1 | /* |
2 | * Copyright (C) 2008, 2009 Google Inc. All rights reserved. |
3 | * |
4 | * Redistribution and use in source and binary forms, with or without |
5 | * modification, are permitted provided that the following conditions are |
6 | * met: |
7 | * |
8 | * * Redistributions of source code must retain the above copyright |
9 | * notice, this list of conditions and the following disclaimer. |
10 | * * Redistributions in binary form must reproduce the above |
11 | * copyright notice, this list of conditions and the following disclaimer |
12 | * in the documentation and/or other materials provided with the |
13 | * distribution. |
14 | * * Neither the name of Google Inc. nor the names of its |
15 | * contributors may be used to endorse or promote products derived from |
16 | * this software without specific prior written permission. |
17 | * |
18 | * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS |
19 | * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT |
20 | * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR |
21 | * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT |
22 | * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, |
23 | * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT |
24 | * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, |
25 | * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY |
26 | * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT |
27 | * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE |
28 | * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
29 | */ |
30 | |
31 | #include "config.h" |
32 | #include "TextEncodingDetector.h" |
33 | |
34 | #include "TextEncoding.h" |
35 | #include <unicode/ucnv.h> |
36 | #include <unicode/ucsdet.h> |
37 | |
38 | namespace WebCore { |
39 | |
40 | bool detectTextEncoding(const char* data, size_t len, |
41 | const char* hintEncodingName, |
42 | TextEncoding* detectedEncoding) |
43 | { |
44 | *detectedEncoding = TextEncoding(); |
45 | int matchesCount = 0; |
46 | UErrorCode status = U_ZERO_ERROR; |
47 | UCharsetDetector* detector = ucsdet_open(&status); |
48 | if (U_FAILURE(status)) |
49 | return false; |
50 | ucsdet_enableInputFilter(detector, true); |
51 | ucsdet_setText(detector, data, static_cast<int32_t>(len), &status); |
52 | if (U_FAILURE(status)) |
53 | return false; |
54 | |
55 | // FIXME: A few things we can do other than improving |
56 | // the ICU detector itself. |
57 | // 1. Use ucsdet_detectAll and pick the most likely one given |
58 | // "the context" (parent-encoding, referrer encoding, etc). |
59 | // 2. 'Emulate' Firefox/IE's non-Universal detectors (e.g. |
60 | // Chinese, Japanese, Russian, Korean and Hebrew) by picking the |
61 | // encoding with a highest confidence among the detector-specific |
62 | // limited set of candidate encodings. |
63 | // Below is a partial implementation of the first part of what's outlined |
64 | // above. |
65 | const UCharsetMatch** matches = ucsdet_detectAll(detector, &matchesCount, &status); |
66 | if (U_FAILURE(status)) { |
67 | ucsdet_close(detector); |
68 | return false; |
69 | } |
70 | |
71 | const char* encoding = 0; |
72 | if (hintEncodingName) { |
73 | TextEncoding hintEncoding(hintEncodingName); |
74 | // 10 is the minimum confidence value consistent with the codepoint |
75 | // allocation in a given encoding. The size of a chunk passed to |
76 | // us varies even for the same html file (apparently depending on |
77 | // the network load). When we're given a rather short chunk, we |
78 | // don't have a sufficiently reliable signal other than the fact that |
79 | // the chunk is consistent with a set of encodings. So, instead of |
80 | // setting an arbitrary threshold, we have to scan all the encodings |
81 | // consistent with the data. |
82 | const int32_t kThreshold = 10; |
83 | for (int i = 0; i < matchesCount; ++i) { |
84 | int32_t confidence = ucsdet_getConfidence(matches[i], &status); |
85 | if (U_FAILURE(status)) { |
86 | status = U_ZERO_ERROR; |
87 | continue; |
88 | } |
89 | if (confidence < kThreshold) |
90 | break; |
91 | const char* matchEncoding = ucsdet_getName(matches[i], &status); |
92 | if (U_FAILURE(status)) { |
93 | status = U_ZERO_ERROR; |
94 | continue; |
95 | } |
96 | if (TextEncoding(matchEncoding) == hintEncoding) { |
97 | encoding = hintEncodingName; |
98 | break; |
99 | } |
100 | } |
101 | } |
102 | // If no match is found so far, just pick the top match. |
103 | // This can happen, say, when a parent frame in EUC-JP refers to |
104 | // a child frame in Shift_JIS and both frames do NOT specify the encoding |
105 | // making us resort to auto-detection (when it IS turned on). |
106 | if (!encoding && matchesCount > 0) |
107 | encoding = ucsdet_getName(matches[0], &status); |
108 | if (U_SUCCESS(status)) { |
109 | *detectedEncoding = TextEncoding(encoding); |
110 | ucsdet_close(detector); |
111 | return true; |
112 | } |
113 | ucsdet_close(detector); |
114 | return false; |
115 | } |
116 | |
117 | } |
118 | |