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
@@ -1,444 +0,0 @@
|
|
1
|
-
/*
|
2
|
-
* contents: Unicode decomposition handling.
|
3
|
-
*
|
4
|
-
* Copyright (C) 2004 Nikolai Weibull <source@pcppopper.org>
|
5
|
-
*/
|
6
|
-
|
7
|
-
#include <ruby.h>
|
8
|
-
#include <stdbool.h>
|
9
|
-
#include <stddef.h>
|
10
|
-
#include <stdint.h>
|
11
|
-
|
12
|
-
#include "unicode.h"
|
13
|
-
|
14
|
-
#include "data/decompose.h"
|
15
|
-
#include "data/compose.h"
|
16
|
-
|
17
|
-
#include "private.h"
|
18
|
-
|
19
|
-
#define COMBINING_CLASS(c) \
|
20
|
-
SPLIT_UNICODE_TABLE_LOOKUP(cclass_data, combining_class_table_part1, combining_class_table_part2, (c), 0)
|
21
|
-
|
22
|
-
|
23
|
-
/* {{{1
|
24
|
-
* Hangul syllable [de]composition constants. A lot of work I'd say.
|
25
|
-
*/
|
26
|
-
enum {
|
27
|
-
SBase = 0xac00,
|
28
|
-
LBase = 0x1100,
|
29
|
-
VBase = 0x1161,
|
30
|
-
TBase = 0x11a7,
|
31
|
-
LCount = 19,
|
32
|
-
VCount = 21,
|
33
|
-
TCount = 28,
|
34
|
-
NCount = (VCount * TCount),
|
35
|
-
SCount = (LCount * NCount),
|
36
|
-
SLast = (SBase + SCount - 1)
|
37
|
-
};
|
38
|
-
|
39
|
-
|
40
|
-
/* {{{1
|
41
|
-
* Return the combinging class of ‘c’.
|
42
|
-
*/
|
43
|
-
int
|
44
|
-
unichar_combining_class(unichar c)
|
45
|
-
{
|
46
|
-
return COMBINING_CLASS(c);
|
47
|
-
}
|
48
|
-
|
49
|
-
|
50
|
-
/* {{{1
|
51
|
-
* Rearrange ‘str’ so that decomposed characters are arranged according to
|
52
|
-
* their combining class. Do this for at most ‘len’ bytes of data.
|
53
|
-
*/
|
54
|
-
static void
|
55
|
-
unicode_canonical_ordering_swap(unichar *str, size_t offset, int next, bool *swapped)
|
56
|
-
{
|
57
|
-
size_t initial = offset + 1;
|
58
|
-
|
59
|
-
size_t j = initial;
|
60
|
-
while (j > 0 && COMBINING_CLASS(str[j - 1]) > next) {
|
61
|
-
unichar c = str[j];
|
62
|
-
str[j] = str[j - 1];
|
63
|
-
str[j - 1] = c;
|
64
|
-
j--;
|
65
|
-
}
|
66
|
-
|
67
|
-
*swapped = *swapped || (j != initial);
|
68
|
-
}
|
69
|
-
|
70
|
-
static bool
|
71
|
-
unicode_canonical_ordering_reorder(unichar *str, size_t len)
|
72
|
-
{
|
73
|
-
bool swapped = false;
|
74
|
-
|
75
|
-
int prev = COMBINING_CLASS(str[0]);
|
76
|
-
for (size_t i = 0; i < len - 1; i++) {
|
77
|
-
int next = COMBINING_CLASS(str[i + 1]);
|
78
|
-
|
79
|
-
if (next != 0 && prev > next)
|
80
|
-
unicode_canonical_ordering_swap(str, i, next, &swapped);
|
81
|
-
else
|
82
|
-
prev = next;
|
83
|
-
}
|
84
|
-
|
85
|
-
return swapped;
|
86
|
-
}
|
87
|
-
|
88
|
-
void
|
89
|
-
unicode_canonical_ordering(unichar *str, size_t len)
|
90
|
-
{
|
91
|
-
while (unicode_canonical_ordering_reorder(str, len))
|
92
|
-
; /* This loop intentionally left empty. */
|
93
|
-
}
|
94
|
-
|
95
|
-
|
96
|
-
/* {{{1
|
97
|
-
* Decompose the character ‘s’ according to the rules outlined in
|
98
|
-
* http://www.unicode.org/unicode/reports/tr15/#Hangul. ‘r’ should be ‹NULL›
|
99
|
-
* or of sufficient length to store the decomposition of ‘s’. Returns the
|
100
|
-
* number of characters stored (or would be if it were non-NULL) in R.
|
101
|
-
*/
|
102
|
-
static size_t
|
103
|
-
decompose_hangul(unichar s, unichar *r)
|
104
|
-
{
|
105
|
-
int SIndex = s - SBase;
|
106
|
-
|
107
|
-
/* If this isn’t a Hangul symbol, then simply return it unchanged. */
|
108
|
-
if (SIndex < 0 || SIndex >= SCount) {
|
109
|
-
if (r != NULL)
|
110
|
-
r[0] = s;
|
111
|
-
return 1;
|
112
|
-
}
|
113
|
-
|
114
|
-
unichar L = LBase + SIndex / NCount;
|
115
|
-
unichar V = VBase + (SIndex % NCount) / TCount;
|
116
|
-
unichar T = TBase + SIndex % TCount;
|
117
|
-
|
118
|
-
if (r != NULL) {
|
119
|
-
r[0] = L;
|
120
|
-
r[1] = V;
|
121
|
-
}
|
122
|
-
|
123
|
-
if (T == TBase)
|
124
|
-
return 2;
|
125
|
-
|
126
|
-
if (r != NULL)
|
127
|
-
r[2] = T;
|
128
|
-
|
129
|
-
return 3;
|
130
|
-
}
|
131
|
-
|
132
|
-
|
133
|
-
/* {{{1
|
134
|
-
* Search the Unicode decomposition table for ‘c’ and depending on the boolean
|
135
|
-
* value of ‘compat’, return its compatibility or canonical decomposition if
|
136
|
-
* found.
|
137
|
-
*/
|
138
|
-
static const char *
|
139
|
-
get_decomposition(int index, bool compat)
|
140
|
-
{
|
141
|
-
int offset;
|
142
|
-
|
143
|
-
if (compat) {
|
144
|
-
offset = decomp_table[index].compat_offset;
|
145
|
-
if (offset == UNICODE_NOT_PRESENT_OFFSET)
|
146
|
-
offset = decomp_table[index].canon_offset;
|
147
|
-
/* Either .compat_offset or .canon_offset can be missing, but
|
148
|
-
* not both. */
|
149
|
-
} else {
|
150
|
-
offset = decomp_table[index].canon_offset;
|
151
|
-
if (offset == UNICODE_NOT_PRESENT_OFFSET)
|
152
|
-
return NULL;
|
153
|
-
}
|
154
|
-
|
155
|
-
return &decomp_expansion_string[offset];
|
156
|
-
}
|
157
|
-
|
158
|
-
static const char *
|
159
|
-
find_decomposition(unichar c, bool compat)
|
160
|
-
{
|
161
|
-
int index;
|
162
|
-
|
163
|
-
if (!unicode_table_lookup(decomp_table, c, &index))
|
164
|
-
return NULL;
|
165
|
-
|
166
|
-
return get_decomposition(index, compat);
|
167
|
-
}
|
168
|
-
|
169
|
-
|
170
|
-
/* {{{1
|
171
|
-
* Copy over the UTF-8 decomposition in ‘decomposition’ to the unichar buffer
|
172
|
-
* ‘chars’. Return the number of unichars in ‘chars’.
|
173
|
-
*/
|
174
|
-
static size_t
|
175
|
-
decomposition_to_wc(const char *decomposition, unichar *chars)
|
176
|
-
{
|
177
|
-
size_t i = 0;
|
178
|
-
for (const char *p = decomposition; *p != '\0'; p = utf_next(p))
|
179
|
-
chars[i++] = utf_char(p);
|
180
|
-
|
181
|
-
return i;
|
182
|
-
}
|
183
|
-
|
184
|
-
|
185
|
-
/* {{{1
|
186
|
-
* Generate the canonical decomposition of ‘c’. The length of the
|
187
|
-
* decomposition is stored in ‘len’.
|
188
|
-
*/
|
189
|
-
/* TODO: clean this up. */
|
190
|
-
unichar *
|
191
|
-
unicode_canonical_decomposition(unichar c, size_t *len)
|
192
|
-
{
|
193
|
-
const char *decomp;
|
194
|
-
unichar *r;
|
195
|
-
|
196
|
-
/* Hangul syllable */
|
197
|
-
if (c >= SBase && c <= SLast) {
|
198
|
-
*len = decompose_hangul(c, NULL);
|
199
|
-
r = ALLOC_N(unichar, *len);
|
200
|
-
decompose_hangul(c, r);
|
201
|
-
} else if ((decomp = find_decomposition(c, false)) != NULL) {
|
202
|
-
*len = utf_length(decomp);
|
203
|
-
r = ALLOC_N(unichar, *len);
|
204
|
-
decomposition_to_wc(decomp, r);
|
205
|
-
} else {
|
206
|
-
r = ALLOC(unichar);
|
207
|
-
*r = c;
|
208
|
-
*len = 1;
|
209
|
-
}
|
210
|
-
|
211
|
-
/* Supposedly following the Unicode 2.1.9 table means that the
|
212
|
-
* decompositions come out in canonical order. I haven't tested this,
|
213
|
-
* but we rely on it here. */
|
214
|
-
return r;
|
215
|
-
}
|
216
|
-
|
217
|
-
|
218
|
-
/* {{{1
|
219
|
-
* Combine Hangul characters ‘a’ and ‘b’ if possible, and store the result in
|
220
|
-
* ‘result’. The combinations tried are L,V => LV and LV,T => LVT in that
|
221
|
-
* order.
|
222
|
-
*/
|
223
|
-
static bool
|
224
|
-
combine_hangul(unichar a, unichar b, unichar *result)
|
225
|
-
{
|
226
|
-
int LIndex = a - LBase;
|
227
|
-
int VIndex = b - VBase;
|
228
|
-
|
229
|
-
if (0 <= LIndex && LIndex < LCount && 0 <= VIndex && VIndex < VCount) {
|
230
|
-
*result = SBase + (LIndex * VCount + VIndex) * TCount;
|
231
|
-
return true;
|
232
|
-
}
|
233
|
-
|
234
|
-
int SIndex = a - SBase;
|
235
|
-
int TIndex = b - TBase;
|
236
|
-
|
237
|
-
if (0 <= SIndex && SIndex < SCount &&
|
238
|
-
(SIndex % TCount) == 0 &&
|
239
|
-
0 < TIndex && TIndex < TCount) {
|
240
|
-
*result = a + TIndex;
|
241
|
-
return true;
|
242
|
-
}
|
243
|
-
|
244
|
-
return false;
|
245
|
-
}
|
246
|
-
|
247
|
-
|
248
|
-
/* {{{1
|
249
|
-
* Try to combine the Unicode characters ‘a’ and ‘b’ storing the result in
|
250
|
-
* ‘result’.
|
251
|
-
*/
|
252
|
-
static uint16_t
|
253
|
-
compose_index(unichar c)
|
254
|
-
{
|
255
|
-
unsigned int page = c >> 8;
|
256
|
-
|
257
|
-
if (page > COMPOSE_TABLE_LAST)
|
258
|
-
return 0;
|
259
|
-
|
260
|
-
return SPLIT_UNICODE_TABLE_LOOKUP_PAGE(compose_data, compose_table, page, c);
|
261
|
-
}
|
262
|
-
|
263
|
-
static bool
|
264
|
-
lookup_compose(const uint16_t table[][2], uint16_t index, unichar c,
|
265
|
-
unichar *result)
|
266
|
-
{
|
267
|
-
if (c != table[index][0])
|
268
|
-
return false;
|
269
|
-
|
270
|
-
*result = table[index][1];
|
271
|
-
|
272
|
-
return true;
|
273
|
-
}
|
274
|
-
|
275
|
-
static bool
|
276
|
-
combine(unichar a, unichar b, unichar *result)
|
277
|
-
{
|
278
|
-
if (combine_hangul(a, b, result))
|
279
|
-
return true;
|
280
|
-
|
281
|
-
uint16_t index_a = compose_index(a);
|
282
|
-
if (index_a >= COMPOSE_FIRST_SINGLE_START && index_a < COMPOSE_SECOND_START)
|
283
|
-
return lookup_compose(compose_first_single,
|
284
|
-
index_a - COMPOSE_FIRST_SINGLE_START,
|
285
|
-
b,
|
286
|
-
result);
|
287
|
-
|
288
|
-
uint16_t index_b = compose_index(b);
|
289
|
-
if (index_b >= COMPOSE_SECOND_SINGLE_START)
|
290
|
-
return lookup_compose(compose_second_single,
|
291
|
-
index_b - COMPOSE_SECOND_SINGLE_START,
|
292
|
-
a,
|
293
|
-
result);
|
294
|
-
|
295
|
-
if (index_a >= COMPOSE_FIRST_START &&
|
296
|
-
index_a < COMPOSE_FIRST_SINGLE_START &&
|
297
|
-
index_b >= COMPOSE_SECOND_START &&
|
298
|
-
index_b < COMPOSE_SECOND_SINGLE_START) {
|
299
|
-
unichar r = compose_array[index_a - COMPOSE_FIRST_START][index_b - COMPOSE_SECOND_START];
|
300
|
-
|
301
|
-
if (r != 0) {
|
302
|
-
*result = r;
|
303
|
-
return true;
|
304
|
-
}
|
305
|
-
}
|
306
|
-
|
307
|
-
return false;
|
308
|
-
}
|
309
|
-
|
310
|
-
|
311
|
-
/* {{{1
|
312
|
-
* Normalize (compose/decompose) characters in ‘str’ so that strings that
|
313
|
-
* actually contain the same characters will be recognized as equal for
|
314
|
-
* comparison for example.
|
315
|
-
*/
|
316
|
-
static size_t
|
317
|
-
normalize_wc_decompose_one(unichar c, NormalizeMode mode, unichar *buf)
|
318
|
-
{
|
319
|
-
bool do_compat = (mode == NORMALIZE_NFKC || mode == NORMALIZE_NFKD);
|
320
|
-
const char *decomp = find_decomposition(c, do_compat);
|
321
|
-
|
322
|
-
if (decomp == NULL) {
|
323
|
-
if (buf != NULL)
|
324
|
-
buf[0] = c;
|
325
|
-
return 1;
|
326
|
-
}
|
327
|
-
|
328
|
-
if (buf != NULL)
|
329
|
-
return decomposition_to_wc(decomp, buf);
|
330
|
-
|
331
|
-
return utf_length(decomp);
|
332
|
-
}
|
333
|
-
|
334
|
-
static void
|
335
|
-
normalize_wc_decompose(const char *str, size_t max_len, bool use_len,
|
336
|
-
NormalizeMode mode, unichar *buf, size_t *buf_len)
|
337
|
-
{
|
338
|
-
size_t n = 0;
|
339
|
-
size_t prev_start = 0;
|
340
|
-
const char *end = str + max_len;
|
341
|
-
for (const char *p = str; (!use_len || p < end) && *p != NUL; p = utf_next(p)) {
|
342
|
-
unichar c = utf_char(p);
|
343
|
-
size_t prev_n = n;
|
344
|
-
|
345
|
-
unichar *base = (buf != NULL) ? buf + n : NULL;
|
346
|
-
if (c >= SBase && c <= SLast)
|
347
|
-
n += decompose_hangul(c, base);
|
348
|
-
else
|
349
|
-
n += normalize_wc_decompose_one(c, mode, base);
|
350
|
-
|
351
|
-
if (buf != NULL && n > 0 && COMBINING_CLASS(buf[prev_n]) == 0) {
|
352
|
-
unicode_canonical_ordering(buf + prev_start,
|
353
|
-
n - prev_start);
|
354
|
-
prev_start = prev_n;
|
355
|
-
}
|
356
|
-
}
|
357
|
-
|
358
|
-
if (buf != NULL && n > 0)
|
359
|
-
unicode_canonical_ordering(buf + prev_start, n - prev_start);
|
360
|
-
|
361
|
-
if (buf != NULL)
|
362
|
-
buf[n] = NUL;
|
363
|
-
|
364
|
-
*buf_len = n;
|
365
|
-
}
|
366
|
-
|
367
|
-
static unichar *
|
368
|
-
normalize_wc_compose(unichar *buf, size_t len)
|
369
|
-
{
|
370
|
-
int new_len = len;
|
371
|
-
size_t prev_start = 0;
|
372
|
-
int prev_cc = 0;
|
373
|
-
|
374
|
-
for (size_t i = 0; i < len; i++) {
|
375
|
-
int cc = COMBINING_CLASS(buf[i]);
|
376
|
-
size_t j = i - (len - new_len);
|
377
|
-
|
378
|
-
if (j > 0 && (prev_cc == 0 || prev_cc < cc) &&
|
379
|
-
combine(buf[prev_start], buf[i], &buf[prev_start])) {
|
380
|
-
new_len--;
|
381
|
-
prev_cc = (j + 1 == prev_start) ?
|
382
|
-
0 : COMBINING_CLASS(buf[j - 1]);
|
383
|
-
} else {
|
384
|
-
if (cc == 0)
|
385
|
-
prev_start = j;
|
386
|
-
|
387
|
-
buf[j] = buf[i];
|
388
|
-
prev_cc = cc;
|
389
|
-
}
|
390
|
-
}
|
391
|
-
|
392
|
-
buf[new_len] = NUL;
|
393
|
-
|
394
|
-
return buf;
|
395
|
-
}
|
396
|
-
|
397
|
-
unichar *
|
398
|
-
_utf_normalize_wc(const char *str, size_t max_len, bool use_len, NormalizeMode mode)
|
399
|
-
{
|
400
|
-
size_t n;
|
401
|
-
normalize_wc_decompose(str, max_len, use_len, mode, NULL, &n);
|
402
|
-
unichar *buf = ALLOC_N(unichar, n + 1);
|
403
|
-
normalize_wc_decompose(str, max_len, use_len, mode, buf, &n);
|
404
|
-
|
405
|
-
/* Just return if we don’t want composition. */
|
406
|
-
if (!(mode == NORMALIZE_NFC || mode == NORMALIZE_NFKC))
|
407
|
-
return buf;
|
408
|
-
|
409
|
-
return normalize_wc_compose(buf, n);
|
410
|
-
}
|
411
|
-
|
412
|
-
|
413
|
-
/* {{{1
|
414
|
-
* Normalize (compose/decompose) characters in ‘str˚ so that strings that
|
415
|
-
* actually contain the same characters will be recognized as equal for
|
416
|
-
* comparison for example.
|
417
|
-
*/
|
418
|
-
char *
|
419
|
-
utf_normalize(const char *str, NormalizeMode mode)
|
420
|
-
{
|
421
|
-
unichar *wcs = _utf_normalize_wc(str, 0, false, mode);
|
422
|
-
char *utf = ucs4_to_utf8(wcs, NULL, NULL);
|
423
|
-
|
424
|
-
free(wcs);
|
425
|
-
return utf;
|
426
|
-
}
|
427
|
-
|
428
|
-
|
429
|
-
/* {{{1
|
430
|
-
* This function is the same as utf_normalize() except that at most ‘len˚
|
431
|
-
* bytes are normalized from ‘str’.
|
432
|
-
*/
|
433
|
-
char *
|
434
|
-
utf_normalize_n(const char *str, NormalizeMode mode, size_t len)
|
435
|
-
{
|
436
|
-
unichar *wcs = _utf_normalize_wc(str, len, true, mode);
|
437
|
-
char *utf = ucs4_to_utf8(wcs, NULL, NULL);
|
438
|
-
|
439
|
-
free(wcs);
|
440
|
-
return utf;
|
441
|
-
}
|
442
|
-
|
443
|
-
|
444
|
-
/* }}}1 */
|
@@ -1,65 +0,0 @@
|
|
1
|
-
break.o: break.c unicode.h data/break.h
|
2
|
-
decompose.o: decompose.c unicode.h private.h data/decompose.h \
|
3
|
-
data/compose.h
|
4
|
-
private.o: private.c private.h
|
5
|
-
properties.o: properties.c unicode.h private.h data/character-tables.h
|
6
|
-
rb_utf_aref.o: rb_utf_aref.c rb_includes.h unicode.h private.h \
|
7
|
-
rb_methods.h
|
8
|
-
rb_utf_aset.o: rb_utf_aset.c rb_includes.h unicode.h private.h \
|
9
|
-
rb_methods.h
|
10
|
-
rb_utf_casecmp.o: rb_utf_casecmp.c rb_includes.h unicode.h private.h \
|
11
|
-
rb_methods.h
|
12
|
-
rb_utf_chomp.o: rb_utf_chomp.c rb_includes.h unicode.h private.h \
|
13
|
-
rb_methods.h
|
14
|
-
rb_utf_chop.o: rb_utf_chop.c rb_includes.h unicode.h private.h \
|
15
|
-
rb_methods.h
|
16
|
-
rb_utf_collate.o: rb_utf_collate.c rb_includes.h unicode.h private.h \
|
17
|
-
rb_methods.h
|
18
|
-
rb_utf_count.o: rb_utf_count.c rb_includes.h unicode.h private.h \
|
19
|
-
rb_methods.h rb_utf_internal_tr.h
|
20
|
-
rb_utf_delete.o: rb_utf_delete.c rb_includes.h unicode.h private.h \
|
21
|
-
rb_methods.h rb_utf_internal_tr.h
|
22
|
-
rb_utf_downcase.o: rb_utf_downcase.c rb_includes.h unicode.h private.h \
|
23
|
-
rb_methods.h
|
24
|
-
rb_utf_each_char.o: rb_utf_each_char.c rb_includes.h unicode.h private.h \
|
25
|
-
rb_methods.h
|
26
|
-
rb_utf_foldcase.o: rb_utf_foldcase.c rb_includes.h unicode.h private.h \
|
27
|
-
rb_methods.h
|
28
|
-
rb_utf_hex.o: rb_utf_hex.c rb_includes.h unicode.h private.h rb_methods.h \
|
29
|
-
rb_utf_internal_bignum.h
|
30
|
-
rb_utf_index.o: rb_utf_index.c rb_includes.h unicode.h private.h \
|
31
|
-
rb_methods.h
|
32
|
-
rb_utf_insert.o: rb_utf_insert.c rb_includes.h unicode.h private.h \
|
33
|
-
rb_methods.h
|
34
|
-
rb_utf_internal_bignum.o: rb_utf_internal_bignum.c rb_includes.h \
|
35
|
-
unicode.h private.h rb_methods.h rb_utf_internal_bignum.h
|
36
|
-
rb_utf_internal_tr.o: rb_utf_internal_tr.c rb_includes.h unicode.h \
|
37
|
-
private.h rb_methods.h rb_utf_internal_tr.h
|
38
|
-
rb_utf_justify.o: rb_utf_justify.c rb_includes.h unicode.h private.h \
|
39
|
-
rb_methods.h
|
40
|
-
rb_utf_length.o: rb_utf_length.c rb_includes.h unicode.h private.h \
|
41
|
-
rb_methods.h
|
42
|
-
rb_utf_lstrip.o: rb_utf_lstrip.c rb_includes.h unicode.h private.h \
|
43
|
-
rb_methods.h
|
44
|
-
rb_utf_normalize.o: rb_utf_normalize.c rb_includes.h unicode.h private.h \
|
45
|
-
rb_methods.h
|
46
|
-
rb_utf_oct.o: rb_utf_oct.c rb_includes.h unicode.h private.h rb_methods.h \
|
47
|
-
rb_utf_internal_bignum.h
|
48
|
-
rb_utf_reverse.o: rb_utf_reverse.c rb_includes.h unicode.h private.h \
|
49
|
-
rb_methods.h
|
50
|
-
rb_utf_rindex.o: rb_utf_rindex.c rb_includes.h unicode.h private.h \
|
51
|
-
rb_methods.h
|
52
|
-
rb_utf_rstrip.o: rb_utf_rstrip.c rb_includes.h unicode.h private.h \
|
53
|
-
rb_methods.h
|
54
|
-
rb_utf_squeeze.o: rb_utf_squeeze.c rb_includes.h unicode.h private.h \
|
55
|
-
rb_methods.h rb_utf_internal_tr.h
|
56
|
-
rb_utf_strip.o: rb_utf_strip.c rb_includes.h unicode.h private.h \
|
57
|
-
rb_methods.h
|
58
|
-
rb_utf_to_i.o: rb_utf_to_i.c rb_includes.h unicode.h private.h \
|
59
|
-
rb_methods.h rb_utf_internal_bignum.h
|
60
|
-
rb_utf_tr.o: rb_utf_tr.c rb_includes.h unicode.h private.h rb_methods.h \
|
61
|
-
rb_utf_internal_tr.h
|
62
|
-
rb_utf_upcase.o: rb_utf_upcase.c rb_includes.h unicode.h private.h \
|
63
|
-
rb_methods.h
|
64
|
-
unicode.o: unicode.c unicode.h private.h rb_methods.h
|
65
|
-
utf.o: utf.c unicode.h private.h
|