULL || m == 0);
|
103
|
+
locale_t l = NULL;
|
104
|
+
if (locale != NULL)
|
105
|
+
l = newlocale(LC_COLLATE_MASK | LC_CTYPE_MASK, locale, NULL);
|
106
|
+
const char *cs = codeset(l);
|
107
|
+
size_t r = strcmp(cs, "UTF-8") != 0 ?
|
108
|
+
recode_ckey(result, m, string, n, l, cs) :
|
109
|
+
ckey(result, m, string, n, l);
|
110
|
+
if (l != NULL)
|
111
|
+
freelocale(l);
|
112
|
+
return r;
|
113
|
+
}
|
114
|
+
|
115
|
+
size_t
|
116
|
+
u_normalized_collation_key(char *result, size_t m, const char *string, size_t n,
|
117
|
+
const char *locale)
|
118
|
+
{
|
119
|
+
assert(string != NULL);
|
120
|
+
assert(result != NULL || m == 0);
|
121
|
+
char buf[2048];
|
122
|
+
size_t n_normalized = u_normalize(buf, sizeof(buf), string, n, U_NORMALIZATION_FORM_KC);
|
123
|
+
if (n_normalized < sizeof(buf))
|
124
|
+
return u_collation_key(result, m, buf, n_normalized, locale);
|
125
|
+
char *normalized = malloc(n_normalized + 1);
|
126
|
+
if (normalized == NULL)
|
127
|
+
return 0;
|
128
|
+
u_normalize(normalized, n_normalized + 1, string, n, U_NORMALIZATION_FORM_KC);
|
129
|
+
size_t n_key = u_collation_key(result, m, normalized, n_normalized, locale);
|
130
|
+
free(normalized);
|
131
|
+
return n_key;
|
132
|
+
}
|
data/ext/u/u_decode.c
ADDED
@@ -0,0 +1,156 @@
|
|
1
|
+
#include <assert.h>
|
2
|
+
#include <stddef.h>
|
3
|
+
#include <stdint.h>
|
4
|
+
#include <stdbool.h>
|
5
|
+
|
6
|
+
#include "u.h"
|
7
|
+
#include "private.h"
|
8
|
+
|
9
|
+
|
10
|
+
// The dfa table and decode function is © 2008–2010 Björn Höhrmann
|
11
|
+
// <bjoern@hoehrmann.de>. See http://bjoern.hoehrmann.de/utf-8/decoder/dfa/
|
12
|
+
// for details.
|
13
|
+
|
14
|
+
enum {
|
15
|
+
ACCEPT = 0,
|
16
|
+
REJECT = 12
|
17
|
+
};
|
18
|
+
|
19
|
+
static const uint8_t dfa[] = {
|
20
|
+
// The first part of the table maps bytes to character classes to
|
21
|
+
// reduce the size of the transition table and create bitmasks.
|
22
|
+
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
23
|
+
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
24
|
+
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
25
|
+
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
26
|
+
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,
|
27
|
+
7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,
|
28
|
+
8,8,2,2,2,2,2,2,2,2,2,2,2,2,2,2, 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,
|
29
|
+
10,3,3,3,3,3,3,3,3,3,3,3,3,4,3,3, 11,6,6,6,5,8,8,8,8,8,8,8,8,8,8,8,
|
30
|
+
|
31
|
+
// The second part is a transition table that maps a combination
|
32
|
+
// of a state of the automaton and a character class to a state.
|
33
|
+
0,12,24,36,60,96,84,12,12,12,48,72, 12,12,12,12,12,12,12,12,12,12,12,12,
|
34
|
+
12, 0,12,12,12,12,12, 0,12, 0,12,12, 12,24,12,12,12,12,12,24,12,24,12,12,
|
35
|
+
12,12,12,12,12,12,12,24,12,12,12,12, 12,24,12,12,12,12,12,12,12,24,12,12,
|
36
|
+
12,12,12,12,12,12,12,36,12,36,12,12, 12,36,12,12,12,12,12,36,12,36,12,12,
|
37
|
+
12,36,12,12,12,12,12,12,12,12,12,12,
|
38
|
+
};
|
39
|
+
|
40
|
+
// This reversal of ‹dfa› is © 2013 Nikolai Weibull.
|
41
|
+
static const uint8_t dfa_r[] = {
|
42
|
+
// The first part of the table maps bytes to character classes to
|
43
|
+
// reduce the size of the transition table and create bitmasks.
|
44
|
+
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
45
|
+
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
46
|
+
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
47
|
+
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
48
|
+
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,
|
49
|
+
7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,
|
50
|
+
8,8,2,2,2,2,2,2,2,2,2,2,2,2,2,2, 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,
|
51
|
+
10,3,3,3,3,3,3,3,3,3,3,3,3,4,3,3, 11,6,6,6,5,8,8,8,8,8,8,8,8,8,8,8,
|
52
|
+
|
53
|
+
// The second part is a transition table that maps a combination
|
54
|
+
// of a state of the automaton and a character class to a state.
|
55
|
+
0,24,12,12,12,12,12,24,12,24,12,12, 12,12,12,12,12,12,12,12,12,12,12,12,
|
56
|
+
12,36, 0,12,12,12,12,48,12,36,12,12, 12,60,12, 0, 0,12,12,72,12,72,12,12,
|
57
|
+
12,60,12, 0,12,12,12,72,12,72, 0,12, 12,12,12,12,12, 0, 0,12,12,12,12,12,
|
58
|
+
12,12,12,12,12,12, 0,12,12,12,12, 0,
|
59
|
+
};
|
60
|
+
|
61
|
+
static inline uint32_t
|
62
|
+
decode(uint32_t *state, uint32_t *c, uint32_t b)
|
63
|
+
{
|
64
|
+
uint32_t type = dfa[b];
|
65
|
+
*c = *state != ACCEPT ? (*c << 6) | (b & 0x3f) : (0xff >> type) & b;
|
66
|
+
*state = dfa[256 + *state + type];
|
67
|
+
return *state;
|
68
|
+
}
|
69
|
+
|
70
|
+
#define REPLACEMENT_CHARACTER ((uint32_t)0xfffd)
|
71
|
+
|
72
|
+
uint32_t
|
73
|
+
u_decode(const char **q, const char *u, const char *end)
|
74
|
+
{
|
75
|
+
assert(u < end);
|
76
|
+
uint32_t c, state = ACCEPT;
|
77
|
+
const unsigned char *p;
|
78
|
+
for (p = (const unsigned char *)u; p < (const unsigned char *)end; p++)
|
79
|
+
switch (decode(&state, &c, *p)) {
|
80
|
+
case ACCEPT:
|
81
|
+
*q = (const char *)p + 1;
|
82
|
+
return c;
|
83
|
+
case REJECT:
|
84
|
+
*q = (const char *)p + 1;
|
85
|
+
return REPLACEMENT_CHARACTER;
|
86
|
+
}
|
87
|
+
*q = (const char *)p;
|
88
|
+
return REPLACEMENT_CHARACTER;
|
89
|
+
}
|
90
|
+
|
91
|
+
int
|
92
|
+
u_decode_n(uint32_t *result, const char *u, size_t n)
|
93
|
+
{
|
94
|
+
const char *q;
|
95
|
+
*result = u_decode(&q, u, u + n);
|
96
|
+
return (int)(q - u);
|
97
|
+
}
|
98
|
+
|
99
|
+
static inline uint32_t
|
100
|
+
decode_r(uint32_t *state, uint32_t *c, uint32_t b, int i)
|
101
|
+
{
|
102
|
+
uint32_t type = dfa_r[b];
|
103
|
+
*state = dfa_r[256 + *state + type];
|
104
|
+
*c |= (*state != ACCEPT ? b & 0x3f : (0xff >> type) & b) << (6*i);
|
105
|
+
return *state;
|
106
|
+
}
|
107
|
+
|
108
|
+
uint32_t
|
109
|
+
u_decode_r(const char **p, const char *begin, const char *u)
|
110
|
+
{
|
111
|
+
assert(begin < u);
|
112
|
+
uint32_t c = 0, state = ACCEPT;
|
113
|
+
int i = 0;
|
114
|
+
const unsigned char *q;
|
115
|
+
for (q = (const unsigned char *)u - 1; (const unsigned char *)begin <= q; q--, i++)
|
116
|
+
switch (decode_r(&state, &c, *q, i)) {
|
117
|
+
case ACCEPT:
|
118
|
+
*p = (const char *)q;
|
119
|
+
return c;
|
120
|
+
case REJECT:
|
121
|
+
*p = (const char *)q;
|
122
|
+
return REPLACEMENT_CHARACTER;
|
123
|
+
}
|
124
|
+
*p = (const char *)begin;
|
125
|
+
return REPLACEMENT_CHARACTER;
|
126
|
+
}
|
127
|
+
|
128
|
+
static inline uint32_t
|
129
|
+
validate(uint32_t *state, uint32_t b)
|
130
|
+
{
|
131
|
+
uint32_t type = dfa[b];
|
132
|
+
return *state = dfa[256 + *state + type];
|
133
|
+
}
|
134
|
+
|
135
|
+
bool
|
136
|
+
u_valid(const char *u, size_t n, const char **end)
|
137
|
+
{
|
138
|
+
uint32_t state = ACCEPT;
|
139
|
+
const unsigned char *p = (const unsigned char *)u;
|
140
|
+
const unsigned char *o = p;
|
141
|
+
const unsigned char *q = p + n;
|
142
|
+
for ( ; p < q; p++)
|
143
|
+
switch (validate(&state, *p)) {
|
144
|
+
case ACCEPT:
|
145
|
+
o = p;
|
146
|
+
break;
|
147
|
+
case REJECT:
|
148
|
+
reject:
|
149
|
+
if (end != NULL)
|
150
|
+
*end = (const char *)o;
|
151
|
+
return false;
|
152
|
+
}
|
153
|
+
if (p == q)
|
154
|
+
return state == ACCEPT;
|
155
|
+
goto reject;
|
156
|
+
}
|
data/ext/u/u_downcase.c
ADDED
@@ -0,0 +1,201 @@
|
|
1
|
+
#include <stdbool.h>
|
2
|
+
#include <stddef.h>
|
3
|
+
#include <stdint.h>
|
4
|
+
#include "u.h"
|
5
|
+
|
6
|
+
#include <assert.h>
|
7
|
+
|
8
|
+
#include <string.h>
|
9
|
+
#include "output.h"
|
10
|
+
|
11
|
+
#include "data/constants.h"
|
12
|
+
#include "attributes.h"
|
13
|
+
#include "u_locale.h"
|
14
|
+
#include "titled.h"
|
15
|
+
#include "case.h"
|
16
|
+
#include "private.h"
|
17
|
+
|
18
|
+
#define LATIN_CAPITAL_LETTER_I ((uint32_t)0x0049)
|
19
|
+
#define LATIN_CAPITAL_LETTER_J ((uint32_t)0x004a)
|
20
|
+
#define LATIN_SMALL_LETTER_I ((uint32_t)0x0069)
|
21
|
+
#define LATIN_CAPITAL_LETTER_I_WITH_GRAVE ((uint32_t)0x00cc)
|
22
|
+
#define LATIN_CAPITAL_LETTER_I_WITH_ACUTE ((uint32_t)0x00cd)
|
23
|
+
#define LATIN_CAPITAL_LETTER_I_WITH_TILDE ((uint32_t)0x0128)
|
24
|
+
#define LATIN_CAPITAL_LETTER_I_WITH_OGONEK ((uint32_t)0x012e)
|
25
|
+
#define LATIN_CAPITAL_LETTER_I_WITH_DOT_ABOVE ((uint32_t)0x0130)
|
26
|
+
#define LATIN_SMALL_LETTER_DOTLESS_I ((uint32_t)0x0131)
|
27
|
+
#define COMBINING_GRAVE_ACCENT ((uint32_t)0x0300)
|
28
|
+
#define COMBINING_ACUTE_ACCENT ((uint32_t)0x0301)
|
29
|
+
#define COMBINING_TILDE ((uint32_t)0x0303)
|
30
|
+
#define COMBINING_DOT_ABOVE ((uint32_t)0x0307)
|
31
|
+
#define GREEK_CAPITAL_LETTER_SIGMA ((uint32_t)0x03a3)
|
32
|
+
#define GREEK_SMALL_LETTER_FINAL_SIGMA ((uint32_t)0x03c2)
|
33
|
+
#define GREEK_SMALL_LETTER_SIGMA ((uint32_t)0x03c3)
|
34
|
+
|
35
|
+
|
36
|
+
static inline bool
|
37
|
+
is_final_sigma(const char *string, const char *p, const char *q, const char *end)
|
38
|
+
{
|
39
|
+
if (p == string)
|
40
|
+
return false;
|
41
|
+
uint32_t c;
|
42
|
+
while (q < end) {
|
43
|
+
c = u_decode(&q, q, end);
|
44
|
+
if (u_char_iscaseignorable(c))
|
45
|
+
continue;
|
46
|
+
if (u_char_iscased(c))
|
47
|
+
return false;
|
48
|
+
break;
|
49
|
+
}
|
50
|
+
while (string < p) {
|
51
|
+
c = u_decode_r(&p, string, p);
|
52
|
+
if (u_char_iscaseignorable(c))
|
53
|
+
continue;
|
54
|
+
if (u_char_iscased(c))
|
55
|
+
return true;
|
56
|
+
return false;
|
57
|
+
}
|
58
|
+
return false;
|
59
|
+
}
|
60
|
+
|
61
|
+
static inline bool
|
62
|
+
has_more_above(const char *q, const char *end)
|
63
|
+
{
|
64
|
+
while (q < end) {
|
65
|
+
switch (u_char_canonical_combining_class(u_decode(&q, q, end))) {
|
66
|
+
case U_CANONICAL_COMBINING_CLASS_ABOVE:
|
67
|
+
return true;
|
68
|
+
case U_CANONICAL_COMBINING_CLASS_NOT_REORDERED:
|
69
|
+
return false;
|
70
|
+
default:
|
71
|
+
break;
|
72
|
+
}
|
73
|
+
}
|
74
|
+
return false;
|
75
|
+
}
|
76
|
+
|
77
|
+
static bool
|
78
|
+
downcase_lithuanian_i(uint32_t c, uint32_t combiner, struct output *output)
|
79
|
+
{
|
80
|
+
output_char(output, c);
|
81
|
+
output_char(output, COMBINING_DOT_ABOVE);
|
82
|
+
if (combiner != '\0')
|
83
|
+
output_char(output, combiner);
|
84
|
+
return true;
|
85
|
+
}
|
86
|
+
|
87
|
+
static inline bool
|
88
|
+
downcase_lithuanian(uint32_t c, const char *q, const char *end,
|
89
|
+
struct output *output)
|
90
|
+
{
|
91
|
+
switch (c) {
|
92
|
+
case LATIN_CAPITAL_LETTER_I:
|
93
|
+
case LATIN_CAPITAL_LETTER_J:
|
94
|
+
case LATIN_CAPITAL_LETTER_I_WITH_OGONEK:
|
95
|
+
if (!has_more_above(q, end))
|
96
|
+
return false;
|
97
|
+
return downcase_lithuanian_i(u_char_downcase(c), '\0', output);
|
98
|
+
case LATIN_CAPITAL_LETTER_I_WITH_GRAVE:
|
99
|
+
return downcase_lithuanian_i(LATIN_SMALL_LETTER_I,
|
100
|
+
COMBINING_GRAVE_ACCENT, output);
|
101
|
+
case LATIN_CAPITAL_LETTER_I_WITH_ACUTE:
|
102
|
+
return downcase_lithuanian_i(LATIN_SMALL_LETTER_I,
|
103
|
+
COMBINING_ACUTE_ACCENT, output);
|
104
|
+
case LATIN_CAPITAL_LETTER_I_WITH_TILDE:
|
105
|
+
return downcase_lithuanian_i(LATIN_SMALL_LETTER_I,
|
106
|
+
COMBINING_TILDE, output);
|
107
|
+
default:
|
108
|
+
return false;
|
109
|
+
}
|
110
|
+
}
|
111
|
+
|
112
|
+
static inline bool
|
113
|
+
is_before_dot(const char *q, const char *end)
|
114
|
+
{
|
115
|
+
while (q < end) {
|
116
|
+
uint32_t c = u_decode(&q, q, end);
|
117
|
+
if (c == COMBINING_DOT_ABOVE)
|
118
|
+
return true;
|
119
|
+
switch (u_char_canonical_combining_class(c)) {
|
120
|
+
case U_CANONICAL_COMBINING_CLASS_ABOVE:
|
121
|
+
case U_CANONICAL_COMBINING_CLASS_NOT_REORDERED:
|
122
|
+
return false;
|
123
|
+
default:
|
124
|
+
break;
|
125
|
+
}
|
126
|
+
}
|
127
|
+
return false;
|
128
|
+
}
|
129
|
+
|
130
|
+
static bool
|
131
|
+
is_i(uint32_t c)
|
132
|
+
{
|
133
|
+
return c == LATIN_CAPITAL_LETTER_I;
|
134
|
+
}
|
135
|
+
|
136
|
+
static inline bool
|
137
|
+
downcase_turkic(uint32_t c,
|
138
|
+
const char *string, const char *p, const char *q, const char *end,
|
139
|
+
struct output *output)
|
140
|
+
{
|
141
|
+
switch (c) {
|
142
|
+
case LATIN_CAPITAL_LETTER_I:
|
143
|
+
output_char(output,
|
144
|
+
is_before_dot(q, end) ?
|
145
|
+
LATIN_SMALL_LETTER_I :
|
146
|
+
LATIN_SMALL_LETTER_DOTLESS_I);
|
147
|
+
return true;
|
148
|
+
case LATIN_CAPITAL_LETTER_I_WITH_DOT_ABOVE:
|
149
|
+
case_simple(LATIN_CAPITAL_LETTER_I,
|
150
|
+
U_GENERAL_CATEGORY_LETTER_UPPERCASE,
|
151
|
+
false, false, output);
|
152
|
+
return true;
|
153
|
+
case COMBINING_DOT_ABOVE:
|
154
|
+
if (!is_after(string, p, is_i))
|
155
|
+
output_char(output, COMBINING_DOT_ABOVE);
|
156
|
+
return true;
|
157
|
+
default:
|
158
|
+
return false;
|
159
|
+
}
|
160
|
+
}
|
161
|
+
|
162
|
+
const char *
|
163
|
+
_u_downcase_step(const char *string, const char *p, const char *end,
|
164
|
+
enum locale locale, struct output *output)
|
165
|
+
{
|
166
|
+
const char *q;
|
167
|
+
uint32_t c = u_decode(&q, p, end);
|
168
|
+
enum u_general_category gc;
|
169
|
+
if (c == GREEK_CAPITAL_LETTER_SIGMA)
|
170
|
+
output_char(output,
|
171
|
+
is_final_sigma(string, p, q, end) ?
|
172
|
+
GREEK_SMALL_LETTER_FINAL_SIGMA :
|
173
|
+
GREEK_SMALL_LETTER_SIGMA);
|
174
|
+
else if (locale == LOCALE_LITHUANIAN &&
|
175
|
+
downcase_lithuanian(c, q, end, output))
|
176
|
+
;
|
177
|
+
else if (locale == LOCALE_TURKIC &&
|
178
|
+
downcase_turkic(c, string, p, q, end, output))
|
179
|
+
;
|
180
|
+
else if (IS(gc = u_char_general_category(c),
|
181
|
+
OR(U_GENERAL_CATEGORY_LETTER_UPPERCASE,
|
182
|
+
OR(U_GENERAL_CATEGORY_LETTER_TITLECASE, 0))))
|
183
|
+
case_simple(c, gc, false, false, output);
|
184
|
+
else
|
185
|
+
output_string(output, p, q - p);
|
186
|
+
return q;
|
187
|
+
}
|
188
|
+
|
189
|
+
size_t
|
190
|
+
u_downcase(char *result, size_t m, const char *u, size_t n,
|
191
|
+
const char *locale)
|
192
|
+
{
|
193
|
+
assert(u != NULL);
|
194
|
+
assert(result != NULL || m == 0);
|
195
|
+
enum locale l = _u_locale_from_string(locale);
|
196
|
+
const char *end = u + n;
|
197
|
+
struct output o = OUTPUT_INIT(result, m);
|
198
|
+
for (const char *p = u; p < end; )
|
199
|
+
p = _u_downcase_step(u, p, end, l, &o);
|
200
|
+
return output_finalize(&o);
|
201
|
+
}
|