OLD | NEW |
---|---|
1 // Copyright (c) 2010 The Chromium Authors. All rights reserved. | 1 // Copyright (c) 2010 The Chromium Authors. All rights reserved. |
2 // Use of this source code is governed by a BSD-style license that can be | 2 // Use of this source code is governed by a BSD-style license that can be |
3 // found in the LICENSE file. | 3 // found in the LICENSE file. |
4 | 4 |
5 #ifndef BASE_I18N_WORD_ITERATOR_H_ | 5 #ifndef BASE_I18N_BREAK_ITERATOR_H_ |
6 #define BASE_I18N_WORD_ITERATOR_H_ | 6 #define BASE_I18N_BREAK_ITERATOR_H_ |
7 #pragma once | 7 #pragma once |
8 | 8 |
9 #include <vector> | |
10 | |
11 #include "base/basictypes.h" | 9 #include "base/basictypes.h" |
12 #include "base/string16.h" | 10 #include "base/string16.h" |
13 | 11 |
14 // The WordIterator class iterates through the words and word breaks | 12 // The BreakIterator class iterates through the words and word breaks |
15 // in a UTF-16 string. | 13 // in a UTF-16 string. |
16 // | 14 // |
17 // It provides two modes, BREAK_WORD and BREAK_LINE, which modify how | 15 // It provides two modes, BREAK_WORD and BREAK_SPACE, which modify how |
18 // trailing non-word characters are aggregated into the returned word. | 16 // trailing non-word characters are aggregated into the returned word. |
19 // | 17 // |
20 // Under BREAK_WORD mode (more common), the non-word characters are | 18 // Under BREAK_WORD mode (more common), the non-word characters are |
21 // not included with a returned word (e.g. in the UTF-16 equivalent of | 19 // not included with a returned word (e.g. in the UTF-16 equivalent of |
22 // the string " foo bar! ", the word breaks are at the periods in | 20 // the string " foo bar! ", the word breaks are at the periods in |
23 // ". .foo. .bar.!. ."). | 21 // ". .foo. .bar.!. ."). |
24 // | 22 // |
25 // Under BREAK_LINE mode (less common), the non-word characters are | 23 // Under BREAK_SPACE mode (less common), the non-word characters are |
26 // included in the word, breaking only when a space-equivalent character | 24 // included in the word, breaking only when a space-equivalent character |
27 // is encountered (e.g. in the UTF16-equivalent of the string " foo bar! ", | 25 // is encountered (e.g. in the UTF16-equivalent of the string " foo bar! ", |
28 // the word breaks are at the periods in ". .foo .bar! ."). | 26 // the word breaks are at the periods in ". .foo .bar! ."). |
29 // | 27 // |
30 // To extract the words from a string, move a BREAK_WORD WordIterator | 28 // To extract the words from a string, move a BREAK_WORD BreakIterator |
31 // through the string and test whether IsWord() is true. E.g., | 29 // through the string and test whether IsWord() is true. E.g., |
32 // WordIterator iter(&str, WordIterator::BREAK_WORD); | 30 // BreakIterator iter(&str, BreakIterator::BREAK_WORD); |
33 // if (!iter.Init()) return false; | 31 // if (!iter.Init()) return false; |
34 // while (iter.Advance()) { | 32 // while (iter.Advance()) { |
35 // if (iter.IsWord()) { | 33 // if (iter.IsWord()) { |
36 // // region [iter.prev(),iter.pos()) contains a word. | 34 // // region [iter.prev(),iter.pos()) contains a word. |
37 // VLOG(1) << "word: " << iter.GetWord(); | 35 // VLOG(1) << "word: " << iter.GetWord(); |
38 // } | 36 // } |
39 // } | 37 // } |
40 | 38 |
39 namespace base { // Prevent conflict with ICU BreakIterator | |
Evan Martin
2010/12/13 20:26:04
I guess this comment isn't necessary -- it's impli
| |
41 | 40 |
42 class WordIterator { | 41 class BreakIterator { |
43 public: | 42 public: |
44 enum BreakType { | 43 enum BreakType { |
45 BREAK_WORD, | 44 BREAK_WORD, |
46 BREAK_LINE | 45 BREAK_SPACE |
47 }; | 46 }; |
48 | 47 |
49 // Requires |str| to live as long as the WordIterator does. | 48 // Requires |str| to live as long as the BreakIterator does. |
50 WordIterator(const string16* str, BreakType break_type); | 49 BreakIterator(const string16* str, BreakType break_type); |
51 ~WordIterator(); | 50 ~BreakIterator(); |
52 | 51 |
53 // Init() must be called before any of the iterators are valid. | 52 // Init() must be called before any of the iterators are valid. |
54 // Returns false if ICU failed to initialize. | 53 // Returns false if ICU failed to initialize. |
55 bool Init(); | 54 bool Init(); |
56 | 55 |
57 // Return the current break position within the string, | 56 // Return the current break position within the string, |
58 // or WordIterator::npos when done. | 57 // or BreakIterator::npos when done. |
59 size_t pos() const { return pos_; } | 58 size_t pos() const { return pos_; } |
60 // Return the value of pos() returned before Advance() was last called. | 59 // Return the value of pos() returned before Advance() was last called. |
61 size_t prev() const { return prev_; } | 60 size_t prev() const { return prev_; } |
62 | 61 |
63 // Advance to the next break. Returns false if we've run past the end of | 62 // Advance to the next break. Returns false if we've run past the end of |
64 // the string. (Note that the very last "word break" is after the final | 63 // the string. (Note that the very last "word break" is after the final |
65 // character in the string, and when we advance to that position it's the | 64 // character in the string, and when we advance to that position it's the |
66 // last time Advance() returns true.) | 65 // last time Advance() returns true.) |
67 bool Advance(); | 66 bool Advance(); |
68 | 67 |
69 // Returns true if the break we just hit is the end of a word. | 68 // Returns true if the break we just hit is the end of a word. |
70 // (Otherwise, the break iterator just skipped over e.g. whitespace | 69 // (Otherwise, the break iterator just skipped over e.g. whitespace |
71 // or punctuation.) | 70 // or punctuation.) |
72 bool IsWord() const; | 71 bool IsWord() const; |
Evan Martin
2010/12/13 20:26:04
Maybe the comments on this function need to be upd
| |
73 | 72 |
74 // Return the word between prev() and pos(). | 73 // Return the word between prev() and pos(). |
75 // Advance() must have been called successfully at least once | 74 // Advance() must have been called successfully at least once |
76 // for pos() to have advanced to somewhere useful. | 75 // for pos() to have advanced to somewhere useful. |
77 string16 GetWord() const; | 76 string16 GetWord() const; |
78 | 77 |
79 private: | 78 private: |
80 // ICU iterator, avoiding ICU ubrk.h dependence. | 79 // ICU iterator, avoiding ICU ubrk.h dependence. |
81 // This is actually an ICU UBreakiterator* type, which turns out to be | 80 // This is actually an ICU UBreakiterator* type, which turns out to be |
82 // a typedef for a void* in the ICU headers. Using void* directly prevents | 81 // a typedef for a void* in the ICU headers. Using void* directly prevents |
83 // callers from needing access to the ICU public headers directory. | 82 // callers from needing access to the ICU public headers directory. |
84 void* iter_; | 83 void* iter_; |
85 | 84 |
86 // The string we're iterating over. | 85 // The string we're iterating over. |
87 const string16* string_; | 86 const string16* string_; |
88 | 87 |
89 // The breaking style (word/line). | 88 // The breaking style (word/line). |
90 BreakType break_type_; | 89 BreakType break_type_; |
91 | 90 |
92 // Previous and current iterator positions. | 91 // Previous and current iterator positions. |
93 size_t prev_, pos_; | 92 size_t prev_, pos_; |
94 | 93 |
95 DISALLOW_COPY_AND_ASSIGN(WordIterator); | 94 DISALLOW_COPY_AND_ASSIGN(BreakIterator); |
96 }; | 95 }; |
97 | 96 |
98 #endif // BASE_I18N_WORD_ITERATOR_H__ | 97 } // namespace base |
98 | |
99 #endif // BASE_I18N_BREAK_ITERATOR_H__ | |
OLD | NEW |