1 | /* Pango |
2 | * break-thai.c: |
3 | * |
4 | * Copyright (C) 2003 Theppitak Karoonboonyanan <thep@linux.thai.net> |
5 | * |
6 | * This library is free software; you can redistribute it and/or |
7 | * modify it under the terms of the GNU Library General Public |
8 | * License as published by the Free Software Foundation; either |
9 | * version 2 of the License, or (at your option) any later version. |
10 | * |
11 | * This library is distributed in the hope that it will be useful, |
12 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
13 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
14 | * Library General Public License for more details. |
15 | * |
16 | * You should have received a copy of the GNU Library General Public |
17 | * License along with this library; if not, write to the Free Software |
18 | * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA |
19 | */ |
20 | |
21 | #include "config.h" |
22 | |
23 | #include "pango-break.h" |
24 | #include "pango-impl-utils.h" |
25 | |
26 | #ifdef HAVE_LIBTHAI |
27 | #include <thai/thwchar.h> |
28 | #include <thai/thbrk.h> |
29 | |
30 | G_LOCK_DEFINE_STATIC (thai_brk); |
31 | |
32 | #ifdef HAVE_TH_BRK_FIND_BREAKS |
33 | static ThBrk *thai_brk = NULL; |
34 | #endif |
35 | |
36 | /* |
37 | * tis_text is assumed to be large enough to hold the converted string, |
38 | * i.e. it must be at least pango_utf8_strlen(text, len)+1 bytes. |
39 | */ |
40 | static thchar_t * |
41 | utf8_to_tis (const char *text, int len, thchar_t *tis_text, int *tis_cnt) |
42 | { |
43 | thchar_t *tis_p; |
44 | const char *text_p; |
45 | |
46 | tis_p = tis_text; |
47 | for (text_p = text; text_p < text + len; text_p = g_utf8_next_char (text_p)) |
48 | *tis_p++ = th_uni2tis (g_utf8_get_char (text_p)); |
49 | *tis_p++ = '\0'; |
50 | |
51 | *tis_cnt = tis_p - tis_text; |
52 | return tis_text; |
53 | } |
54 | |
55 | #endif |
56 | static void |
57 | break_thai (const char *text, |
58 | int len, |
59 | const PangoAnalysis *analysis G_GNUC_UNUSED, |
60 | PangoLogAttr *attrs, |
61 | int attrs_len G_GNUC_UNUSED) |
62 | { |
63 | #ifdef HAVE_LIBTHAI |
64 | thchar_t tis_stack[512]; |
65 | int brk_stack[512]; |
66 | thchar_t *tis_text; |
67 | int *brk_pnts; |
68 | int cnt; |
69 | |
70 | cnt = pango_utf8_strlen (text, len) + 1; |
71 | |
72 | tis_text = tis_stack; |
73 | if (cnt > (int) G_N_ELEMENTS (tis_stack)) |
74 | tis_text = g_new (thchar_t, cnt); |
75 | |
76 | utf8_to_tis (text, len, tis_text, &cnt); |
77 | |
78 | brk_pnts = brk_stack; |
79 | if (cnt > (int) G_N_ELEMENTS (brk_stack)) |
80 | brk_pnts = g_new (int, cnt); |
81 | |
82 | /* find line break positions */ |
83 | |
84 | G_LOCK (thai_brk); |
85 | #ifdef HAVE_TH_BRK_FIND_BREAKS |
86 | if (thai_brk == NULL) |
87 | thai_brk = th_brk_new(NULL); |
88 | len = th_brk_find_breaks(thai_brk, tis_text, brk_pnts, cnt); |
89 | #else |
90 | len = th_brk (tis_text, brk_pnts, cnt); |
91 | #endif |
92 | G_UNLOCK (thai_brk); |
93 | |
94 | for (cnt = 0; cnt < len; cnt++) |
95 | { |
96 | if (!attrs[brk_pnts[cnt]].is_line_break) |
97 | { |
98 | /* Insert line breaks where there wasn't one. |
99 | * Satisfy invariants by marking it as char break too. |
100 | */ |
101 | attrs[brk_pnts[cnt]].is_char_break = TRUE; |
102 | attrs[brk_pnts[cnt]].is_line_break = TRUE; |
103 | } |
104 | if (!(attrs[brk_pnts[cnt]].is_word_start || |
105 | attrs[brk_pnts[cnt]].is_word_end)) |
106 | { |
107 | /* If we find a break in the middle of a sequence |
108 | * of characters, end and start a word. We must |
109 | * be careful only to do that if default_break |
110 | * did not already find a word start or end, |
111 | * otherwise we mess up the sequence. |
112 | */ |
113 | attrs[brk_pnts[cnt]].is_word_start = TRUE; |
114 | attrs[brk_pnts[cnt]].is_word_end = TRUE; |
115 | } |
116 | } |
117 | |
118 | if (brk_pnts != brk_stack) |
119 | g_free (brk_pnts); |
120 | |
121 | if (tis_text != tis_stack) |
122 | g_free (tis_text); |
123 | #endif |
124 | } |
125 | |