Creates BreakIterator::GetWordBreakStatus.

base/i18n/break_iterator.cc

 // Copyright (c) 2011 The Chromium Authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 #include "base/i18n/break_iterator.h"
 
 #include "base/logging.h"
 #include "third_party/icu/source/common/unicode/ubrk.h"
 #include "third_party/icu/source/common/unicode/uchar.h"
 #include "third_party/icu/source/common/unicode/ustring.h"
@@ skipping 120 matching lines @@
   prev_ = npos;
   if (U_FAILURE(status)) {
     NOTREACHED() << "ubrk_setText failed";
     return false;
   }
   string_ = StringPiece16(text, length);
   return true;
 }
 
 bool BreakIterator::IsWord() const {
-  int32_t status = ubrk_getRuleStatus(static_cast<UBreakIterator*>(iter_));
+  return IsWordBreak() == IS_WORD_BREAK;
+}
+
+BreakIterator::WordBreakStatus BreakIterator::IsWordBreak() const {
   if (break_type_ != BREAK_WORD && break_type_ != RULE_BASED)
-    return false;
-  return status != UBRK_WORD_NONE;
+    return IS_NOT_WORD_BREAK;

[please use gerrit instead, 2015/08/07 20:53:10]

Call ubrk_getRuleStatus() before the if statement.

The original code called urbk_getRuleStatus() before checking break_type_. This
seems unusual, but there could be a good reason for it. Either way, the code
that's using IsWord() implicitly assumes that ubrk_getRuleStatus() is always
called inside of IsWord(). Let's not break that assumption.

[Julius, 2015/08/10 16:06:37]

Done.

+  return ubrk_getRuleStatus(static_cast<UBreakIterator*>(iter_)) ==
+                 UBRK_WORD_NONE
+             ? IS_SKIPPABLE_WORD
+             : IS_WORD_BREAK;
 }
 
 bool BreakIterator::IsEndOfWord(size_t position) const {
   if (break_type_ != BREAK_WORD && break_type_ != RULE_BASED)
     return false;
 
   UBreakIterator* iter = static_cast<UBreakIterator*>(iter_);
   UBool boundary = ubrk_isBoundary(iter, static_cast<int32_t>(position));
   int32_t status = ubrk_getRuleStatus(iter);
   return (!!boundary && status != UBRK_WORD_NONE);
@@ skipping 22 matching lines @@
   return GetStringPiece().as_string();
 }
 
 StringPiece16 BreakIterator::GetStringPiece() const {
   DCHECK(prev_ != npos && pos_ != npos);
   return string_.substr(prev_, pos_ - prev_);
 }
 
 }  // namespace i18n
 }  // namespace base