diff options
Diffstat (limited to 'external/icu/icu4c-khmerbreakengine.patch.1')
| -rw-r--r-- | external/icu/icu4c-khmerbreakengine.patch.1 | 269 |
1 files changed, 24 insertions, 245 deletions
diff --git a/external/icu/icu4c-khmerbreakengine.patch.1 b/external/icu/icu4c-khmerbreakengine.patch.1 index 7992da6fc18f..272d0b8ab204 100644 --- a/external/icu/icu4c-khmerbreakengine.patch.1 +++ b/external/icu/icu4c-khmerbreakengine.patch.1 @@ -1,7 +1,7 @@ diff -ur icu.org/source/common/dictbe.cpp icu/source/common/dictbe.cpp ---- icu.org/source/common/dictbe.cpp 2018-10-02 00:39:56.000000000 +0200 -+++ icu/source/common/dictbe.cpp 2018-10-20 00:14:46.462039038 +0200 -@@ -29,7 +29,19 @@ +--- icu.org/source/common/dictbe.cpp 2020-04-22 22:04:20.000000000 +0200 ++++ icu/source/common/dictbe.cpp 2020-05-11 18:55:07.702282061 +0200 +@@ -32,7 +32,19 @@ ****************************************************************** */ @@ -22,7 +22,7 @@ diff -ur icu.org/source/common/dictbe.cpp icu/source/common/dictbe.cpp } DictionaryBreakEngine::~DictionaryBreakEngine() { -@@ -76,6 +88,169 @@ +@@ -79,6 +91,169 @@ fSet.compact(); } @@ -192,7 +192,7 @@ diff -ur icu.org/source/common/dictbe.cpp icu/source/common/dictbe.cpp /* ****************************************************************** * PossibleWord -@@ -282,7 +282,7 @@ +@@ -108,7 +283,7 @@ ~PossibleWord() {} // Fill the list of candidates if needed, select the longest, and return the number found @@ -201,7 +201,7 @@ diff -ur icu.org/source/common/dictbe.cpp icu/source/common/dictbe.cpp // Select the currently marked candidate, point after it in the text, and invalidate self int32_t acceptMarked( UText *text ); -@@ -303,12 +303,12 @@ +@@ -129,12 +304,12 @@ }; @@ -216,7 +216,7 @@ diff -ur icu.org/source/common/dictbe.cpp icu/source/common/dictbe.cpp // Dictionary leaves text after longest prefix, not longest word. Back up. if (count <= 0) { utext_setNativeIndex(text, start); -@@ -803,51 +978,28 @@ +@@ -815,53 +990,30 @@ * KhmerBreakEngine */ @@ -241,6 +241,8 @@ diff -ur icu.org/source/common/dictbe.cpp icu/source/common/dictbe.cpp + : DictionaryBreakEngine((1 << UBRK_WORD) | (1 << UBRK_LINE)), fDictionary(adoptDictionary) { + UTRACE_ENTRY(UTRACE_UBRK_CREATE_BREAK_ENGINE); + UTRACE_DATA1(UTRACE_INFO, "dictbe=%s", "Khmr"); - fKhmerWordSet.applyPattern(UNICODE_STRING_SIMPLE("[[:Khmr:]&[:LineBreak=SA:]]"), status); + + clusterLimit = 3; @@ -277,10 +279,10 @@ diff -ur icu.org/source/common/dictbe.cpp icu/source/common/dictbe.cpp + fIgnoreSet.compact(); + fBaseSet.compact(); + fPuncSet.compact(); + UTRACE_EXIT_STATUS(status); } - KhmerBreakEngine::~KhmerBreakEngine() { -@@ -859,180 +1011,204 @@ +@@ -874,180 +1026,204 @@ int32_t rangeStart, int32_t rangeEnd, UVector32 &foundBreaks ) const { @@ -637,8 +639,8 @@ diff -ur icu.org/source/common/dictbe.cpp icu/source/common/dictbe.cpp #if !UCONFIG_NO_NORMALIZATION diff -ur icu.org/source/common/dictbe.h icu/source/common/dictbe.h ---- icu.org/source/common/dictbe.h 2018-09-29 02:34:41.000000000 +0200 -+++ icu/source/common/dictbe.h 2018-10-19 14:21:00.339942804 +0200 +--- icu.org/source/common/dictbe.h 2020-04-22 22:04:20.000000000 +0200 ++++ icu/source/common/dictbe.h 2020-05-11 19:08:24.754634732 +0200 @@ -34,7 +34,8 @@ * threads without synchronization.</p> */ @@ -735,206 +737,15 @@ diff -ur icu.org/source/common/dictbe.h icu/source/common/dictbe.h * <p>Virtual destructor.</p> */ virtual ~DictionaryBreakEngine(); -@@ -68,7 +139,7 @@ - * <p>Find any breaks within a run in the supplied text.</p> - * - * @param text A UText representing the text. The iterator is left at -- * the end of the run of characters which the engine is capable of handling -+ * the end of the run of characters which the engine is capable of handling - * that starts from the first character in the range. - * @param startPos The start of the run within the supplied text. - * @param endPos The end of the run within the supplied text. -@@ -218,118 +289,120 @@ - - }; - --/******************************************************************* -- * BurmeseBreakEngine -- */ -- --/** -- * <p>BurmeseBreakEngine is a kind of DictionaryBreakEngine that uses a -- * DictionaryMatcher and heuristics to determine Burmese-specific breaks.</p> -- * -- * <p>After it is constructed a BurmeseBreakEngine may be shared between -- * threads without synchronization.</p> -- */ --class BurmeseBreakEngine : public DictionaryBreakEngine { -- private: -- /** -- * The set of characters handled by this engine -- * @internal -- */ -- -- UnicodeSet fBurmeseWordSet; +@@ -293,11 +364,13 @@ + */ + + 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. -- */ -- BurmeseBreakEngine(DictionaryMatcher *adoptDictionary, UErrorCode &status); -- -- /** -- * <p>Virtual destructor.</p> -- */ -- virtual ~BurmeseBreakEngine(); -- -- 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, -- UVector32 &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, -- UVector32 &foundBreaks ) const; -- --}; -- -+/******************************************************************* -+ * BurmeseBreakEngine -+ */ -+ -+/** -+ * <p>BurmeseBreakEngine is a kind of DictionaryBreakEngine that uses a -+ * DictionaryMatcher and heuristics to determine Burmese-specific breaks.</p> -+ * -+ * <p>After it is constructed a BurmeseBreakEngine may be shared between -+ * threads without synchronization.</p> -+ */ -+class BurmeseBreakEngine : public DictionaryBreakEngine { -+ private: -+ /** -+ * The set of characters handled by this engine -+ * @internal -+ */ -+ -+ UnicodeSet fBurmeseWordSet; -+ 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. -+ */ -+ BurmeseBreakEngine(DictionaryMatcher *adoptDictionary, UErrorCode &status); -+ -+ /** -+ * <p>Virtual destructor.</p> -+ */ -+ virtual ~BurmeseBreakEngine(); -+ -+ 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, -+ UVector32 &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 fBeginWordSet; + UnicodeSet fPuncSet; + DictionaryMatcher *fDictionary; @@ -942,44 +753,12 @@ diff -ur icu.org/source/common/dictbe.h icu/source/common/dictbe.h + const uint32_t BADSNLP = 256 * 20; + const uint32_t kuint32max = 0x7FFFFFFF; + -+ 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, -+ UVector32 &foundBreaks ) const; -+ -+}; -+ - #if !UCONFIG_NO_NORMALIZATION - - /******************************************************************* + public: + + /** diff -ur icu.org/source/common/dictionarydata.cpp icu/source/common/dictionarydata.cpp ---- icu.org/source/common/dictionarydata.cpp 2018-09-29 02:34:41.000000000 +0200 -+++ icu/source/common/dictionarydata.cpp 2018-10-19 14:21:00.339942804 +0200 +--- icu.org/source/common/dictionarydata.cpp 2020-04-22 22:04:20.000000000 +0200 ++++ icu/source/common/dictionarydata.cpp 2020-05-11 18:50:43.703113749 +0200 @@ -44,7 +44,7 @@ int32_t UCharsDictionaryMatcher::matches(UText *text, int32_t maxLength, int32_t limit, @@ -1027,8 +806,8 @@ diff -ur icu.org/source/common/dictionarydata.cpp icu/source/common/dictionaryda if (values != NULL) { values[wordCount] = bt.getValue(); diff -ur icu.org/source/common/dictionarydata.h icu/source/common/dictionarydata.h ---- icu.org/source/common/dictionarydata.h 2018-09-29 02:34:41.000000000 +0200 -+++ icu/source/common/dictionarydata.h 2018-10-19 14:21:00.339942804 +0200 +--- icu.org/source/common/dictionarydata.h 2020-04-22 22:04:20.000000000 +0200 ++++ icu/source/common/dictionarydata.h 2020-05-11 18:50:43.704113746 +0200 @@ -21,6 +21,7 @@ #include "unicode/utext.h" #include "unicode/udata.h" |