ICU-10015 Dictionary Break, sync between ICU4C & J. Most changes for this ticket...

X-SVN-Rev: 35115

diff --git a/icu4c/source/common/dictbe.cpp b/icu4c/source/common/dictbe.cpp
index 45104265b6f4df823dd3f5cf5d2c7c89b6caf5dc..20447f45c8add95d87f09b277f6f033cbab51188 100644 (file)
--- a/icu4c/source/common/dictbe.cpp
+++ b/icu4c/source/common/dictbe.cpp
@@ -1,6 +1,6 @@
 /**
  *******************************************************************************
- * Copyright (C) 2006-2013, International Business Machines Corporation
+ * Copyright (C) 2006-2014, International Business Machines Corporation
  * and others. All Rights Reserved.
  *******************************************************************************
  */
@@ -49,6 +49,9 @@ DictionaryBreakEngine::findBreaks( UText *text,
     int32_t result = 0;
 
     // Find the span of characters included in the set.
+    //   The span to break begins at the current position in the text, and
+    //   extends towards the start or end of the text, depending on 'reverse'.
+
     int32_t start = (int32_t)utext_getNativeIndex(text);
     int32_t current;
     int32_t rangeStart;

diff --git a/icu4c/source/common/dictionarydata.cpp b/icu4c/source/common/dictionarydata.cpp
index 039871b5be8a1c145beb22df3c037506a4dd700f..f174f9d450843da4c94047e24171ed5a2a21fe8d 100644 (file)
--- a/icu4c/source/common/dictionarydata.cpp
+++ b/icu4c/source/common/dictionarydata.cpp
@@ -1,6 +1,6 @@
 /*
 *******************************************************************************
-* Copyright (C) 2013, International Business Machines
+* Copyright (C) 2014, International Business Machines
 * Corporation and others.  All Rights Reserved.
 *******************************************************************************
 * dictionarydata.h
@@ -118,7 +118,7 @@ int32_t BytesDictionaryMatcher::matches(UText *text, int32_t maxLength, int32_t
             if (count < limit) {
                 if (values != NULL) {
                     values[count] = bt.getValue();
-            }
+                }
                 lengths[count++] = numChars;
             }
             if (result == USTRINGTRIE_FINAL_VALUE) {

diff --git a/icu4c/source/common/rbbi.cpp b/icu4c/source/common/rbbi.cpp
index 6ab57a7c1175d5769e091c4c70edf8b09b364e28..f091a3ac49eae009353ca644ec6a9f6e5c75c571 100644 (file)
--- a/icu4c/source/common/rbbi.cpp
+++ b/icu4c/source/common/rbbi.cpp
@@ -1,6 +1,6 @@
 /*
 ***************************************************************************
-*   Copyright (C) 1999-2013 International Business Machines Corporation
+*   Copyright (C) 1999-2014 International Business Machines Corporation
 *   and others. All rights reserved.
 ***************************************************************************
 */
@@ -592,6 +592,7 @@ int32_t RuleBasedBreakIterator::next(void) {
     }
 
     int32_t startPos = current();
+    fDictionaryCharCount = 0;
     int32_t result = handleNext(fData->fForwardTable);
     if (fDictionaryCharCount > 0) {
         result = checkDictionary(startPos, result, FALSE);
@@ -646,7 +647,6 @@ int32_t RuleBasedBreakIterator::previous(void) {
     // break position before the current position (we back our internal
     // iterator up one step to prevent handlePrevious() from returning
     // the current position), but not necessarily the last one before
-
     // where we started
 
     int32_t start = current();
@@ -679,11 +679,11 @@ int32_t RuleBasedBreakIterator::previous(void) {
     // the result position that we are to return (in lastResult.)  If
     // the backwards rules overshot and the above loop had to do two or more
     // next()s to move up to the desired return position, we will have a valid
-    // tag value. But, if handlePrevious() took us to exactly the correct result positon,
+    // tag value. But, if handlePrevious() took us to exactly the correct result position,
     // we wont have a tag value for that position, which is only set by handleNext().
 
-    // set the current iteration position to be the last break position
-    // before where we started, and then return that value
+    // Set the current iteration position to be the last break position
+    // before where we started, and then return that value.
     utext_setNativeIndex(fText, lastResult);
     fLastRuleStatusIndex  = lastTag;       // for use by getRuleStatus()
     fLastStatusIndexValid = breakTagValid;
@@ -1703,6 +1703,7 @@ int32_t RuleBasedBreakIterator::checkDictionary(int32_t startPos,
     // If we found breaks, build a new break cache. The first and last entries must
     // be the original starting and ending position.
     if (foundBreakCount > 0) {
+        U_ASSERT(foundBreakCount == breaks.size());
         int32_t totalBreaks = foundBreakCount;
         if (startPos < breaks.elementAti(0)) {
             totalBreaks += 1;