Creates BreakIterator::GetWordBreakStatus.

base/i18n/break_iterator.h

 // 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.
 
 #ifndef BASE_I18N_BREAK_ITERATOR_H_
 #define BASE_I18N_BREAK_ITERATOR_H_
 
 #include "base/basictypes.h"
 #include "base/i18n/base_i18n_export.h"
 #include "base/strings/string16.h"
@@ skipping 53 matching lines @@
     // TODO(jshin): Remove this after reviewing call sites.
     // If call sites really need break only on space-like characters
     // implement it separately.
     BREAK_SPACE = BREAK_LINE,
     BREAK_NEWLINE,
     BREAK_CHARACTER,
     // But don't remove this one!
     RULE_BASED,
   };
 
+  enum WordBreakStatus {
+    // The end of text that the iterator recognizes as word characters.
+    // Non-word characters are things like punctuation and spaces.
+    IS_WORD_BREAK,
+    // Characters that the iterator can skip past, such as punctuation,
+    // whitespace, and, if using RULE_BASED mode, characters from another
+    // character set.
+    IS_SKIPPABLE_WORD,
+    // Only used if not in BREAK_WORD or RULE_BASED mode. This is returned for
+    // newlines, line breaks, and character breaks.
+    IS_LINE_OR_CHAR_BREAK
+  };
+
   // Requires |str| to live as long as the BreakIterator does.
   BreakIterator(const StringPiece16& str, BreakType break_type);
   // Make a rule-based iterator. BreakType == RULE_BASED is implied.
   // TODO(andrewhayden): This signature could easily be misinterpreted as
   // "(const string16& str, const string16& locale)". We should do something
   // better.
   BreakIterator(const StringPiece16& str, const string16& rules);
   ~BreakIterator();
 
   // Init() must be called before any of the iterators are valid.
@@ skipping 10 matching lines @@
   // if Init() had been called again. Any old state is lost. Returns true
   // unless there is an error setting the text.
   bool SetText(const base::char16* text, const size_t length);
 
   // Under BREAK_WORD mode, returns true if the break we just hit is the
   // end of a word. (Otherwise, the break iterator just skipped over e.g.
   // whitespace or punctuation.)  Under BREAK_LINE and BREAK_NEWLINE modes,
   // this distinction doesn't apply and it always returns false.
   bool IsWord() const;
 
+  // Under BREAK_WORD mode:
+  //  - Returns IS_SKIPPABLE_WORD if non-word characters, such as punctuation or
+  //    spaces, are found.
+  //  - Returns IS_WORD_BREAK if the break we just hit is the end of a sequence
+  //    of word characters.
+  // Under RULE_BASED mode:
+  //  - Returns IS_SKIPPABLE_WORD if characters outside the rules' character set
+  //    or non-word characters, such as punctuation or spaces, are found.
+  //  - Returns IS_WORD_BREAK if the break we just hit is the end of a sequence
+  //    of word characters that are in the rules' character set.
+  // Not under BREAK_WORD or RULE_BASED mode:
+  //  - Returns IS_LINE_OR_CHAR_BREAK.
+  BreakIterator::WordBreakStatus GetWordBreakStatus() const;
+
   // Under BREAK_WORD mode, returns true if |position| is at the end of word or
   // at the start of word. It always returns false under BREAK_LINE and
   // BREAK_NEWLINE modes.
   bool IsEndOfWord(size_t position) const;
   bool IsStartOfWord(size_t position) const;
 
   // Under BREAK_CHARACTER mode, returns whether |position| is a Unicode
   // grapheme boundary.
   bool IsGraphemeBoundary(size_t position) const;
 
@@ skipping 30 matching lines @@
   // Previous and current iterator positions.
   size_t prev_, pos_;
 
   DISALLOW_COPY_AND_ASSIGN(BreakIterator);
 };
 
 }  // namespace i18n
 }  // namespace base
 
 #endif  // BASE_I18N_BREAK_ITERATOR_H_