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
||
75 type
== lex_Composed_Vowel
);
78 static inline BOOL
is_joiner( int type
)
80 return (type
== lex_ZWJ
|| type
== lex_ZWNJ
);
83 static INT
consonant_header(LPCWSTR input
, INT cChar
, INT start
, INT next
,
86 if (!is_consonant( lex(input
[next
]) )) return -1;
88 if ((next
< cChar
) && lex(input
[next
]) == lex_Nukta
)
90 if ((next
< cChar
) && lex(input
[next
])==lex_Halant
)
93 if((next
< cChar
) && is_joiner( lex(input
[next
]) ))
95 if ((next
< cChar
) && is_consonant( lex(input
[next
]) ))
98 else if ((next
< cChar
) && is_joiner( lex(input
[next
]) ) && lex(input
[next
+1])==lex_Halant
)
101 if ((next
< cChar
) && is_consonant( lex(input
[next
]) ))
107 static INT
parse_consonant_syllable(LPCWSTR input
, INT cChar
, INT start
,
108 INT
*main
, INT next
, lexical_function lex
)
114 check
= consonant_header(input
,cChar
,start
,next
,lex
);
120 } while (check
!= -1);
121 if (headers
|| is_consonant( lex(input
[next
]) ))
128 if ((next
< cChar
) && lex(input
[next
]) == lex_Nukta
)
130 if ((next
< cChar
) && lex(input
[next
]) == lex_Anudatta
)
133 if ((next
< cChar
) && lex(input
[next
]) == lex_Halant
)
136 if((next
< cChar
) && is_joiner( lex(input
[next
]) ))
139 else if (next
< cChar
)
141 while((next
< cChar
) && is_matra( lex(input
[next
]) ))
143 if ((next
< cChar
) && lex(input
[next
]) == lex_Nukta
)
145 if ((next
< cChar
) && lex(input
[next
]) == lex_Halant
)
148 if ((next
< cChar
) && lex(input
[next
]) == lex_Modifier
)
150 if ((next
< cChar
) && lex(input
[next
]) == lex_Vedic
)
155 static INT
parse_vowel_syllable(LPCWSTR input
, INT cChar
, INT start
,
156 INT next
, lexical_function lex
)
158 if ((next
< cChar
) && lex(input
[next
]) == lex_Nukta
)
160 if ((next
< cChar
) && is_joiner( lex(input
[next
]) ) && lex(input
[next
+1])==lex_Halant
&& is_consonant( lex(input
[next
+2]) ))
162 else if ((next
< cChar
) && lex(input
[next
])==lex_Halant
&& is_consonant( lex(input
[next
+1]) ))
164 else if ((next
< cChar
) && lex(input
[next
])==lex_ZWJ
&& is_consonant( lex(input
[next
+1]) ))
167 if ((next
< cChar
) && is_matra( lex(input
[next
]) ))
169 while((next
< cChar
) && is_matra( lex(input
[next
]) ))
171 if ((next
< cChar
) && lex(input
[next
]) == lex_Nukta
)
173 if ((next
< cChar
) && lex(input
[next
]) == lex_Halant
)
177 if ((next
< cChar
) && lex(input
[next
]) == lex_Modifier
)
179 if ((next
< cChar
) && lex(input
[next
]) == lex_Vedic
)
184 static INT
Indic_process_next_syllable( LPCWSTR input
, INT cChar
, INT start
, INT
* main
, INT next
, lexical_function lex
)
186 if (lex(input
[next
])==lex_Vowel
)
189 return parse_vowel_syllable(input
, cChar
, start
, next
+1, lex
);
191 else if ((cChar
> next
+3) && lex(input
[next
]) == lex_Ra
&& lex(input
[next
+1]) == lex_Halant
&& lex(input
[next
+2]) == lex_Vowel
)
194 return parse_vowel_syllable(input
, cChar
, start
, next
+3, lex
);
197 else if (start
== next
&& lex(input
[next
])==lex_NBSP
)
200 return parse_vowel_syllable(input
, cChar
, start
, next
+1, lex
);
202 else if (start
== next
&& (cChar
> next
+3) && lex(input
[next
]) == lex_Ra
&& lex(input
[next
+1]) == lex_Halant
&& lex(input
[next
+2]) == lex_NBSP
)
205 return parse_vowel_syllable(input
, cChar
, start
, next
+3, lex
);
208 return parse_consonant_syllable(input
, cChar
, start
, main
, next
, lex
);
211 static BOOL
Consonant_is_post_base_form(HDC hdc
, SCRIPT_ANALYSIS
*psa
, ScriptCache
* psc
, LPCWSTR pwChar
, IndicSyllable
*s
, lexical_function lexical
, BOOL modern
)
213 if (is_consonant(lexical(pwChar
[s
->base
])) && s
->base
> s
->start
&& lexical(pwChar
[s
->base
-1]) == lex_Halant
)
216 return (SHAPE_does_GSUB_feature_apply_to_chars(hdc
, psa
, psc
, &pwChar
[s
->base
-1], 1, 2, "pstf") > 0);
220 cc
[0] = pwChar
[s
->base
];
221 cc
[1] = pwChar
[s
->base
-1];
222 return (SHAPE_does_GSUB_feature_apply_to_chars(hdc
, psa
, psc
, cc
, 1, 2, "pstf") > 0);
228 static BOOL
Consonant_is_below_base_form(HDC hdc
, SCRIPT_ANALYSIS
*psa
, ScriptCache
* psc
, LPCWSTR pwChar
, IndicSyllable
*s
, lexical_function lexical
, BOOL modern
)
230 if (is_consonant(lexical(pwChar
[s
->base
])) && s
->base
> s
->start
&& lexical(pwChar
[s
->base
-1]) == lex_Halant
)
233 return (SHAPE_does_GSUB_feature_apply_to_chars(hdc
, psa
, psc
, &pwChar
[s
->base
-1], 1, 2, "blwf") > 0);
237 cc
[0] = pwChar
[s
->base
];
238 cc
[1] = pwChar
[s
->base
-1];
239 return (SHAPE_does_GSUB_feature_apply_to_chars(hdc
, psa
, psc
, cc
, 1, 2, "blwf") > 0);
245 static BOOL
Consonant_is_pre_base_form(HDC hdc
, SCRIPT_ANALYSIS
*psa
, ScriptCache
* psc
, LPCWSTR pwChar
, IndicSyllable
*s
, lexical_function lexical
, BOOL modern
)
247 if (is_consonant(lexical(pwChar
[s
->base
])) && s
->base
> s
->start
&& lexical(pwChar
[s
->base
-1]) == lex_Halant
)
250 return (SHAPE_does_GSUB_feature_apply_to_chars(hdc
, psa
, psc
, &pwChar
[s
->base
-1], 1, 2, "pref") > 0);
254 cc
[0] = pwChar
[s
->base
];
255 cc
[1] = pwChar
[s
->base
-1];
256 return (SHAPE_does_GSUB_feature_apply_to_chars(hdc
, psa
, psc
, cc
, 1, 2, "pref") > 0);
262 static BOOL
Consonant_is_ralf(HDC hdc
, SCRIPT_ANALYSIS
*psa
, ScriptCache
* psc
, LPCWSTR pwChar
, IndicSyllable
*s
, lexical_function lexical
)
264 if ((lexical(pwChar
[s
->start
])==lex_Ra
) && s
->end
> s
->start
&& lexical(pwChar
[s
->start
+1]) == lex_Halant
)
265 return (SHAPE_does_GSUB_feature_apply_to_chars(hdc
, psa
, psc
, &pwChar
[s
->start
], 1, 2, "rphf") > 0);
269 static int FindBaseConsonant(HDC hdc
, SCRIPT_ANALYSIS
*psa
, ScriptCache
* psc
, LPCWSTR input
, IndicSyllable
*s
, lexical_function lex
, BOOL modern
)
275 /* remove ralf from consideration */
276 if (Consonant_is_ralf(hdc
, psa
, psc
, input
, s
, lex
))
282 /* try to find a base consonant */
283 if (!is_consonant( lex(input
[s
->base
]) ))
285 for (i
= s
->end
; i
>= s
->start
; i
--)
286 if (is_consonant( lex(input
[i
]) ))
293 while ((blwf
= Consonant_is_below_base_form(hdc
, psa
, psc
, input
, s
, lex
, modern
)) || Consonant_is_post_base_form(hdc
, psa
, psc
, input
, s
, lex
, modern
) || (pref
= Consonant_is_pre_base_form(hdc
, psa
, psc
, input
, s
, lex
, modern
)))
295 if (blwf
&& s
->blwf
== -1)
296 s
->blwf
= s
->base
- 1;
297 if (pref
&& s
->pref
== -1)
298 s
->pref
= s
->base
- 1;
300 for (i
= s
->base
-1; i
>= s
->start
; i
--)
301 if (is_consonant( lex(input
[i
]) ))
311 if (s
->ralf
== s
->base
)
317 void Indic_ParseSyllables( HDC hdc
, SCRIPT_ANALYSIS
*psa
, ScriptCache
* psc
, LPCWSTR input
, const int cChar
, IndicSyllable
**syllables
, int *syllable_count
, lexical_function lex
, BOOL modern
)
327 ERR("Failure to have required functions\n");
331 debug_output_string(input
, cChar
, lex
);
334 while((next
< cChar
) && lex(input
[next
]) == lex_Generic
)
339 next
= Indic_process_next_syllable(input
, cChar
, 0, ¢er
, index
, lex
);
343 *syllables
= HeapReAlloc(GetProcessHeap(),0,*syllables
, sizeof(IndicSyllable
)*(*syllable_count
+1));
345 *syllables
= HeapAlloc(GetProcessHeap(),0,sizeof(IndicSyllable
));
346 (*syllables
)[*syllable_count
].start
= index
;
347 (*syllables
)[*syllable_count
].base
= center
;
348 (*syllables
)[*syllable_count
].ralf
= -1;
349 (*syllables
)[*syllable_count
].blwf
= -1;
350 (*syllables
)[*syllable_count
].pref
= -1;
351 (*syllables
)[*syllable_count
].end
= next
-1;
352 FindBaseConsonant(hdc
, psa
, psc
, input
, &(*syllables
)[*syllable_count
], lex
, modern
);
354 *syllable_count
= (*syllable_count
)+1;
356 else if (index
< cChar
)
358 TRACE("Processing failed at %i\n",index
);
362 TRACE("Processed %i of %i characters into %i syllables\n",index
,cChar
,*syllable_count
);
365 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
)
371 ERR("Failure to have required functions\n");
375 Indic_ParseSyllables(hdc
, psa
, psc
, input
, cChar
, syllables
, syllable_count
, lex
, modern
);
376 for (i
= 0; i
< *syllable_count
; i
++)
377 reorder_f(input
, &(*syllables
)[i
], lex
);