|
Packit |
0ec9dd |
/* Pango
|
|
Packit |
0ec9dd |
* break-arabic.c:
|
|
Packit |
0ec9dd |
*
|
|
Packit |
0ec9dd |
* Copyright (C) 2006 Red Hat Software
|
|
Packit |
0ec9dd |
* Copyright (C) 2006 Sharif FarsiWeb, Inc.
|
|
Packit |
0ec9dd |
* Authors: Behdad Esfahbod <besfahbo@redhat.com>
|
|
Packit |
0ec9dd |
* Roozbeh Pournader <roozbeh@farsiweb.info>
|
|
Packit |
0ec9dd |
*
|
|
Packit |
0ec9dd |
* This library is free software; you can redistribute it and/or
|
|
Packit |
0ec9dd |
* modify it under the terms of the GNU Library General Public
|
|
Packit |
0ec9dd |
* License as published by the Free Software Foundation; either
|
|
Packit |
0ec9dd |
* version 2 of the License, or (at your option) any later version.
|
|
Packit |
0ec9dd |
*
|
|
Packit |
0ec9dd |
* This library is distributed in the hope that it will be useful,
|
|
Packit |
0ec9dd |
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
Packit |
0ec9dd |
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
|
Packit |
0ec9dd |
* Library General Public License for more details.
|
|
Packit |
0ec9dd |
*
|
|
Packit |
0ec9dd |
* You should have received a copy of the GNU Library General Public
|
|
Packit |
0ec9dd |
* License along with this library; if not, write to the
|
|
Packit |
0ec9dd |
* Free Software Foundation, Inc., 59 Temple Place - Suite 330,
|
|
Packit |
0ec9dd |
* Boston, MA 02111-1307, USA.
|
|
Packit |
0ec9dd |
*/
|
|
Packit |
0ec9dd |
|
|
Packit |
0ec9dd |
#include "config.h"
|
|
Packit |
0ec9dd |
|
|
Packit |
0ec9dd |
#include "pango-break.h"
|
|
Packit |
0ec9dd |
|
|
Packit |
0ec9dd |
#define ALEF_WITH_MADDA_ABOVE 0x0622
|
|
Packit |
0ec9dd |
#define YEH_WITH_HAMZA_ABOVE 0x0626
|
|
Packit |
0ec9dd |
#define ALEF 0x0627
|
|
Packit |
0ec9dd |
#define WAW 0x0648
|
|
Packit |
0ec9dd |
#define YEH 0x064A
|
|
Packit |
0ec9dd |
|
|
Packit |
0ec9dd |
#define MADDAH_ABOVE 0x0653
|
|
Packit |
0ec9dd |
#define HAMZA_ABOVE 0x0654
|
|
Packit |
0ec9dd |
#define HAMZA_BELOW 0x0655
|
|
Packit |
0ec9dd |
|
|
Packit |
0ec9dd |
/*
|
|
Packit |
0ec9dd |
* Arabic characters with canonical decompositions that are not just
|
|
Packit |
0ec9dd |
* ligatures. The characters U+06C0, U+06C2, and U+06D3 are intentionally
|
|
Packit |
0ec9dd |
* excluded as they are marked as "not an independent letter" in Unicode
|
|
Packit |
0ec9dd |
* Character Database's NamesList.txt
|
|
Packit |
0ec9dd |
*/
|
|
Packit |
0ec9dd |
#define IS_COMPOSITE(c) (ALEF_WITH_MADDA_ABOVE <= (c) && (c) <= YEH_WITH_HAMZA_ABOVE)
|
|
Packit |
0ec9dd |
|
|
Packit |
0ec9dd |
/* If a character is the second part of a composite Arabic character with an Alef */
|
|
Packit |
0ec9dd |
#define IS_COMPOSITE_WITH_ALEF(c) (MADDAH_ABOVE <= (c) && (c) <= HAMZA_BELOW)
|
|
Packit |
0ec9dd |
|
|
Packit |
0ec9dd |
static void
|
|
Packit |
0ec9dd |
break_arabic (const char *text,
|
|
Packit |
0ec9dd |
int length,
|
|
Packit |
0ec9dd |
const PangoAnalysis *analysis G_GNUC_UNUSED,
|
|
Packit |
0ec9dd |
PangoLogAttr *attrs,
|
|
Packit |
0ec9dd |
int attrs_len G_GNUC_UNUSED)
|
|
Packit |
0ec9dd |
{
|
|
Packit |
0ec9dd |
int i;
|
|
Packit |
0ec9dd |
const char *p;
|
|
Packit |
0ec9dd |
gunichar prev_wc, this_wc;
|
|
Packit |
0ec9dd |
|
|
Packit |
0ec9dd |
/* See http://bugzilla.gnome.org/show_bug.cgi?id=350132 for issues this
|
|
Packit |
0ec9dd |
* module tries to solve.
|
|
Packit |
0ec9dd |
*/
|
|
Packit |
0ec9dd |
|
|
Packit |
0ec9dd |
for (p = text, i = 0, prev_wc = 0;
|
|
Packit |
0ec9dd |
p < text + length;
|
|
Packit |
0ec9dd |
p = g_utf8_next_char (p), i++, prev_wc = this_wc)
|
|
Packit |
0ec9dd |
{
|
|
Packit |
0ec9dd |
this_wc = g_utf8_get_char (p);
|
|
Packit |
0ec9dd |
|
|
Packit |
0ec9dd |
/*
|
|
Packit |
0ec9dd |
* Unset backspace_deletes_character for various combinations.
|
|
Packit |
0ec9dd |
*
|
|
Packit |
0ec9dd |
* A few more combinations may need to be handled here, but are not
|
|
Packit |
0ec9dd |
* handled yet, as expectations of users is not known or may differ
|
|
Packit |
0ec9dd |
* among different languages or users:
|
|
Packit |
0ec9dd |
* some letters combined with U+0658 ARABIC MARK NOON GHUNNA;
|
|
Packit |
0ec9dd |
* combinations considered one letter in Azerbaijani (WAW+SUKUN and
|
|
Packit |
0ec9dd |
* FARSI_YEH+HAMZA_ABOVE); combinations of YEH and ALEF_MAKSURA with
|
|
Packit |
0ec9dd |
* HAMZA_BELOW (Qur'anic); TATWEEL+HAMZA_ABOVE (Qur'anic).
|
|
Packit |
0ec9dd |
*
|
|
Packit |
0ec9dd |
* FIXME: Ordering these in some other way may lower the time spent here, or not.
|
|
Packit |
0ec9dd |
*/
|
|
Packit |
0ec9dd |
if (G_UNLIKELY (
|
|
Packit |
0ec9dd |
IS_COMPOSITE (this_wc) ||
|
|
Packit |
0ec9dd |
(prev_wc == ALEF && IS_COMPOSITE_WITH_ALEF (this_wc)) ||
|
|
Packit |
0ec9dd |
(this_wc == HAMZA_ABOVE && (prev_wc == WAW || prev_wc == YEH))
|
|
Packit |
0ec9dd |
))
|
|
Packit |
0ec9dd |
attrs[i+1].backspace_deletes_character = FALSE;
|
|
Packit |
0ec9dd |
}
|
|
Packit |
0ec9dd |
}
|