2 * Implementation of Indic Syllables for the Uniscribe Script Processor
4 * Copyright 2011 CodeWeavers, Aric Stewart
6 * This library is free software; you can redistribute it and/or
7 * modify it under the terms of the GNU Lesser General Public
8 * License as published by the Free Software Foundation; either
9 * version 2.1 of the License, or (at your option) any later version.
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 * Lesser General Public License for more details.
16 * You should have received a copy of the GNU Lesser 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
34 #include "wine/debug.h"
35 #include "usp10_internal.h"
37 WINE_DEFAULT_DEBUG_CHANNEL(uniscribe
);
39 static void debug_output_string(LPCWSTR str
, int cChar
, lexical_function f
)
42 if (TRACE_ON(uniscribe
))
44 for (i
= 0; i
< cChar
; i
++)
48 case lex_Consonant
: TRACE("C"); break;
49 case lex_Ra
: TRACE("Ra"); break;
50 case lex_Vowel
: TRACE("V"); break;
51 case lex_Nukta
: TRACE("N"); break;
52 case lex_Halant
: TRACE("H"); break;
53 case lex_ZWNJ
: TRACE("Zwnj"); break;
54 case lex_ZWJ
: TRACE("Zwj"); break;
55 case lex_Matra_post
: TRACE("Mp");break;
56 case lex_Matra_above
: TRACE("Ma");break;
57 case lex_Matra_below
: TRACE("Mb");break;
58 case lex_Matra_pre
: TRACE("Mm");break;
59 case lex_Modifier
: TRACE("Sm"); break;
60 case lex_Vedic
: TRACE("Vd"); break;
61 case lex_Anudatta
: TRACE("A"); break;
62 case lex_Composed_Vowel
: TRACE("t"); break;
71 static inline BOOL
is_matra( int type
)
73 return (type
== lex_Matra_above
|| type
== lex_Matra_below
||
74 type
== lex_Matra_pre
|| type
== lex_Matra_post
);
77 static inline BOOL
is_joiner( int type
)
79 return (type
== lex_ZWJ
|| type
== lex_ZWNJ
);
82 static INT
consonant_header(LPCWSTR input
, INT cChar
, INT start
, INT next
,
85 if (!is_consonant( lex(input
[next
]) )) return -1;
87 if ((next
< cChar
) && lex(input
[next
]) == lex_Nukta
)
89 if (lex(input
[next
])==lex_Halant
)
92 if((next
< cChar
) && is_joiner( lex(input
[next
]) ))
94 if ((next
< cChar
) && is_consonant( lex(input
[next
]) ))
97 else if (is_joiner( lex(input
[next
]) ) && lex(input
[next
+1])==lex_Halant
)
100 if ((next
< cChar
) && is_consonant( lex(input
[next
]) ))
106 static INT
parse_consonant_syllable(LPCWSTR input
, INT cChar
, INT start
,
107 INT
*main
, INT next
, lexical_function lex
)
113 check
= consonant_header(input
,cChar
,start
,next
,lex
);
119 } while (check
!= -1);
120 if (headers
|| is_consonant( lex(input
[next
]) ))
127 if ((next
< cChar
) && lex(input
[next
]) == lex_Nukta
)
129 if ((next
< cChar
) && lex(input
[next
]) == lex_Anudatta
)
132 if ((next
< cChar
) && lex(input
[next
]) == lex_Halant
)
135 if((next
< cChar
) && is_joiner( lex(input
[next
]) ))
138 else if (next
< cChar
)
140 while((next
< cChar
) && is_matra( lex(input
[next
]) ))
142 if ((next
< cChar
) && lex(input
[next
]) == lex_Nukta
)
144 if ((next
< cChar
) && lex(input
[next
]) == lex_Halant
)
147 if ((next
< cChar
) && lex(input
[next
]) == lex_Modifier
)
149 if ((next
< cChar
) && lex(input
[next
]) == lex_Vedic
)
154 static INT
parse_vowel_syllable(LPCWSTR input
, INT cChar
, INT start
,
155 INT next
, lexical_function lex
)
157 if ((next
< cChar
) && lex(input
[next
]) == lex_Nukta
)
159 if ((next
< cChar
) && is_joiner( lex(input
[next
]) ) && lex(input
[next
+1])==lex_Halant
&& is_consonant( lex(input
[next
+2]) ))
161 else if ((next
< cChar
) && lex(input
[next
])==lex_Halant
&& is_consonant( lex(input
[next
+1]) ))
163 else if ((next
< cChar
) && lex(input
[next
])==lex_ZWJ
&& is_consonant( lex(input
[next
+1]) ))
166 if (is_matra( lex(input
[next
]) ))
168 while((next
< cChar
) && is_matra( lex(input
[next
]) ))
170 if ((next
< cChar
) && lex(input
[next
]) == lex_Nukta
)
172 if ((next
< cChar
) && lex(input
[next
]) == lex_Halant
)
176 if ((next
< cChar
) && lex(input
[next
]) == lex_Modifier
)
178 if ((next
< cChar
) && lex(input
[next
]) == lex_Vedic
)
183 static INT
Indic_process_next_syllable( LPCWSTR input
, INT cChar
, INT start
, INT
* main
, INT next
, lexical_function lex
)
185 if (lex(input
[next
])==lex_Vowel
)
188 return parse_vowel_syllable(input
, cChar
, start
, next
+1, lex
);
190 else if ((cChar
> next
+3) && lex(input
[next
]) == lex_Ra
&& lex(input
[next
+1]) == lex_Halant
&& lex(input
[next
+2]) == lex_Vowel
)
193 return parse_vowel_syllable(input
, cChar
, start
, next
+3, lex
);
196 else if (start
== next
&& lex(input
[next
])==lex_NBSP
)
199 return parse_vowel_syllable(input
, cChar
, start
, next
+1, lex
);
201 else if (start
== next
&& (cChar
> next
+3) && lex(input
[next
]) == lex_Ra
&& lex(input
[next
+1]) == lex_Halant
&& lex(input
[next
+2]) == lex_NBSP
)
204 return parse_vowel_syllable(input
, cChar
, start
, next
+3, lex
);
207 return parse_consonant_syllable(input
, cChar
, start
, main
, next
, lex
);
210 static BOOL
Consonent_is_post_base_form(HDC hdc
, SCRIPT_ANALYSIS
*psa
, ScriptCache
* psc
, LPWSTR pwChar
, IndicSyllable
*s
, lexical_function lexical
, BOOL modern
)
212 if (is_consonant(lexical(pwChar
[s
->base
])) && s
->base
> s
->start
&& lexical(pwChar
[s
->base
-1]) == lex_Halant
)
215 return (SHAPE_does_GSUB_feature_apply_to_chars(hdc
, psa
, psc
, &pwChar
[s
->base
-1], 1, 2, "pstf") > 0);
219 cc
[0] = pwChar
[s
->base
];
220 cc
[1] = pwChar
[s
->base
-1];
221 return (SHAPE_does_GSUB_feature_apply_to_chars(hdc
, psa
, psc
, cc
, 1, 2, "pstf") > 0);
227 static BOOL
Consonent_is_below_base_form(HDC hdc
, SCRIPT_ANALYSIS
*psa
, ScriptCache
* psc
, LPWSTR pwChar
, IndicSyllable
*s
, lexical_function lexical
, BOOL modern
)
229 if (is_consonant(lexical(pwChar
[s
->base
])) && s
->base
> s
->start
&& lexical(pwChar
[s
->base
-1]) == lex_Halant
)
232 return (SHAPE_does_GSUB_feature_apply_to_chars(hdc
, psa
, psc
, &pwChar
[s
->base
-1], 1, 2, "blwf") > 0);
236 cc
[0] = pwChar
[s
->base
];
237 cc
[1] = pwChar
[s
->base
-1];
238 return (SHAPE_does_GSUB_feature_apply_to_chars(hdc
, psa
, psc
, cc
, 1, 2, "blwf") > 0);
244 static BOOL
Consonent_is_pre_base_form(HDC hdc
, SCRIPT_ANALYSIS
*psa
, ScriptCache
* psc
, LPWSTR pwChar
, IndicSyllable
*s
, lexical_function lexical
, BOOL modern
)
246 if (is_consonant(lexical(pwChar
[s
->base
])) && s
->base
> s
->start
&& lexical(pwChar
[s
->base
-1]) == lex_Halant
)
249 return (SHAPE_does_GSUB_feature_apply_to_chars(hdc
, psa
, psc
, &pwChar
[s
->base
-1], 1, 2, "pref") > 0);
253 cc
[0] = pwChar
[s
->base
];
254 cc
[1] = pwChar
[s
->base
-1];
255 return (SHAPE_does_GSUB_feature_apply_to_chars(hdc
, psa
, psc
, cc
, 1, 2, "pref") > 0);
261 static BOOL
Consonent_is_ralf(HDC hdc
, SCRIPT_ANALYSIS
*psa
, ScriptCache
* psc
, LPWSTR pwChar
, IndicSyllable
*s
, lexical_function lexical
)
263 if ((lexical(pwChar
[s
->start
])==lex_Ra
) && s
->end
> s
->start
&& lexical(pwChar
[s
->start
+1]) == lex_Halant
)
264 return (SHAPE_does_GSUB_feature_apply_to_chars(hdc
, psa
, psc
, &pwChar
[s
->start
], 1, 2, "rphf") > 0);
268 static int FindBaseConsonant(HDC hdc
, SCRIPT_ANALYSIS
*psa
, ScriptCache
* psc
, LPWSTR input
, IndicSyllable
*s
, lexical_function lex
, BOOL modern
)
274 /* remove ralf from consideration */
275 if (Consonent_is_ralf(hdc
, psa
, psc
, input
, s
, lex
))
281 /* try to find a base consonant */
282 if (!is_consonant( lex(input
[s
->base
]) ))
284 for (i
= s
->end
; i
>= s
->start
; i
--)
285 if (is_consonant( lex(input
[i
]) ))
292 while ((blwf
= Consonent_is_below_base_form(hdc
, psa
, psc
, input
, s
, lex
, modern
)) || Consonent_is_post_base_form(hdc
, psa
, psc
, input
, s
, lex
, modern
) || (pref
= Consonent_is_pre_base_form(hdc
, psa
, psc
, input
, s
, lex
, modern
)))
294 if (blwf
&& s
->blwf
== -1)
295 s
->blwf
= s
->base
- 1;
296 if (pref
&& s
->pref
== -1)
297 s
->pref
= s
->base
- 1;
299 for (i
= s
->base
-1; i
>= s
->start
; i
--)
300 if (is_consonant( lex(input
[i
]) ))
310 if (s
->ralf
== s
->base
)
316 void Indic_ReorderCharacters( HDC hdc
, SCRIPT_ANALYSIS
*psa
, ScriptCache
* psc
, LPWSTR input
, int cChar
, IndicSyllable
**syllables
, int *syllable_count
, lexical_function lex
, reorder_function reorder_f
, BOOL modern
)
324 if (!lex
|| ! reorder_f
)
326 ERR("Failure to have required functions\n");
330 debug_output_string(input
, cChar
, lex
);
333 while((next
< cChar
) && lex(input
[next
]) == lex_Generic
)
336 next
= Indic_process_next_syllable(input
, cChar
, 0, ¢er
, index
, lex
);
340 *syllables
= HeapReAlloc(GetProcessHeap(),0,*syllables
, sizeof(IndicSyllable
)*(*syllable_count
+1));
342 *syllables
= HeapAlloc(GetProcessHeap(),0,sizeof(IndicSyllable
));
343 (*syllables
)[*syllable_count
].start
= index
;
344 (*syllables
)[*syllable_count
].base
= center
;
345 (*syllables
)[*syllable_count
].ralf
= -1;
346 (*syllables
)[*syllable_count
].blwf
= -1;
347 (*syllables
)[*syllable_count
].pref
= -1;
348 (*syllables
)[*syllable_count
].end
= next
-1;
349 FindBaseConsonant(hdc
, psa
, psc
, input
, &(*syllables
)[*syllable_count
], lex
, modern
);
350 reorder_f(input
, &(*syllables
)[*syllable_count
], lex
);
352 *syllable_count
= (*syllable_count
)+1;
354 else if (index
< cChar
)
356 TRACE("Processing failed at %i\n",index
);
360 TRACE("Processed %i of %i characters into %i syllables\n",index
,cChar
,*syllable_count
);