diff -r 000000000000 -r 1fb32624e06b fontservices/textshaperplugin/IcuSource/layout/IndicClassTables.cpp --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/fontservices/textshaperplugin/IcuSource/layout/IndicClassTables.cpp Tue Feb 02 02:02:46 2010 +0200 @@ -0,0 +1,406 @@ +/* + * + * (C) Copyright IBM Corp. 1998-2005 - All Rights Reserved + * + */ + +#include "LETypes.h" +#include "LEScripts.h" +#include "OpenTypeTables.h" +#include "OpenTypeUtilities.h" +#include "IndicReordering.h" + +U_NAMESPACE_BEGIN + +#define ARRAY_SIZE(array) (sizeof array / sizeof array[0]) + +// Split matra table indices +#define _x1 (1 << CF_INDEX_SHIFT) +#define _x2 (2 << CF_INDEX_SHIFT) +#define _x3 (3 << CF_INDEX_SHIFT) +#define _x4 (4 << CF_INDEX_SHIFT) +#define _x5 (5 << CF_INDEX_SHIFT) +#define _x6 (6 << CF_INDEX_SHIFT) +#define _x7 (7 << CF_INDEX_SHIFT) +#define _x8 (8 << CF_INDEX_SHIFT) +#define _x9 (9 << CF_INDEX_SHIFT) + +// simple classes +#define _xx (CC_RESERVED) +#define _ma (CC_VOWEL_MODIFIER | CF_POS_ABOVE) +#define _mp (CC_VOWEL_MODIFIER | CF_POS_AFTER) +#define _sa (CC_STRESS_MARK | CF_POS_ABOVE) +#define _sb (CC_STRESS_MARK | CF_POS_BELOW) +#define _iv (CC_INDEPENDENT_VOWEL) +// Added by Nokia -- Devanagai independent vowel A -> special case +#define _ia (CC_INDEPENDENT_VOWEL_A) + +#define _i2 (CC_INDEPENDENT_VOWEL_2) +#define _ct (CC_CONSONANT | CF_CONSONANT) +#define _cn (CC_CONSONANT_WITH_NUKTA | CF_CONSONANT) +// Added by Nokia -- Special case Kannada and Bengali Ra to satisfy usecase (Initial)Ra + ZWJ + Virama + Consonant +#define _cr (CC_CONSONANT_KANNADA_BENGALI_RA | CF_CONSONANT) + +#define _nu (CC_NUKTA) +#define _dv (CC_DEPENDENT_VOWEL) +// Added by Nokia -- Devanagari vowel CANDRA E -> special case with independent vowel A +#define _de (CC_DEPENDENT_VOWEL_CANDRA_E | CF_POS_ABOVE) + +// Added by Nokia -- Gurmukhi bearer ARA +#define _ba (CC_GUR_BEARER_A) +// Added by Nokia -- Gurmukhi bearer IRI +#define _bi (CC_GUR_BEARER_I) +// Added by Nokia -- Gurmukhi bearer URA +#define _bu (CC_GUR_BEARER_U) + +// Added by Nokia -- Gurmukhi dependent vowel which combined with bearer ARA to form an independent vowel +#define _ga (CC_GUR_DEPENDENT_VOWEL_A) +// Added by Nokia -- Gurmukhi dependent vowel which combined with bearer IRI to form an independent vowel +#define _gi (CC_GUR_DEPENDENT_VOWEL_I) +// Added by Nokia -- Gurmukhi dependent vowel which combined with bearer URA to form an independent vowel +#define _gu (CC_GUR_DEPENDENT_VOWEL_U) + +#define _gar (_ga | CF_POS_AFTER) +#define _gaa (_ga | CF_POS_ABOVE) +#define _gil (_gi | CF_POS_BEFORE) +#define _gir (_gi | CF_POS_AFTER) +#define _gia (_gi | CF_POS_ABOVE) +#define _gub (_gu | CF_POS_BELOW) +#define _gua (_gu | CF_POS_ABOVE) + +#define _dl (_dv | CF_POS_BEFORE) +#define _db (_dv | CF_POS_BELOW) +#define _da (_dv | CF_POS_ABOVE) +#define _dr (_dv | CF_POS_AFTER) +#define _lm (_dv | CF_LENGTH_MARK) +#define _l1 (CC_SPLIT_VOWEL_PIECE_1 | CF_POS_BEFORE) +#define _a1 (CC_SPLIT_VOWEL_PIECE_1 | CF_POS_ABOVE) +#define _r2 (CC_SPLIT_VOWEL_PIECE_2 | CF_POS_AFTER) +#define _m2 (CC_SPLIT_VOWEL_PIECE_2 | CF_LENGTH_MARK) +#define _m3 (CC_SPLIT_VOWEL_PIECE_3 | CF_LENGTH_MARK) +#define _vr (CC_VIRAMA) + +// split matras +#define _s1 (_dv | _x1) +#define _s2 (_dv | _x2) +#define _s3 (_dv | _x3) +#define _s4 (_dv | _x4) +#define _s5 (_dv | _x5) +#define _s6 (_dv | _x6) +#define _s7 (_dv | _x7) +#define _s8 (_dv | _x8) +#define _s9 (_dv | _x9) + +// consonants with special forms +// NOTE: this assumes that no consonants with nukta have +// special forms... (Bengali RA?) +#define _bb (_ct | CF_BELOW_BASE) +#define _pb (_ct | CF_POST_BASE) +#define _vt (_bb | CF_VATTU) +#define _rv (_vt | CF_REPH) +#define _rp (_pb | CF_REPH) +#define _rb (_bb | CF_REPH) + +// Added by Nokia -- for Kannada Ra & Bengali Ra +#define _b2 (_cr | CF_BELOW_BASE) +#define _kr (_b2 | CF_REPH) + +// Added by Nokia -- for Bengali Ya +#define _by (_pb | CF_CONSONANT_BENGALI_YA) + +// Added by Nokia -- for Tamil independent vowel O +#define _to (CC_INDEPENDENT_VOWEL_TAMIL_O | CF_POS_AFTER) + +// Added by Nokia -- for Malayalam CHILLU +#define _pbc (_pb | CF_CONSONANT_MLYL_CHILLU) +#define _bbc (_bb | CF_CONSONANT_MLYL_CHILLU) +#define _ctc (_ct | CF_CONSONANT_MLYL_CHILLU) + +// +// Character class tables +// +static const IndicClassTable::CharClass devaCharClasses[] = +{ + _xx, _ma, _ma, _mp, _iv, _ia, _iv, _iv, _iv, _iv, _iv, _iv, _iv, _iv, _iv, _iv, // 0900 - 090F + _iv, _iv, _iv, _iv, _iv, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, // 0910 - 091F + _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _cn, _ct, _ct, _ct, _ct, _ct, _ct, // 0920 - 092F + _rv, _cn, _ct, _ct, _cn, _ct, _ct, _ct, _ct, _ct, _xx, _xx, _nu, _xx, _dr, _dl, // 0930 - 093F + _dr, _db, _db, _db, _db, _de, _da, _da, _da, _dr, _dr, _dr, _dr, _vr, _xx, _xx, // 0940 - 094F + _xx, _sa, _sb, _sa, _sa, _xx, _xx, _xx, _cn, _cn, _cn, _cn, _cn, _cn, _cn, _cn, // 0950 - 095F + _iv, _iv, _db, _db, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, // 0960 - 096F + _xx // 0970 +}; + +// Nokia: Changed 09AF character from _pb to _by which represents a Bengali Ya which may or may not take +// post base form, depending on its position within a consonant cluster. +// Also changed 09B0 from _rb to special case _kr to satiisfy the usecase Ra + ZWJ + Virama + Consonant. +// Changed 09E2 and 09E3 to _db here, not _dv since the Bengali rendering rules want them below +// the base consonant. +static const IndicClassTable::CharClass bengCharClasses[] = +{ + _xx, _ma, _mp, _mp, _xx, _i2, _iv, _iv, _iv, _iv, _iv, _iv, _iv, _xx, _xx, _i2, // 0980 - 098F + _iv, _xx, _xx, _iv, _iv, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, // 0990 - 099F + _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _xx, _ct, _ct, _bb, _ct, _ct, _by, // 09A0 - 09AF + _kr, _xx, _ct, _xx, _xx, _xx, _ct, _ct, _ct, _ct, _xx, _xx, _nu, _xx, _r2, _dl, // 09B0 - 09BF + _dr, _db, _db, _db, _db, _xx, _xx, _l1, _dl, _xx, _xx, _s1, _s2, _vr, _xx, _xx, // 09C0 - 09CF + _xx, _xx, _xx, _xx, _xx, _xx, _xx, _m2, _xx, _xx, _xx, _xx, _cn, _cn, _xx, _cn, // 09D0 - 09DF + _iv, _iv, _db, _db, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, // 09E0 - 09EF + _ct, _ct, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx // 09F0 - 09FA +}; + +// Nokia: Changed 0A05 from _iv to _ba and 0A72&0A73 from _ct to _bi&_bu, changed 0A3E,0A48&0A4C +// from _d* to _ga*, 0A3F,0A40&0A47 from _d* to _gi*, 0A41,0A42&0A4B from _d* to _gu* to support +// decompositions of the independent vowels into a sequence of a vowel bearer and a depending vowel sign. +// Also changed 0A2F from _bb to _pb since YA is expected to post-base when combined in constant cluster. +static const IndicClassTable::CharClass punjCharClasses[] = +{ + _xx, _ma, _ma, _mp, _xx, _ba, _iv, _iv, _iv, _iv, _iv, _xx, _xx, _xx, _xx, _iv, // 0A00 - 0A0F + _iv, _xx, _xx, _iv, _iv, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, // 0A10 - 0A1F + _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _xx, _ct, _ct, _ct, _ct, _ct, _pb, // 0A20 - 0A2F + _vt, _xx, _ct, _cn, _xx, _bb, _cn, _xx, _ct, _bb, _xx, _xx, _nu, _xx, _gar, _gil, // 0A30 - 0A3F + _gir, _gub, _gub, _xx, _xx, _xx, _xx, _gia, _gaa, _xx, _xx, _gua, _gaa, _vr, _xx, _xx, // 0A40 - 0A4F + _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _cn, _cn, _cn, _ct, _xx, _cn, _xx, // 0A50 - 0A5F + _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, // 0A60 - 0A6F + _ma, _ma, _bi, _bu, _xx // 0A70 - 0A74 +}; + +static const IndicClassTable::CharClass gujrCharClasses[] = +{ + _xx, _ma, _ma, _mp, _xx, _iv, _iv, _iv, _iv, _iv, _iv, _iv, _xx, _iv, _xx, _iv, // 0A80 - 0A8F + _iv, _iv, _xx, _iv, _iv, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, // 0A90 - 0A9F + _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _xx, _ct, _ct, _ct, _ct, _ct, _ct, // 0AA0 - 0AAF + _rv, _xx, _ct, _ct, _xx, _ct, _ct, _ct, _ct, _ct, _xx, _xx, _nu, _xx, _dr, _dl, // 0AB0 - 0ABF + _dr, _db, _db, _db, _db, _da, _xx, _da, _da, _dr, _xx, _dr, _dr, _vr, _xx, _xx, // 0AC0 - 0ACF + _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, // 0AD0 - 0ADF + _iv, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx // 0AE0 - 0AEF +}; + +static const IndicClassTable::CharClass oryaCharClasses[] = +{ + _xx, _ma, _mp, _mp, _xx, _iv, _iv, _iv, _iv, _iv, _iv, _iv, _iv, _xx, _xx, _iv, // 0B00 - 0B0F + _iv, _xx, _xx, _iv, _iv, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, // 0B10 - 0B1F + _ct, _ct, _ct, _ct, _bb, _ct, _ct, _ct, _bb, _xx, _ct, _ct, _bb, _bb, _bb, _pb, // 0B20 - 0B2F + _rb, _xx, _bb, _bb, _xx, _ct, _ct, _ct, _ct, _ct, _xx, _xx, _nu, _xx, _r2, _da, // 0B30 - 0B3F + _dr, _db, _db, _db, _xx, _xx, _xx, _l1, _s1, _xx, _xx, _s2, _s3, _vr, _xx, _xx, // 0B40 - 0B4F + _xx, _xx, _xx, _xx, _xx, _xx, _m2, _m2, _xx, _xx, _xx, _xx, _cn, _cn, _xx, _cn, // 0B50 - 0B5F + _iv, _iv, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, // 0B60 - 0B6F + _xx, _ct // 0B70 - 0B71 +}; + +// Nokia: Changed 0B92 to _to for the exceptional case of Tamil independent vowel O +static const IndicClassTable::CharClass tamlCharClasses[] = +{ + _xx, _xx, _ma, _mp, _xx, _iv, _iv, _iv, _iv, _iv, _iv, _xx, _xx, _xx, _iv, _iv, // 0B80 - 0B8F + _iv, _xx, _to, _iv, _iv, _ct, _xx, _xx, _xx, _ct, _ct, _xx, _ct, _xx, _ct, _ct, // 0B90 - 0B9F + _xx, _xx, _xx, _ct, _ct, _xx, _xx, _xx, _ct, _ct, _ct, _xx, _xx, _xx, _ct, _ct, // 0BA0 - 0BAF + _ct, _ct, _ct, _ct, _ct, _ct, _xx, _ct, _ct, _ct, _xx, _xx, _xx, _xx, _r2, _dr, // 0BB0 - 0BBF + _da, _dr, _dr, _xx, _xx, _xx, _l1, _l1, _dl, _xx, _s1, _s2, _s3, _vr, _xx, _xx, // 0BC0 - 0BCF + _xx, _xx, _xx, _xx, _xx, _xx, _xx, _m2, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, // 0BD0 - 0BDF + _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, // 0BE0 - 0BEF + _xx, _xx, _xx // 0BF0 - 0BF2 +}; + +// FIXME: Should some of the bb's be pb's? (KA, NA, MA, YA, VA, etc. (approx 13)) +// Nokia: Changed 0C31 character to _bb as with _ct the consonant cluster is not rendered correctly. +static const IndicClassTable::CharClass teluCharClasses[] = +{ + _xx, _mp, _mp, _mp, _xx, _iv, _iv, _iv, _iv, _iv, _iv, _iv, _iv, _xx, _iv, _iv, // 0C00 - 0C0F + _iv, _xx, _iv, _iv, _iv, _bb, _bb, _bb, _bb, _bb, _bb, _bb, _bb, _bb, _bb, _bb, // 0C10 - 0C1F + _bb, _bb, _bb, _bb, _bb, _bb, _bb, _bb, _bb, _xx, _bb, _bb, _bb, _bb, _bb, _bb, // 0C20 - 0C2F + _bb, _bb, _bb, _bb, _xx, _bb, _bb, _bb, _bb, _bb, _xx, _xx, _xx, _xx, _da, _da, // 0C30 - 0C3F + _da, _dr, _dr, _dr, _dr, _xx, _a1, _da, _s1, _xx, _da, _da, _da, _vr, _xx, _xx, // 0C40 - 0C4F + _xx, _xx, _xx, _xx, _xx, _da, _m2, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, // 0C50 - 0C5F + _iv, _iv, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx // 0C60 - 0C6F +}; + +// U+CC3 and U+CC4 are _lm here not _dr since the Kannada rendering +// rules want them below and to the right of the entire cluster +// +// There's some information about this in: +// +// http://brahmi.sourceforge.net/docs/KannadaComputing.html + +// Nokia: Changed 0CBC character class to _nu (NUKTA), as post Unicode 4.0, this +// code point is classified as the Kannada NUKTA character. +// Also changed 0CB0 from _rb to special case _cr to satiisfy the usecase Ra + ZWJ + Virama + Consonant +// see _cr and _rb above +static const IndicClassTable::CharClass kndaCharClasses[] = +{ + _xx, _xx, _mp, _mp, _xx, _iv, _iv, _iv, _iv, _iv, _iv, _iv, _iv, _xx, _iv, _iv, // 0C80 - 0C8F + _iv, _xx, _iv, _iv, _iv, _bb, _bb, _bb, _bb, _bb, _bb, _bb, _bb, _bb, _bb, _bb, // 0C90 - 0C9F + _bb, _bb, _bb, _bb, _bb, _bb, _bb, _bb, _bb, _xx, _bb, _bb, _bb, _bb, _bb, _bb, // 0CA0 - 0CAF + _kr, _ct, _bb, _bb, _xx, _bb, _bb, _bb, _bb, _bb, _xx, _xx, _nu, _xx, _dr, _da, // 0CB0 - 0CBF + _s1, _dr, _r2, _lm, _lm, _xx, _a1, _s2, _s3, _xx, _s4, _s5, _da, _vr, _xx, _xx, // 0CC0 - 0CCF + _xx, _xx, _xx, _xx, _xx, _m3, _m2, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _bb, _xx, // 0CD0 - 0CDF + _iv, _iv, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx // 0CE0 - 0CEF +}; + +// FIXME: this is correct for old-style Malayalam (MAL) but not for reformed Malayalam (MLR) +// FIXME: should there be a REPH for old-style Malayalam? +static const IndicClassTable::CharClass mlymCharClasses[] = +{ + _xx, _xx, _mp, _mp, _xx, _iv, _iv, _iv, _iv, _iv, _iv, _iv, _iv, _xx, _iv, _iv, // 0D00 - 0D0F +// _iv, _xx, _iv, _iv, _iv, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _bb, // 0D10 - 0D1F //old + _iv, _xx, _iv, _iv, _iv, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, _ct, // 0D10 - 0D1F +// _ct, _ct, _ct, _bb, _ct, _bb, _bb, _ct, _ct, _xx, _ct, _ct, _ct, _ct, _ct, _pb, // 0D20 - 0D2F //old + _ct, _ct, _ct, _ctc, _ct, _ct, _ct, _ct, _ctc, _xx, _ct, _ct, _ct, _ct, _ct, _pb, // 0D20 - 0D2F +// _pb, _cn, _bb, _ct, _ct, _pb, _ct, _ct, _ct, _ct, _xx, _xx, _xx, _xx, _r2, _dr, // 0D30 - 0D3F //old + _pbc, _cn, _bbc, _ctc, _ct, _pb, _ct, _ct, _ct, _ct, _xx, _xx, _xx, _xx, _r2, _dr, // 0D30 - 0D3F + _dr, _dr, _dr, _dr, _xx, _xx, _l1, _l1, _dl, _xx, _s1, _s2, _s3, _vr, _xx, _xx, // 0D40 - 0D4F + _xx, _xx, _xx, _xx, _xx, _xx, _xx, _m2, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, // 0D50 - 0D5F + _iv, _iv, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx, _xx // 0D60 - 0D6F +}; + +// +// Split matra tables +// +static const SplitMatra bengSplitTable[] = {{0x09C7, 0x09BE}, {0x09C7, 0x09D7}}; + +static const SplitMatra oryaSplitTable[] = {{0x0B47, 0x0B56}, {0x0B47, 0x0B3E}, {0x0B47, 0x0B57}}; + +static const SplitMatra tamlSplitTable[] = {{0x0BC6, 0x0BBE}, {0x0BC7, 0x0BBE}, {0x0BC6, 0x0BD7}}; + +static const SplitMatra teluSplitTable[] = {{0x0C46, 0x0C56}}; + +static const SplitMatra kndaSplitTable[] = {{0x0CBF, 0x0CD5}, {0x0CC6, 0x0CD5}, {0x0CC6, 0x0CD6}, {0x0CC6, 0x0CC2}, + {0x0CC6, 0x0CC2, 0x0CD5}}; + +static const SplitMatra mlymSplitTable[] = {{0x0D46, 0x0D3E}, {0x0D47, 0x0D3E}, {0x0D46, 0x0D57}}; + +// +// Script Flags +// + +// FIXME: post 'GSUB' reordering of MATRA_PRE's for Malayalam and Tamil +// FIXME: reformed Malayalam needs to reorder VATTU to before base glyph... +// FIXME: eyelash RA only for Devanagari?? +#define DEVA_SCRIPT_FLAGS (SF_EYELASH_RA | SF_NO_POST_BASE_LIMIT) +#define BENG_SCRIPT_FLAGS (SF_REPH_AFTER_BELOW | SF_NO_POST_BASE_LIMIT) +#define PUNJ_SCRIPT_FLAGS (SF_NO_POST_BASE_LIMIT | SF_MATRAS_AFTER_POSTBASE) +#define GUJR_SCRIPT_FLAGS (SF_NO_POST_BASE_LIMIT) +#define ORYA_SCRIPT_FLAGS (SF_REPH_AFTER_BELOW | SF_NO_POST_BASE_LIMIT) +#define TAML_SCRIPT_FLAGS (SF_MPRE_FIXUP | SF_NO_POST_BASE_LIMIT) +#define TELU_SCRIPT_FLAGS (SF_MATRAS_AFTER_BASE | 3) +#define KNDA_SCRIPT_FLAGS (SF_MATRAS_AFTER_BASE | 3) +#define MLYM_SCRIPT_FLAGS (SF_MPRE_FIXUP | SF_NO_POST_BASE_LIMIT) + +// +// Indic Class Tables +// +static const IndicClassTable devaClassTable = {0x0900, 0x0970, 2, DEVA_SCRIPT_FLAGS, devaCharClasses, NULL}; + +static const IndicClassTable bengClassTable = {0x0980, 0x09FA, 3, BENG_SCRIPT_FLAGS, bengCharClasses, bengSplitTable}; + +static const IndicClassTable punjClassTable = {0x0A00, 0x0A74, 2, PUNJ_SCRIPT_FLAGS, punjCharClasses, NULL}; + +static const IndicClassTable gujrClassTable = {0x0A80, 0x0AEF, 2, GUJR_SCRIPT_FLAGS, gujrCharClasses, NULL}; + +static const IndicClassTable oryaClassTable = {0x0B00, 0x0B71, 3, ORYA_SCRIPT_FLAGS, oryaCharClasses, oryaSplitTable}; + +static const IndicClassTable tamlClassTable = {0x0B80, 0x0BF2, 3, TAML_SCRIPT_FLAGS, tamlCharClasses, tamlSplitTable}; + +static const IndicClassTable teluClassTable = {0x0C00, 0x0C6F, 3, TELU_SCRIPT_FLAGS, teluCharClasses, teluSplitTable}; + +static const IndicClassTable kndaClassTable = {0x0C80, 0x0CEF, 4, KNDA_SCRIPT_FLAGS, kndaCharClasses, kndaSplitTable}; + +static const IndicClassTable mlymClassTable = {0x0D00, 0x0D6F, 3, MLYM_SCRIPT_FLAGS, mlymCharClasses, mlymSplitTable}; + +// +// IndicClassTable addresses +// +static const IndicClassTable * const indicClassTables[] = { + NULL, /* 'zyyy' (COMMON) */ + NULL, /* 'qaai' (INHERITED) */ + NULL, /* 'arab' (ARABIC) */ + NULL, /* 'armn' (ARMENIAN) */ + &bengClassTable, /* 'beng' (BENGALI) */ + NULL, /* 'bopo' (BOPOMOFO) */ + NULL, /* 'cher' (CHEROKEE) */ + NULL, /* 'qaac' (COPTIC) */ + NULL, /* 'cyrl' (CYRILLIC) */ + NULL, /* 'dsrt' (DESERET) */ + &devaClassTable, /* 'deva' (DEVANAGARI) */ + NULL, /* 'ethi' (ETHIOPIC) */ + NULL, /* 'geor' (GEORGIAN) */ + NULL, /* 'goth' (GOTHIC) */ + NULL, /* 'grek' (GREEK) */ + &gujrClassTable, /* 'gujr' (GUJARATI) */ + &punjClassTable, /* 'guru' (GURMUKHI) */ + NULL, /* 'hani' (HAN) */ + NULL, /* 'hang' (HANGUL) */ + NULL, /* 'hebr' (HEBREW) */ + NULL, /* 'hira' (HIRAGANA) */ + &kndaClassTable, /* 'knda' (KANNADA) */ + NULL, /* 'kata' (KATAKANA) */ + NULL, /* 'khmr' (KHMER) */ + NULL, /* 'laoo' (LAO) */ + NULL, /* 'latn' (LATIN) */ + &mlymClassTable, /* 'mlym' (MALAYALAM) */ + NULL, /* 'mong' (MONGOLIAN) */ + NULL, /* 'mymr' (MYANMAR) */ + NULL, /* 'ogam' (OGHAM) */ + NULL, /* 'ital' (OLD-ITALIC) */ + &oryaClassTable, /* 'orya' (ORIYA) */ + NULL, /* 'runr' (RUNIC) */ + NULL, /* 'sinh' (SINHALA) */ + NULL, /* 'syrc' (SYRIAC) */ + &tamlClassTable, /* 'taml' (TAMIL) */ + &teluClassTable, /* 'telu' (TELUGU) */ + NULL, /* 'thaa' (THAANA) */ + NULL, /* 'thai' (THAI) */ + NULL, /* 'tibt' (TIBETAN) */ + NULL, /* 'cans' (CANADIAN-ABORIGINAL) */ + NULL, /* 'yiii' (YI) */ + NULL, /* 'tglg' (TAGALOG) */ + NULL, /* 'hano' (HANUNOO) */ + NULL, /* 'buhd' (BUHID) */ + NULL, /* 'tagb' (TAGBANWA) */ + NULL, /* 'brai' (BRAILLE) */ + NULL, /* 'cprt' (CYPRIOT) */ + NULL, /* 'limb' (LIMBU) */ + NULL, /* 'linb' (LINEAR_B) */ + NULL, /* 'osma' (OSMANYA) */ + NULL, /* 'shaw' (SHAVIAN) */ + NULL, /* 'tale' (TAI_LE) */ + NULL, /* 'ugar' (UGARITIC) */ + NULL /* 'hrkt' (KATAKANA_OR_HIRAGANA) */ +}; + +IndicClassTable::CharClass IndicClassTable::getCharClass(LEUnicode ch) const +{ + if (ch == C_SIGN_ZWJ) { + return CF_CONSONANT | CC_ZERO_WIDTH_MARK; + } + + if (ch == C_SIGN_ZWNJ) { + return CC_ZERO_WIDTH_MARK; + } + + if (ch < firstChar || ch > lastChar) { + return CC_RESERVED; + } + + return classTable[ch - firstChar]; +} + +const IndicClassTable *IndicClassTable::getScriptClassTable(le_int32 scriptCode) +{ + if (scriptCode < 0 || scriptCode >= scriptCodeCount) { + return NULL; + } + + return indicClassTables[scriptCode]; +} + +le_int32 IndicReordering::getWorstCaseExpansion(le_int32 scriptCode) +{ + const IndicClassTable *classTable = IndicClassTable::getScriptClassTable(scriptCode); + + if (classTable == NULL) { + return 1; + } + + return classTable->getWorstCaseExpansion(); +} + +U_NAMESPACE_END