ICU-7647 Add/use LaoBreakEngine and laodict.txt; more useful messages in gendict
X-SVN-Rev: 34229
This commit is contained in:
Peter Edberg
parent
4d9fad13ef
commit
bf4126616b
@ -1,6 +1,6 @@
|
||||
/*
|
||||
************************************************************************************
|
||||
* Copyright (C) 2006-2012, International Business Machines Corporation
|
||||
* Copyright (C) 2006-2013, International Business Machines Corporation
|
||||
* and others. All Rights Reserved.
|
||||
************************************************************************************
|
||||
*/
|
||||
@ -229,6 +229,9 @@ ICULanguageBreakFactory::loadEngineFor(UChar32 c, int32_t breakType) {
|
||||
case USCRIPT_THAI:
|
||||
engine = new ThaiBreakEngine(m, status);
|
||||
break;
|
||||
case USCRIPT_LAO:
|
||||
engine = new LaoBreakEngine(m, status);
|
||||
break;
|
||||
case USCRIPT_KHMER:
|
||||
engine = new KhmerBreakEngine(m, status);
|
||||
break;
|
||||
|
||||
@ -88,10 +88,10 @@ DictionaryBreakEngine::setCharacters( const UnicodeSet &set ) {
|
||||
|
||||
/*
|
||||
******************************************************************
|
||||
* PossibleWord
|
||||
*/
|
||||
|
||||
|
||||
// Helper class for improving readability of the Thai word break
|
||||
// Helper class for improving readability of the Thai/Lao/Khmer word break
|
||||
// algorithm. The implementation is completely inline.
|
||||
|
||||
// List size, limited by the maximum number of words in the dictionary
|
||||
@ -183,6 +183,11 @@ PossibleWord::markCurrent() {
|
||||
mark = current;
|
||||
}
|
||||
|
||||
/*
|
||||
******************************************************************
|
||||
* ThaiBreakEngine
|
||||
*/
|
||||
|
||||
// How many words in a row are "good enough"?
|
||||
#define THAI_LOOKAHEAD 3
|
||||
|
||||
@ -415,6 +420,203 @@ foundBest:
|
||||
return wordsFound;
|
||||
}
|
||||
|
||||
/*
|
||||
******************************************************************
|
||||
* LaoBreakEngine
|
||||
*/
|
||||
|
||||
// How many words in a row are "good enough"?
|
||||
#define LAO_LOOKAHEAD 3
|
||||
|
||||
// Will not combine a non-word with a preceding dictionary word longer than this
|
||||
#define LAO_ROOT_COMBINE_THRESHOLD 3
|
||||
|
||||
// Will not combine a non-word that shares at least this much prefix with a
|
||||
// dictionary word, with a preceding word
|
||||
#define LAO_PREFIX_COMBINE_THRESHOLD 3
|
||||
|
||||
// Minimum word size
|
||||
#define LAO_MIN_WORD 2
|
||||
|
||||
// Minimum number of characters for two words
|
||||
#define LAO_MIN_WORD_SPAN (LAO_MIN_WORD * 2)
|
||||
|
||||
LaoBreakEngine::LaoBreakEngine(DictionaryMatcher *adoptDictionary, UErrorCode &status)
|
||||
: DictionaryBreakEngine((1<<UBRK_WORD) | (1<<UBRK_LINE)),
|
||||
fDictionary(adoptDictionary)
|
||||
{
|
||||
fLaoWordSet.applyPattern(UNICODE_STRING_SIMPLE("[[:Laoo:]&[:LineBreak=SA:]]"), status);
|
||||
if (U_SUCCESS(status)) {
|
||||
setCharacters(fLaoWordSet);
|
||||
}
|
||||
fMarkSet.applyPattern(UNICODE_STRING_SIMPLE("[[:Laoo:]&[:LineBreak=SA:]&[:M:]]"), status);
|
||||
fMarkSet.add(0x0020);
|
||||
fEndWordSet = fLaoWordSet;
|
||||
fEndWordSet.remove(0x0EC0, 0x0EC4); // prefix vowels
|
||||
fBeginWordSet.add(0x0E81, 0x0EAE); // basic consonants (including holes for corresponding Thai characters)
|
||||
fBeginWordSet.add(0x0EDC, 0x0EDD); // digraph consonants (no Thai equivalent)
|
||||
fBeginWordSet.add(0x0EC0, 0x0EC4); // prefix vowels
|
||||
|
||||
// Compact for caching.
|
||||
fMarkSet.compact();
|
||||
fEndWordSet.compact();
|
||||
fBeginWordSet.compact();
|
||||
}
|
||||
|
||||
LaoBreakEngine::~LaoBreakEngine() {
|
||||
delete fDictionary;
|
||||
}
|
||||
|
||||
int32_t
|
||||
LaoBreakEngine::divideUpDictionaryRange( UText *text,
|
||||
int32_t rangeStart,
|
||||
int32_t rangeEnd,
|
||||
UStack &foundBreaks ) const {
|
||||
if ((rangeEnd - rangeStart) < LAO_MIN_WORD_SPAN) {
|
||||
return 0; // Not enough characters for two words
|
||||
}
|
||||
|
||||
uint32_t wordsFound = 0;
|
||||
int32_t wordLength;
|
||||
int32_t current;
|
||||
UErrorCode status = U_ZERO_ERROR;
|
||||
PossibleWord words[LAO_LOOKAHEAD];
|
||||
UChar32 uc;
|
||||
|
||||
utext_setNativeIndex(text, rangeStart);
|
||||
|
||||
while (U_SUCCESS(status) && (current = (int32_t)utext_getNativeIndex(text)) < rangeEnd) {
|
||||
wordLength = 0;
|
||||
|
||||
// Look for candidate words at the current position
|
||||
int candidates = words[wordsFound%LAO_LOOKAHEAD].candidates(text, fDictionary, rangeEnd);
|
||||
|
||||
// If we found exactly one, use that
|
||||
if (candidates == 1) {
|
||||
wordLength = words[wordsFound % LAO_LOOKAHEAD].acceptMarked(text);
|
||||
wordsFound += 1;
|
||||
}
|
||||
// If there was more than one, see which one can take us forward the most words
|
||||
else if (candidates > 1) {
|
||||
// If we're already at the end of the range, we're done
|
||||
if ((int32_t)utext_getNativeIndex(text) >= rangeEnd) {
|
||||
goto foundBest;
|
||||
}
|
||||
do {
|
||||
int wordsMatched = 1;
|
||||
if (words[(wordsFound + 1) % LAO_LOOKAHEAD].candidates(text, fDictionary, rangeEnd) > 0) {
|
||||
if (wordsMatched < 2) {
|
||||
// Followed by another dictionary word; mark first word as a good candidate
|
||||
words[wordsFound%LAO_LOOKAHEAD].markCurrent();
|
||||
wordsMatched = 2;
|
||||
}
|
||||
|
||||
// If we're already at the end of the range, we're done
|
||||
if ((int32_t)utext_getNativeIndex(text) >= rangeEnd) {
|
||||
goto foundBest;
|
||||
}
|
||||
|
||||
// See if any of the possible second words is followed by a third word
|
||||
do {
|
||||
// If we find a third word, stop right away
|
||||
if (words[(wordsFound + 2) % LAO_LOOKAHEAD].candidates(text, fDictionary, rangeEnd)) {
|
||||
words[wordsFound % LAO_LOOKAHEAD].markCurrent();
|
||||
goto foundBest;
|
||||
}
|
||||
}
|
||||
while (words[(wordsFound + 1) % LAO_LOOKAHEAD].backUp(text));
|
||||
}
|
||||
}
|
||||
while (words[wordsFound % LAO_LOOKAHEAD].backUp(text));
|
||||
foundBest:
|
||||
wordLength = words[wordsFound % LAO_LOOKAHEAD].acceptMarked(text);
|
||||
wordsFound += 1;
|
||||
}
|
||||
|
||||
// We come here after having either found a word or not. We look ahead to the
|
||||
// next word. If it's not a dictionary word, we will combine it withe the word we
|
||||
// just found (if there is one), but only if the preceding word does not exceed
|
||||
// the threshold.
|
||||
// The text iterator should now be positioned at the end of the word we found.
|
||||
if ((int32_t)utext_getNativeIndex(text) < rangeEnd && wordLength < LAO_ROOT_COMBINE_THRESHOLD) {
|
||||
// if it is a dictionary word, do nothing. If it isn't, then if there is
|
||||
// no preceding word, or the non-word shares less than the minimum threshold
|
||||
// of characters with a dictionary word, then scan to resynchronize
|
||||
if (words[wordsFound % LAO_LOOKAHEAD].candidates(text, fDictionary, rangeEnd) <= 0
|
||||
&& (wordLength == 0
|
||||
|| words[wordsFound%LAO_LOOKAHEAD].longestPrefix() < LAO_PREFIX_COMBINE_THRESHOLD)) {
|
||||
// Look for a plausible word boundary
|
||||
//TODO: This section will need a rework for UText.
|
||||
int32_t remaining = rangeEnd - (current+wordLength);
|
||||
UChar32 pc = utext_current32(text);
|
||||
int32_t chars = 0;
|
||||
for (;;) {
|
||||
utext_next32(text);
|
||||
uc = utext_current32(text);
|
||||
// TODO: Here we're counting on the fact that the SA languages are all
|
||||
// in the BMP. This should get fixed with the UText rework.
|
||||
chars += 1;
|
||||
if (--remaining <= 0) {
|
||||
break;
|
||||
}
|
||||
if (fEndWordSet.contains(pc) && fBeginWordSet.contains(uc)) {
|
||||
// Maybe. See if it's in the dictionary.
|
||||
int candidates = words[(wordsFound + 1) % LAO_LOOKAHEAD].candidates(text, fDictionary, rangeEnd);
|
||||
utext_setNativeIndex(text, current + wordLength + chars);
|
||||
if (candidates > 0) {
|
||||
break;
|
||||
}
|
||||
}
|
||||
pc = uc;
|
||||
}
|
||||
|
||||
// Bump the word count if there wasn't already one
|
||||
if (wordLength <= 0) {
|
||||
wordsFound += 1;
|
||||
}
|
||||
|
||||
// Update the length with the passed-over characters
|
||||
wordLength += chars;
|
||||
}
|
||||
else {
|
||||
// Back up to where we were for next iteration
|
||||
utext_setNativeIndex(text, current+wordLength);
|
||||
}
|
||||
}
|
||||
|
||||
// Never stop before a combining mark.
|
||||
int32_t currPos;
|
||||
while ((currPos = (int32_t)utext_getNativeIndex(text)) < rangeEnd && fMarkSet.contains(utext_current32(text))) {
|
||||
utext_next32(text);
|
||||
wordLength += (int32_t)utext_getNativeIndex(text) - currPos;
|
||||
}
|
||||
|
||||
// Look ahead for possible suffixes if a dictionary word does not follow.
|
||||
// We do this in code rather than using a rule so that the heuristic
|
||||
// resynch continues to function. For example, one of the suffix characters
|
||||
// could be a typo in the middle of a word.
|
||||
// NOT CURRENTLY APPLICABLE TO LAO
|
||||
|
||||
// Did we find a word on this iteration? If so, push it on the break stack
|
||||
if (wordLength > 0) {
|
||||
foundBreaks.push((current+wordLength), status);
|
||||
}
|
||||
}
|
||||
|
||||
// Don't return a break for the end of the dictionary range if there is one there.
|
||||
if (foundBreaks.peeki() >= rangeEnd) {
|
||||
(void) foundBreaks.popi();
|
||||
wordsFound -= 1;
|
||||
}
|
||||
|
||||
return wordsFound;
|
||||
}
|
||||
|
||||
/*
|
||||
******************************************************************
|
||||
* KhmerBreakEngine
|
||||
*/
|
||||
|
||||
// How many words in a row are "good enough"?
|
||||
#define KHMER_LOOKAHEAD 3
|
||||
|
||||
|
||||
@ -1,6 +1,6 @@
|
||||
/**
|
||||
*******************************************************************************
|
||||
* Copyright (C) 2006,2012, International Business Machines Corporation *
|
||||
* Copyright (C) 2006,2012-2013, International Business Machines Corporation *
|
||||
* and others. All Rights Reserved. *
|
||||
*******************************************************************************
|
||||
*/
|
||||
@ -186,6 +186,118 @@ class ThaiBreakEngine : public DictionaryBreakEngine {
|
||||
|
||||
};
|
||||
|
||||
/*******************************************************************
|
||||
* LaoBreakEngine
|
||||
*/
|
||||
|
||||
/**
|
||||
* <p>LaoBreakEngine is a kind of DictionaryBreakEngine that uses a
|
||||
* dictionary and heuristics to determine Lao-specific breaks.</p>
|
||||
*
|
||||
* <p>After it is constructed a LaoBreakEngine may be shared between
|
||||
* threads without synchronization.</p>
|
||||
*/
|
||||
class LaoBreakEngine : public DictionaryBreakEngine {
|
||||
private:
|
||||
/**
|
||||
* The set of characters handled by this engine
|
||||
* @internal
|
||||
*/
|
||||
|
||||
UnicodeSet fLaoWordSet;
|
||||
UnicodeSet fEndWordSet;
|
||||
UnicodeSet fBeginWordSet;
|
||||
UnicodeSet fMarkSet;
|
||||
DictionaryMatcher *fDictionary;
|
||||
|
||||
public:
|
||||
|
||||
/**
|
||||
* <p>Default constructor.</p>
|
||||
*
|
||||
* @param adoptDictionary A DictionaryMatcher to adopt. Deleted when the
|
||||
* engine is deleted.
|
||||
*/
|
||||
LaoBreakEngine(DictionaryMatcher *adoptDictionary, UErrorCode &status);
|
||||
|
||||
/**
|
||||
* <p>Virtual destructor.</p>
|
||||
*/
|
||||
virtual ~LaoBreakEngine();
|
||||
|
||||
protected:
|
||||
/**
|
||||
* <p>Divide up a range of known dictionary characters handled by this break engine.</p>
|
||||
*
|
||||
* @param text A UText representing the text
|
||||
* @param rangeStart The start of the range of dictionary characters
|
||||
* @param rangeEnd The end of the range of dictionary characters
|
||||
* @param foundBreaks Output of C array of int32_t break positions, or 0
|
||||
* @return The number of breaks found
|
||||
*/
|
||||
virtual int32_t divideUpDictionaryRange( UText *text,
|
||||
int32_t rangeStart,
|
||||
int32_t rangeEnd,
|
||||
UStack &foundBreaks ) const;
|
||||
|
||||
};
|
||||
|
||||
/*******************************************************************
|
||||
* KhmerBreakEngine
|
||||
*/
|
||||
|
||||
/**
|
||||
* <p>KhmerBreakEngine is a kind of DictionaryBreakEngine that uses a
|
||||
* DictionaryMatcher and heuristics to determine Khmer-specific breaks.</p>
|
||||
*
|
||||
* <p>After it is constructed a KhmerBreakEngine may be shared between
|
||||
* threads without synchronization.</p>
|
||||
*/
|
||||
class KhmerBreakEngine : public DictionaryBreakEngine {
|
||||
private:
|
||||
/**
|
||||
* The set of characters handled by this engine
|
||||
* @internal
|
||||
*/
|
||||
|
||||
UnicodeSet fKhmerWordSet;
|
||||
UnicodeSet fEndWordSet;
|
||||
UnicodeSet fBeginWordSet;
|
||||
UnicodeSet fMarkSet;
|
||||
DictionaryMatcher *fDictionary;
|
||||
|
||||
public:
|
||||
|
||||
/**
|
||||
* <p>Default constructor.</p>
|
||||
*
|
||||
* @param adoptDictionary A DictionaryMatcher to adopt. Deleted when the
|
||||
* engine is deleted.
|
||||
*/
|
||||
KhmerBreakEngine(DictionaryMatcher *adoptDictionary, UErrorCode &status);
|
||||
|
||||
/**
|
||||
* <p>Virtual destructor.</p>
|
||||
*/
|
||||
virtual ~KhmerBreakEngine();
|
||||
|
||||
protected:
|
||||
/**
|
||||
* <p>Divide up a range of known dictionary characters.</p>
|
||||
*
|
||||
* @param text A UText representing the text
|
||||
* @param rangeStart The start of the range of dictionary characters
|
||||
* @param rangeEnd The end of the range of dictionary characters
|
||||
* @param foundBreaks Output of C array of int32_t break positions, or 0
|
||||
* @return The number of breaks found
|
||||
*/
|
||||
virtual int32_t divideUpDictionaryRange( UText *text,
|
||||
int32_t rangeStart,
|
||||
int32_t rangeEnd,
|
||||
UStack &foundBreaks ) const;
|
||||
|
||||
};
|
||||
|
||||
#if !UCONFIG_NO_NORMALIZATION
|
||||
|
||||
/*******************************************************************
|
||||
@ -251,63 +363,6 @@ class CjkBreakEngine : public DictionaryBreakEngine {
|
||||
|
||||
#endif
|
||||
|
||||
/*******************************************************************
|
||||
* KhmerBreakEngine
|
||||
*/
|
||||
|
||||
/**
|
||||
* <p>KhmerBreakEngine is a kind of DictionaryBreakEngine that uses a
|
||||
* DictionaryMatcher and heuristics to determine Khmer-specific breaks.</p>
|
||||
*
|
||||
* <p>After it is constructed a KhmerBreakEngine may be shared between
|
||||
* threads without synchronization.</p>
|
||||
*/
|
||||
class KhmerBreakEngine : public DictionaryBreakEngine {
|
||||
private:
|
||||
/**
|
||||
* The set of characters handled by this engine
|
||||
* @internal
|
||||
*/
|
||||
|
||||
UnicodeSet fKhmerWordSet;
|
||||
UnicodeSet fEndWordSet;
|
||||
UnicodeSet fBeginWordSet;
|
||||
UnicodeSet fMarkSet;
|
||||
DictionaryMatcher *fDictionary;
|
||||
|
||||
public:
|
||||
|
||||
/**
|
||||
* <p>Default constructor.</p>
|
||||
*
|
||||
* @param adoptDictionary A DictionaryMatcher to adopt. Deleted when the
|
||||
* engine is deleted.
|
||||
*/
|
||||
KhmerBreakEngine(DictionaryMatcher *adoptDictionary, UErrorCode &status);
|
||||
|
||||
/**
|
||||
* <p>Virtual destructor.</p>
|
||||
*/
|
||||
virtual ~KhmerBreakEngine();
|
||||
|
||||
protected:
|
||||
/**
|
||||
* <p>Divide up a range of known dictionary characters.</p>
|
||||
*
|
||||
* @param text A UText representing the text
|
||||
* @param rangeStart The start of the range of dictionary characters
|
||||
* @param rangeEnd The end of the range of dictionary characters
|
||||
* @param foundBreaks Output of C array of int32_t break positions, or 0
|
||||
* @return The number of breaks found
|
||||
*/
|
||||
virtual int32_t divideUpDictionaryRange( UText *text,
|
||||
int32_t rangeStart,
|
||||
int32_t rangeEnd,
|
||||
UStack &foundBreaks ) const;
|
||||
|
||||
};
|
||||
|
||||
|
||||
U_NAMESPACE_END
|
||||
|
||||
/* DICTBE_H */
|
||||
|
||||
@ -1,5 +1,5 @@
|
||||
## Makefile.in for ICU data
|
||||
## Copyright (c) 1999-2012, International Business Machines Corporation and
|
||||
## Copyright (c) 1999-2013, International Business Machines Corporation and
|
||||
## others. All Rights Reserved.
|
||||
|
||||
## Source directory information
|
||||
@ -527,6 +527,9 @@ $(BRKBLDDIR)/%.dict: $(TOOLBINDIR)/gendict$(TOOLEXEEXT) $(DAT_FILES)
|
||||
$(BRKBLDDIR)/thaidict.dict: $(TOOLBINDIR)/gendict$(TOOLEXEEXT) $(DAT_FILES)
|
||||
$(INVOKE) $(TOOLBINDIR)/gendict --bytes --transform offset-0x0e00 -c -i $(BUILDDIR) $(BRKSRCDIR)/thaidict.txt $(BRKBLDDIR)/thaidict.dict
|
||||
|
||||
$(BRKBLDDIR)/laodict.dict: $(TOOLBINDIR)/gendict$(TOOLEXEEXT) $(DAT_FILES)
|
||||
$(INVOKE) $(TOOLBINDIR)/gendict --bytes --transform offset-0x0e80 -c -i $(BUILDDIR) $(BRKSRCDIR)/laodict.txt $(BRKBLDDIR)/laodict.dict
|
||||
|
||||
# TODO: figure out why combining characters are here?
|
||||
$(BRKBLDDIR)/khmerdict.dict: $(TOOLBINDIR)/gendict$(TOOLEXEEXT) $(DAT_FILES)
|
||||
$(INVOKE) $(TOOLBINDIR)/gendict --bytes --transform offset-0x1780 -c -i $(BUILDDIR) $(BRKSRCDIR)/khmerdict.txt $(BRKBLDDIR)/khmerdict.dict
|
||||
|
||||
@ -34,7 +34,7 @@ BRK_RES_ALIAS_SOURCE = $(BRK_RES_SYNTHETIC_ALIAS)
|
||||
|
||||
|
||||
# List of dictionary files (dict).
|
||||
BRK_DICT_SOURCE = cjdict.txt khmerdict.txt thaidict.txt
|
||||
BRK_DICT_SOURCE = cjdict.txt khmerdict.txt laodict.txt thaidict.txt
|
||||
|
||||
|
||||
# List of break iterator files (brk).
|
||||
|
||||
11250
icu4c/source/data/brkitr/laodict.txt
Normal file
11250
icu4c/source/data/brkitr/laodict.txt
Normal file
File diff suppressed because it is too large
Load Diff
@ -20,6 +20,7 @@ root{
|
||||
Hira:process(dependency){"cjdict.dict"}
|
||||
Kata:process(dependency){"cjdict.dict"}
|
||||
Khmr:process(dependency){"khmerdict.dict"}
|
||||
Laoo:process(dependency){"laodict.dict"}
|
||||
Thai:process(dependency){"thaidict.dict"}
|
||||
}
|
||||
}
|
||||
|
||||
@ -743,6 +743,10 @@ $(ICUBRK)\thaidict.dict:
|
||||
@echo Creating $(ICUBRK)\thaidict.dict
|
||||
@"$(ICUTOOLS)\gendict\$(CFG)\gendict" -c --bytes --transform offset-0xe00 $(ICUSRCDATA_RELATIVE_PATH)\$(ICUBRK)\thaidict.txt "$(ICUBLD_PKG)\$(ICUBRK)\thaidict.dict"
|
||||
|
||||
$(ICUBRK)\laodict.dict:
|
||||
@echo Creating $(ICUBRK)\laodict.dict
|
||||
@"$(ICUTOOLS)\gendict\$(CFG)\gendict" -c --bytes --transform offset-0xe00 $(ICUSRCDATA_RELATIVE_PATH)\$(ICUBRK)\laodict.txt "$(ICUBLD_PKG)\$(ICUBRK)\laodict.dict"
|
||||
|
||||
$(ICUBRK)\khmerdict.dict:
|
||||
@echo Creating $(ICUBRK)\khmerdict.dict
|
||||
@"$(ICUTOOLS)\gendict\$(CFG)\gendict" -c --bytes --transform offset-0x1780 $(ICUSRCDATA_RELATIVE_PATH)\$(ICUBRK)\khmerdict.txt "$(ICUBLD_PKG)\$(ICUBRK)\khmerdict.dict"
|
||||
|
||||
@ -1,6 +1,6 @@
|
||||
<?xml version="1.0" encoding="UTF-8" ?>
|
||||
<!--
|
||||
Copyright (c) 2010-2012 International Business Machines Corporation and others. All rights reserved.
|
||||
Copyright (c) 2010-2013 International Business Machines Corporation and others. All rights reserved.
|
||||
-->
|
||||
<!DOCTYPE ldml SYSTEM "http://www.unicode.org/repos/cldr/trunk/common/dtd/ldml.dtd"
|
||||
[
|
||||
@ -25,6 +25,7 @@
|
||||
</icu:boundaries>
|
||||
<icu:dictionaries>
|
||||
<icu:dictionary type="Thai" icu:dependency="thaidict.dict"/>
|
||||
<icu:dictionary type="Laoo" icu:dependency="laodict.dict"/>
|
||||
<icu:dictionary type="Khmr" icu:dependency="khmerdict.dict"/>
|
||||
<icu:dictionary type="Hani" icu:dependency="cjdict.dict"/>
|
||||
<icu:dictionary type="Hira" icu:dependency="cjdict.dict"/>
|
||||
|
||||
13
icu4c/source/test/testdata/rbbitst.txt
vendored
13
icu4c/source/test/testdata/rbbitst.txt
vendored
@ -691,6 +691,19 @@ Bangkok)•</data>
|
||||
<data>•ใช•มั้ย•</data>
|
||||
<data>•มั๊ยล่ะ•ที่รัก•</data>
|
||||
|
||||
##########################################################################################
|
||||
#
|
||||
# Lao Tests
|
||||
#
|
||||
##########################################################################################
|
||||
<locale en>
|
||||
# Basic check for #7647
|
||||
<line>
|
||||
<data>•ສະບາຍດີ•</data>
|
||||
<data>•ດີ•ຂອບໃຈ•</data>
|
||||
<data>•ເຈົ້າ•ເວົ້າ•ພາສາ•ອັງກິດ•ໄດ້•ບໍ່•</data>
|
||||
<data>•ກະລຸນາ•ເວົ້າ•ຊ້າ•ໆ•</data>
|
||||
|
||||
##########################################################################################
|
||||
#
|
||||
# Khmer Tests
|
||||
|
||||
@ -365,6 +365,9 @@ int main(int argc, char **argv) {
|
||||
UBool hasValues = FALSE;
|
||||
UBool hasValuelessContents = FALSE;
|
||||
int lineCount = 0;
|
||||
int wordCount = 0;
|
||||
int minlen = 255;
|
||||
int maxlen = 0;
|
||||
UBool isOk = TRUE;
|
||||
while (readLine(f, fileLine, status)) {
|
||||
lineCount++;
|
||||
@ -401,9 +404,15 @@ int main(int argc, char **argv) {
|
||||
}
|
||||
dict.addWord(fileLine.tempSubString(0, keyLen), (int32_t)value, status);
|
||||
hasValues = TRUE;
|
||||
wordCount++;
|
||||
if (keyLen < minlen) minlen = keyLen;
|
||||
if (keyLen > maxlen) maxlen = keyLen;
|
||||
} else {
|
||||
dict.addWord(fileLine.tempSubString(0, keyLen), 0, status);
|
||||
hasValuelessContents = FALSE;
|
||||
hasValuelessContents = TRUE;
|
||||
wordCount++;
|
||||
if (keyLen < minlen) minlen = keyLen;
|
||||
if (keyLen > maxlen) maxlen = keyLen;
|
||||
}
|
||||
|
||||
if (status.isFailure()) {
|
||||
@ -412,6 +421,7 @@ int main(int argc, char **argv) {
|
||||
exit(status.reset());
|
||||
}
|
||||
}
|
||||
if (verbose) { printf("Processed %d lines, added %d words, minlen %d, maxlen %d\n", lineCount, wordCount, minlen, maxlen); }
|
||||
|
||||
if (!isOk && status.isSuccess()) {
|
||||
status.set(U_ILLEGAL_ARGUMENT_ERROR);
|
||||
@ -420,7 +430,7 @@ int main(int argc, char **argv) {
|
||||
fprintf(stderr, "warning: file contained both valued and unvalued strings!\n");
|
||||
}
|
||||
|
||||
if (verbose) { puts("Serializing data..."); }
|
||||
if (verbose) { printf("Serializing data...isBytesTrie? %d\n", isBytesTrie); }
|
||||
int32_t outDataSize;
|
||||
const void *outData;
|
||||
UnicodeString usp;
|
||||
@ -434,7 +444,7 @@ int main(int argc, char **argv) {
|
||||
outData = usp.getBuffer();
|
||||
}
|
||||
if (status.isFailure()) {
|
||||
fprintf(stderr, "gendict: got failure of type %s while serializing\n", status.errorName());
|
||||
fprintf(stderr, "gendict: got failure of type %s while serializing, if U_ILLEGAL_ARGUMENT_ERROR possibly due to duplicate dictionary entries\n", status.errorName());
|
||||
exit(status.reset());
|
||||
}
|
||||
if (verbose) { puts("Opening output file..."); }
|
||||
|
||||
Loading…
Reference in New Issue
Block a user