1
/* -*- Mode: C; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
2
/* ***** BEGIN LICENSE BLOCK *****
3
* Version: MPL 1.1/GPL 2.0/LGPL 2.1
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/
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
15
* The Original Code is Mozilla Universal charset detector code.
17
* The Initial Developer of the Original Code is
18
* Shy Shalom <shooshX@gmail.com>
19
* Portions created by the Initial Developer are Copyright (C) 2005
20
* the Initial Developer: All Rights Reserved.
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.
36
* ***** END LICENSE BLOCK ***** */
38
#ifndef nsHebrewProber_h__
39
#define nsHebrewProber_h__
41
#include "nsSBCharSetProber.h"
43
// This prober doesn't actually recognize a language or a charset.
44
// It is a helper prober for the use of the Hebrew model probers
45
class nsHebrewProber: public nsCharSetProber
48
nsHebrewProber(void) :mLogicalProb(0), mVisualProb(0) { Reset(); }
50
virtual ~nsHebrewProber(void) {}
51
virtual nsProbingState HandleData(const char* aBuf, PRUint32 aLen);
52
virtual const char* GetCharSetName();
53
virtual void Reset(void);
55
virtual nsProbingState GetState(void);
57
virtual float GetConfidence(void) { return (float)0.0; }
58
virtual void SetOpion() {}
60
void SetModelProbers(nsCharSetProber *logicalPrb, nsCharSetProber *visualPrb)
61
{ mLogicalProb = logicalPrb; mVisualProb = visualPrb; }
64
virtual void DumpStatus();
68
static PRBool isFinal(char c);
69
static PRBool isNonFinal(char c);
71
PRInt32 mFinalCharLogicalScore, mFinalCharVisualScore;
73
// The two last characters seen in the previous buffer.
74
char mPrev, mBeforePrev;
76
// These probers are owned by the group prober.
77
nsCharSetProber *mLogicalProb, *mVisualProb;
81
* ** General ideas of the Hebrew charset recognition **
83
* Four main charsets exist in Hebrew:
84
* "ISO-8859-8" - Visual Hebrew
85
* "windows-1255" - Logical Hebrew
86
* "ISO-8859-8-I" - Logical Hebrew
87
* "x-mac-hebrew" - ?? Logical Hebrew ??
89
* Both "ISO" charsets use a completely identical set of code points, whereas
90
* "windows-1255" and "x-mac-hebrew" are two different proper supersets of
91
* these code points. windows-1255 defines additional characters in the range
92
* 0x80-0x9F as some misc punctuation marks as well as some Hebrew-specific
93
* diacritics and additional 'Yiddish' ligature letters in the range 0xc0-0xd6.
94
* x-mac-hebrew defines similar additional code points but with a different
97
* As far as an average Hebrew text with no diacritics is concerned, all four
98
* charsets are identical with respect to code points. Meaning that for the
99
* main Hebrew alphabet, all four map the same values to all 27 Hebrew letters
100
* (including final letters).
102
* The dominant difference between these charsets is their directionality.
103
* "Visual" directionality means that the text is ordered as if the renderer is
104
* not aware of a BIDI rendering algorithm. The renderer sees the text and
105
* draws it from left to right. The text itself when ordered naturally is read
106
* backwards. A buffer of Visual Hebrew generally looks like so:
107
* "[last word of first line spelled backwards] [whole line ordered backwards
108
* and spelled backwards] [first word of first line spelled backwards]
109
* [end of line] [last word of second line] ... etc' "
110
* adding punctuation marks, numbers and English text to visual text is
111
* naturally also "visual" and from left to right.
113
* "Logical" directionality means the text is ordered "naturally" according to
114
* the order it is read. It is the responsibility of the renderer to display
115
* the text from right to left. A BIDI algorithm is used to place general
116
* punctuation marks, numbers and English text in the text.
118
* Texts in x-mac-hebrew are almost impossible to find on the Internet. From
119
* what little evidence I could find, it seems that its general directionality
122
* To sum up all of the above, the Hebrew probing mechanism knows about two
124
* Visual Hebrew - "ISO-8859-8" - backwards text - Words and sentences are
125
* backwards while line order is natural. For charset recognition purposes
126
* the line order is unimportant (In fact, for this implementation, even
127
* word order is unimportant).
128
* Logical Hebrew - "windows-1255" - normal, naturally ordered text.
130
* "ISO-8859-8-I" is a subset of windows-1255 and doesn't need to be
131
* specifically identified.
132
* "x-mac-hebrew" is also identified as windows-1255. A text in x-mac-hebrew
133
* that contain special punctuation marks or diacritics is displayed with
134
* some unconverted characters showing as question marks. This problem might
135
* be corrected using another model prober for x-mac-hebrew. Due to the fact
136
* that x-mac-hebrew texts are so rare, writing another model prober isn't
137
* worth the effort and performance hit.
141
* The prober is divided between two nsSBCharSetProbers and an nsHebrewProber,
142
* all of which are managed, created, fed data, inquired and deleted by the
143
* nsSBCSGroupProber. The two nsSBCharSetProbers identify that the text is in
144
* fact some kind of Hebrew, Logical or Visual. The final decision about which
145
* one is it is made by the nsHebrewProber by combining final-letter scores
146
* with the scores of the two nsSBCharSetProbers to produce a final answer.
148
* The nsSBCSGroupProber is responsible for stripping the original text of HTML
149
* tags, English characters, numbers, low-ASCII punctuation characters, spaces
150
* and new lines. It reduces any sequence of such characters to a single space.
151
* The buffer fed to each prober in the SBCS group prober is pure text in
153
* The two nsSBCharSetProbers (model probers) share the same language model:
155
* The first nsSBCharSetProber uses the model normally as any other
156
* nsSBCharSetProber does, to recognize windows-1255, upon which this model was
157
* built. The second nsSBCharSetProber is told to make the pair-of-letter
158
* lookup in the language model backwards. This in practice exactly simulates
159
* a visual Hebrew model using the windows-1255 logical Hebrew model.
161
* The nsHebrewProber is not using any language model. All it does is look for
162
* final-letter evidence suggesting the text is either logical Hebrew or visual
163
* Hebrew. Disjointed from the model probers, the results of the nsHebrewProber
164
* alone are meaningless. nsHebrewProber always returns 0.00 as confidence
165
* since it never identifies a charset by itself. Instead, the pointer to the
166
* nsHebrewProber is passed to the model probers as a helper "Name Prober".
167
* When the Group prober receives a positive identification from any prober,
168
* it asks for the name of the charset identified. If the prober queried is a
169
* Hebrew model prober, the model prober forwards the call to the
170
* nsHebrewProber to make the final decision. In the nsHebrewProber, the
171
* decision is made according to the final-letters scores maintained and Both
172
* model probers scores. The answer is returned in the form of the name of the
173
* charset identified, either "windows-1255" or "ISO-8859-8".
176
#endif /* nsHebrewProber_h__ */