1 : /* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
2 : /* ***** BEGIN LICENSE BLOCK *****
3 : * Version: MPL 1.1/GPL 2.0/LGPL 2.1
4 : *
5 : * The contents of this file are subject to the Mozilla Public License Version
6 : * 1.1 (the "License"); you may not use this file except in compliance with
7 : * the License. You may obtain a copy of the License at
8 : * http://www.mozilla.org/MPL/
9 : *
10 : * Software distributed under the License is distributed on an "AS IS" basis,
11 : * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
12 : * for the specific language governing rights and limitations under the
13 : * License.
14 : *
15 : * The Original Code is Novell code.
16 : *
17 : * The Initial Developer of the Original Code is Novell.
18 : * Portions created by the Initial Developer are Copyright (C) 2006
19 : * the Initial Developer. All Rights Reserved.
20 : *
21 : * Contributor(s):
22 : * Robert O'Callahan <robert@ocallahan.org> (Original Author)
23 : *
24 : * Alternatively, the contents of this file may be used under the terms of
25 : * either the GNU General Public License Version 2 or later (the "GPL"), or
26 : * the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
27 : * in which case the provisions of the GPL or the LGPL are applicable instead
28 : * of those above. If you wish to allow use of your version of this file only
29 : * under the terms of either the GPL or the LGPL, and not to allow others to
30 : * use your version of this file under the terms of the MPL, indicate your
31 : * decision by deleting the provisions above and replace them with the notice
32 : * and other provisions required by the GPL or the LGPL. If you do not delete
33 : * the provisions above, a recipient may use your version of this file under
34 : * the terms of any one of the MPL, the GPL or the LGPL.
35 : *
36 : * ***** END LICENSE BLOCK ***** */
37 :
38 : #ifndef NSLINEBREAKER_H_
39 : #define NSLINEBREAKER_H_
40 :
41 : #include "nsString.h"
42 : #include "nsTArray.h"
43 : #include "nsILineBreaker.h"
44 :
45 : class nsIAtom;
46 : class nsHyphenator;
47 :
48 : /**
49 : * A receiver of line break data.
50 : */
51 0 : class nsILineBreakSink {
52 : public:
53 : /**
54 : * Sets the break data for a substring of the associated text chunk.
55 : * One or more of these calls will be performed; the union of all substrings
56 : * will cover the entire text chunk. Substrings may overlap (i.e., we may
57 : * set the break-before state of a character more than once).
58 : * @param aBreakBefore the break-before states for the characters in the substring.
59 : * These are enum values from gfxTextRun::CompressedGlyph:
60 : * FLAG_BREAK_TYPE_NONE - no linebreak is allowed here
61 : * FLAG_BREAK_TYPE_NORMAL - a normal (whitespace) linebreak
62 : * FLAG_BREAK_TYPE_HYPHEN - a hyphenation point
63 : */
64 : virtual void SetBreaks(PRUint32 aStart, PRUint32 aLength, PRUint8* aBreakBefore) = 0;
65 :
66 : /**
67 : * Indicates which characters should be capitalized. Only called if
68 : * BREAK_NEED_CAPITALIZATION was requested.
69 : */
70 : virtual void SetCapitalization(PRUint32 aStart, PRUint32 aLength, bool* aCapitalize) = 0;
71 : };
72 :
73 : /**
74 : * A line-breaking state machine. You feed text into it via AppendText calls
75 : * and it computes the possible line breaks. Because break decisions can
76 : * require a lot of context, the breaks for a piece of text are sometimes not
77 : * known until later text has been seen (or all text ends). So breaks are
78 : * returned via a call to SetBreaks on the nsILineBreakSink object passed
79 : * with each text chunk, which might happen during the corresponding AppendText
80 : * call, or might happen during a later AppendText call or even a Reset()
81 : * call.
82 : *
83 : * The linebreak results MUST NOT depend on how the text is broken up
84 : * into AppendText calls.
85 : *
86 : * The current strategy is that we break the overall text into
87 : * whitespace-delimited "words". Then those words are passed to the nsILineBreaker
88 : * service for deeper analysis if they contain a "complex" character as described
89 : * below.
90 : *
91 : * This class also handles detection of which characters should be capitalized
92 : * for text-transform:capitalize. This is a good place to handle that because
93 : * we have all the context we need.
94 : */
95 : class nsLineBreaker {
96 : public:
97 : nsLineBreaker();
98 : ~nsLineBreaker();
99 :
100 0 : static inline bool IsSpace(PRUnichar u) { return NS_IsSpace(u); }
101 :
102 0 : static inline bool IsComplexASCIIChar(PRUnichar u)
103 : {
104 : return !((0x0030 <= u && u <= 0x0039) ||
105 : (0x0041 <= u && u <= 0x005A) ||
106 : (0x0061 <= u && u <= 0x007A) ||
107 0 : (0x000a == u));
108 : }
109 :
110 0 : static inline bool IsComplexChar(PRUnichar u)
111 : {
112 0 : return IsComplexASCIIChar(u) ||
113 0 : NS_NeedsPlatformNativeHandling(u) ||
114 : (0x1100 <= u && u <= 0x11ff) || // Hangul Jamo
115 : (0x2000 <= u && u <= 0x21ff) || // Punctuations and Symbols
116 : (0x2e80 <= u && u <= 0xd7ff) || // several CJK blocks
117 : (0xf900 <= u && u <= 0xfaff) || // CJK Compatibility Idographs
118 0 : (0xff00 <= u && u <= 0xffef); // Halfwidth and Fullwidth Forms
119 : }
120 :
121 : // Break opportunities exist at the end of each run of breakable whitespace
122 : // (see IsSpace above). Break opportunities can also exist between pairs of
123 : // non-whitespace characters, as determined by nsILineBreaker. We pass a whitespace-
124 : // delimited word to nsILineBreaker if it contains at least one character
125 : // matching IsComplexChar.
126 : // We provide flags to control on a per-chunk basis where breaks are allowed.
127 : // At any character boundary, exactly one text chunk governs whether a
128 : // break is allowed at that boundary.
129 : //
130 : // We operate on text after whitespace processing has been applied, so
131 : // other characters (e.g. tabs and newlines) may have been converted to
132 : // spaces.
133 :
134 : /**
135 : * Flags passed with each chunk of text.
136 : */
137 : enum {
138 : /*
139 : * Do not introduce a break opportunity at the start of this chunk of text.
140 : */
141 : BREAK_SUPPRESS_INITIAL = 0x01,
142 : /**
143 : * Do not introduce a break opportunity in the interior of this chunk of text.
144 : * Also, whitespace in this chunk is treated as non-breakable.
145 : */
146 : BREAK_SUPPRESS_INSIDE = 0x02,
147 : /**
148 : * The sink currently is already set up to have no breaks in it;
149 : * if no breaks are possible, nsLineBreaker does not need to call
150 : * SetBreaks on it. This is useful when handling large quantities of
151 : * preformatted text; the textruns will never have any breaks set on them,
152 : * and there is no need to ever actually scan the text for breaks, except
153 : * at the end of textruns in case context is needed for following breakable
154 : * text.
155 : */
156 : BREAK_SKIP_SETTING_NO_BREAKS = 0x04,
157 : /**
158 : * We need to be notified of characters that should be capitalized
159 : * (as in text-transform:capitalize) in this chunk of text.
160 : */
161 : BREAK_NEED_CAPITALIZATION = 0x08,
162 : /**
163 : * Auto-hyphenation is enabled, so we need to get a hyphenator
164 : * (if available) and use it to find breakpoints.
165 : */
166 : BREAK_USE_AUTO_HYPHENATION = 0x10
167 : };
168 :
169 : /**
170 : * Append "invisible whitespace". This acts like whitespace, but there is
171 : * no actual text associated with it. Only the BREAK_SUPPRESS_INSIDE flag
172 : * is relevant here.
173 : */
174 : nsresult AppendInvisibleWhitespace(PRUint32 aFlags);
175 :
176 : /**
177 : * Feed Unicode text into the linebreaker for analysis. aLength must be
178 : * nonzero.
179 : * @param aSink can be null if the breaks are not actually needed (we may
180 : * still be setting up state for later breaks)
181 : */
182 : nsresult AppendText(nsIAtom* aLangGroup, const PRUnichar* aText, PRUint32 aLength,
183 : PRUint32 aFlags, nsILineBreakSink* aSink);
184 : /**
185 : * Feed 8-bit text into the linebreaker for analysis. aLength must be nonzero.
186 : * @param aSink can be null if the breaks are not actually needed (we may
187 : * still be setting up state for later breaks)
188 : */
189 : nsresult AppendText(nsIAtom* aLangGroup, const PRUint8* aText, PRUint32 aLength,
190 : PRUint32 aFlags, nsILineBreakSink* aSink);
191 : /**
192 : * Reset all state. This means the current run has ended; any outstanding
193 : * calls through nsILineBreakSink are made, and all outstanding references to
194 : * nsILineBreakSink objects are dropped.
195 : * After this call, this linebreaker can be reused.
196 : * This must be called at least once between any call to AppendText() and
197 : * destroying the object.
198 : * @param aTrailingBreak this is set to true when there is a break opportunity
199 : * at the end of the text. This will normally only be declared true when there
200 : * is breakable whitespace at the end.
201 : */
202 : nsresult Reset(bool* aTrailingBreak);
203 :
204 : private:
205 : // This is a list of text sources that make up the "current word" (i.e.,
206 : // run of text which does not contain any whitespace). All the mLengths
207 : // are are nonzero, these cannot overlap.
208 0 : struct TextItem {
209 0 : TextItem(nsILineBreakSink* aSink, PRUint32 aSinkOffset, PRUint32 aLength,
210 : PRUint32 aFlags)
211 0 : : mSink(aSink), mSinkOffset(aSinkOffset), mLength(aLength), mFlags(aFlags) {}
212 :
213 : nsILineBreakSink* mSink;
214 : PRUint32 mSinkOffset;
215 : PRUint32 mLength;
216 : PRUint32 mFlags;
217 : };
218 :
219 : // State for the nonwhitespace "word" that started in previous text and hasn't
220 : // finished yet.
221 :
222 : // When the current word ends, this computes the linebreak opportunities
223 : // *inside* the word (excluding either end) and sets them through the
224 : // appropriate sink(s). Then we clear the current word state.
225 : nsresult FlushCurrentWord();
226 :
227 : void UpdateCurrentWordLangGroup(nsIAtom *aLangGroup);
228 :
229 : void FindHyphenationPoints(nsHyphenator *aHyphenator,
230 : const PRUnichar *aTextStart,
231 : const PRUnichar *aTextLimit,
232 : PRUint8 *aBreakState);
233 :
234 : nsAutoTArray<PRUnichar,100> mCurrentWord;
235 : // All the items that contribute to mCurrentWord
236 : nsAutoTArray<TextItem,2> mTextItems;
237 : nsIAtom* mCurrentWordLangGroup;
238 : bool mCurrentWordContainsMixedLang;
239 : bool mCurrentWordContainsComplexChar;
240 :
241 : // True if the previous character was breakable whitespace
242 : bool mAfterBreakableSpace;
243 : // True if a break must be allowed at the current position because
244 : // a run of breakable whitespace ends here
245 : bool mBreakHere;
246 : };
247 :
248 : #endif /*NSLINEBREAKER_H_*/
|