1 /***************************************************************************
3 * Open \______ \ ____ ____ | | _\_ |__ _______ ___
4 * Source | _// _ \_/ ___\| |/ /| __ \ / _ \ \/ /
5 * Jukebox | | ( <_> ) \___| < | \_\ ( <_> > < <
6 * Firmware |____|_ /\____/ \___ >__|_ \|___ /\____/__/\_ \
10 * Copyright (C) 2005 by Gadi Cohen
12 * Largely based on php_hebrev by Zeev Suraski <zeev@php.net>
13 * Heavily modified by Gadi Cohen aka Kinslayer <dragon@wastelands.net>
15 * This program is free software; you can redistribute it and/or
16 * modify it under the terms of the GNU General Public License
17 * as published by the Free Software Foundation; either version 2
18 * of the License, or (at your option) any later version.
20 * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
21 * KIND, either express or implied.
23 ****************************************************************************/
29 #include "rbunicode.h"
31 #include "scroll_engine.h"
34 /* #define _HEB_BUFFER_LENGTH (MAX_PATH + LCD_WIDTH/2 + 3 + 2 + 2) * 2 */
35 #define _HEB_BLOCK_TYPE_ENG 1
36 #define _HEB_BLOCK_TYPE_HEB 0
37 #define _HEB_ORIENTATION_LTR 1
38 #define _HEB_ORIENTATION_RTL 0
40 #define ischar(c) ((c > 0x0589 && c < 0x0700) || \
41 (c >= 0xfb50 && c <= 0xfefc) ? 1 : 0)
42 #define _isblank(c) ((c==' ' || c=='\t') ? 1 : 0)
43 #define _isnewline(c) ((c=='\n' || c=='\r') ? 1 : 0)
44 #define XOR(a,b) ((a||b) && !(a&&b))
46 static const arab_t
* arab_lookup(unsigned short uchar
)
48 if (uchar
>= 0x621 && uchar
<= 0x63a)
49 return &(jointable
[uchar
- 0x621]);
50 if (uchar
>= 0x640 && uchar
<= 0x64a)
51 return &(jointable
[uchar
- 0x621 - 5]);
52 if (uchar
>= 0x671 && uchar
<= 0x6d5)
53 return &(jointable
[uchar
- 0x621 - 5 - 38]);
54 if (uchar
== 0x200D) /* Support for the zero-width joiner */
59 static void arabjoin(unsigned short * stringprt
, int length
)
61 bool connected
= false;
62 unsigned short * writeprt
= stringprt
;
64 const arab_t
* prev
= 0;
66 const arab_t
* ligature
= 0;
70 for (i
= 0; i
<= length
; i
++) {
71 cur
= arab_lookup(uchar
= *stringprt
++);
73 /* Skip non-arabic chars */
76 /* Finish the last char */
78 *writeprt
++ = prev
->final
;
81 *writeprt
++ = prev
->isolated
;
90 /* nothing to do for arabic char if the previous was non-arabic */
96 /* if it's LAM, check for LAM+ALEPH ligatures */
97 if (prev
->isolated
== 0xfedd) {
98 switch (cur
->isolated
) {
100 ligature
= &(lamaleph
[0]);
103 ligature
= &(lamaleph
[1]);
106 ligature
= &(lamaleph
[2]);
109 ligature
= &(lamaleph
[3]);
113 if (ligature
) { /* replace the 2 glyphs by their ligature */
117 if (connected
) { /* previous char has something connected to it */
118 if (prev
->medial
&& cur
->final
) /* Can we connect to it? */
119 *writeprt
++ = prev
->medial
;
121 *writeprt
++ = prev
->final
;
125 if (prev
->initial
&& cur
->final
) { /* Can we connect to it? */
126 *writeprt
++ = prev
->initial
;
129 *writeprt
++ = prev
->isolated
;
136 unsigned short *bidi_l2v(const unsigned char *str
, int orientation
)
138 int length
= utf8length(str
);
139 static unsigned short utf16_buf
[SCROLL_LINE_SIZE
];
140 static unsigned short bidi_buf
[SCROLL_LINE_SIZE
];
141 unsigned short *heb_str
, *target
, *tmp
; /* *broken_str */
142 int block_start
, block_end
, block_type
, block_length
, i
;
145 int begin, end, char_count, orig_begin;
149 target
= tmp
= utf16_buf
;
151 str
= utf8decode(str
, target
++);
154 if (target
== utf16_buf
) /* empty string */
157 /* properly join any arabic chars */
158 arabjoin(utf16_buf
, length
);
160 block_start
=block_end
=block_length
=0;
166 target
= heb_str
+ length
;
172 block_type
= _HEB_BLOCK_TYPE_HEB
;
174 block_type
= _HEB_BLOCK_TYPE_ENG
;
177 while((XOR(ischar(*(tmp
+1)),block_type
)
178 || _isblank(*(tmp
+1)) || ispunct((int)*(tmp
+1))
180 && block_end
< length
-1) {
186 if (block_type
!= orientation
) {
187 while ((_isblank(*tmp
) || ispunct((int)*tmp
))
188 && *tmp
!='/' && *tmp
!='-' && block_end
>block_start
) {
194 for (i
=block_start
; i
<=block_end
; i
++) {
195 *target
= (block_type
== orientation
) ?
196 *(utf16_buf
+i
) : *(utf16_buf
+block_end
-i
+block_start
);
197 if (block_type
!=orientation
) {
209 target
+= orientation
? 1 : -1;
211 block_type
= !block_type
;
212 block_start
=block_end
+1;
213 } while(block_end
<length
-1);
217 #if 0 /* Is this code really necessary? */
218 broken_str
= utf16_buf
;
224 while ((!max_chars
|| char_count
<max_chars
) && begin
>0) {
227 if (begin
<=0 || _isnewline(heb_str
[begin
])) {
228 while(begin
>0 && _isnewline(heb_str
[begin
-1])) {
235 if (char_count
==max_chars
) { /* try to avoid breaking words */
236 int new_char_count
= char_count
;
237 int new_begin
= begin
;
239 while (new_char_count
>0) {
240 if (_isblank(heb_str
[new_begin
]) ||
241 _isnewline(heb_str
[new_begin
])) {
247 if (new_char_count
>0) {
248 char_count
=new_char_count
;
254 /* if (_isblank(heb_str[begin])) {
258 /* skip leading newlines */
259 while (begin
<=end
&& _isnewline(heb_str
[begin
])) {
264 for (i
=begin
; i
<=end
; i
++) {
265 *target
= heb_str
[i
];
269 for (i
=orig_begin
; i
<=end
&& _isnewline(heb_str
[i
]); i
++) {
270 *target
= heb_str
[i
];