organicmaps/icu
13
Fork 0
mirror of https://github.com/unicode-org/icu.git synced 2025-04-08 06:53:45 +00:00
Code Issues 1 Projects Releases Wiki Activity Actions 2

ICU-4996 characters w/ nukta are never vattu and don't have below- or post-base forms.

Browse source 
X-SVN-Rev: 20677
This commit is contained in:
Eric Mader 2006-11-22 22:50:53 +00:00
parent 87c1c18c6d
commit a07b078753
1 changed files with 24 additions and 8 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

32
icu4c/source/layout/IndicReordering.cpp
View file

@ -535,35 +535,48 @@ le_int32 IndicReordering::reorder(const LEUnicode *chars, le_int32 charCount, le
lastConsonant -= 1;
}
IndicClassTable::CharClass charClass = CC_RESERVED;
IndicClassTable::CharClass nextClass = CC_RESERVED;
le_int32 baseConsonant = lastConsonant;
le_int32 postBase = lastConsonant + 1;
le_int32 postBaseLimit = classTable->scriptFlags & SF_POST_BASE_LIMIT_MASK;
le_bool seenVattu = FALSE;
le_bool seenBelowBaseForm = FALSE;
le_bool hasNukta = FALSE;
le_bool hasBelowBaseForm = FALSE;
le_bool hasPostBaseForm = FALSE;
if (postBase < markStart && classTable->isNukta(chars[postBase])) {
charClass = CC_NUKTA;
postBase += 1;
}
while (baseConsonant > baseLimit) {
IndicClassTable::CharClass charClass = classTable->getCharClass(chars[baseConsonant]);
nextClass = charClass;
hasNukta = IndicClassTable::isNukta(nextClass);
charClass = classTable->getCharClass(chars[baseConsonant]);
hasBelowBaseForm = IndicClassTable::hasBelowBaseForm(charClass) && !hasNukta;
hasPostBaseForm = IndicClassTable::hasPostBaseForm(charClass) && !hasNukta;
if (IndicClassTable::isConsonant(charClass)) {
if (postBaseLimit == 0 || seenVattu ||
(baseConsonant > baseLimit && !classTable->isVirama(chars[baseConsonant - 1])) ||
!IndicClassTable::hasPostOrBelowBaseForm(charClass)) {
!(hasBelowBaseForm || hasPostBaseForm)) {
break;
}
seenVattu = IndicClassTable::isVattu(charClass);
// consonants with nuktas are never vattus
seenVattu = IndicClassTable::isVattu(charClass) && !hasNukta;
if (IndicClassTable::hasPostBaseForm(charClass)) {
// consonants with nuktas never have below- or post-base forms
if (hasPostBaseForm) {
if (seenBelowBaseForm) {
break;
}
postBase = baseConsonant;
} else if (IndicClassTable::hasBelowBaseForm(charClass)) {
} else if (hasBelowBaseForm) {
seenBelowBaseForm = TRUE;
}
@ -591,14 +604,17 @@ le_int32 IndicReordering::reorder(const LEUnicode *chars, le_int32 charCount, le
LEUnicode ch = chars[i];
// Don't put 'blwf' on first consonant.
FeatureMask features = (i == baseLimit? tagArray2 : tagArray1);
IndicClassTable::CharClass charClass = classTable->getCharClass(ch);
charClass = classTable->getCharClass(ch);
nextClass = classTable->getCharClass(chars[i + 1]);
hasNukta = IndicClassTable::isNukta(nextClass);
if (IndicClassTable::isConsonant(charClass)) {
if (IndicClassTable::isVattu(charClass) && supressVattu) {
if (IndicClassTable::isVattu(charClass) && !hasNukta && supressVattu) {
features = tagArray4;
}
supressVattu = IndicClassTable::isVattu(charClass);
supressVattu = IndicClassTable::isVattu(charClass) && !hasNukta;
} else if (IndicClassTable::isVirama(charClass) && chars[i + 1] == C_SIGN_ZWNJ)
{
features = tagArray4;