CharacterEncodingDetector.h revision 34581f44cde67960fbac3ba1f191a2c063ea5145 
1/*
2 * Copyright (C) 2013 The Android Open Source Project
3 *
4 * Licensed under the Apache License, Version 2.0 (the "License");
5 * you may not use this file except in compliance with the License.
6 * You may obtain a copy of the License at
7 *
8 *      http://www.apache.org/licenses/LICENSE-2.0
9 *
10 * Unless required by applicable law or agreed to in writing, software
11 * distributed under the License is distributed on an "AS IS" BASIS,
12 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13 * See the License for the specific language governing permissions and
14 * limitations under the License.
15 */
16
17#ifndef _CHARACTER_ENCODING_DETECTOR_H
18#define _CHARACTER_ENCODING_DETECTOR_H
19
20#include <media/mediascanner.h>
21
22#include "StringArray.h"
23
24#include "unicode/ucnv.h"
25#include "unicode/ucsdet.h"
26#include "unicode/ustring.h"
27
28namespace android {
29
30class CharacterEncodingDetector {
31
32    public:
33    CharacterEncodingDetector();
34        ~CharacterEncodingDetector();
35
36        void addTag(const char *name, const char *value);
37        size_t size();
38
39        void detectAndConvert();
40        status_t getTag(int index, const char **name, const char**value);
41
42    private:
43        const UCharsetMatch *getPreferred(
44                const char *input, size_t len,
45                const UCharsetMatch** ucma, size_t matches,
46                bool *goodmatch, int *highestmatch);
47
48        bool isFrequent(const uint16_t *values, uint32_t c);
49
50        // cached name and value strings, for native encoding support.
51        // TODO: replace these with byte blob arrays that don't require the data to be
52        // singlenullbyte-terminated
53        StringArray     mNames;
54        StringArray     mValues;
55
56        UConverter*     mUtf8Conv;
57};
58
59
60
61};  // namespace android
62
63#endif
64