xref: /third_party/icu/icu4c/source/i18n/csrutf8.cpp (revision 2e5b6d6d) 
1// © 2016 and later: Unicode, Inc. and others.
2// License & terms of use: http://www.unicode.org/copyright.html
3/*
4 **********************************************************************
5 *   Copyright (C) 2005-2014, International Business Machines
6 *   Corporation and others.  All Rights Reserved.
7 **********************************************************************
8 */
9
10#include "unicode/utypes.h"
11
12#if !UCONFIG_NO_CONVERSION
13
14#include "csrutf8.h"
15#include "csmatch.h"
16
17U_NAMESPACE_BEGIN
18
19CharsetRecog_UTF8::~CharsetRecog_UTF8()
20{
21    // nothing to do
22}
23
24const char *CharsetRecog_UTF8::getName() const
25{
26    return "UTF-8";
27}
28
29UBool CharsetRecog_UTF8::match(InputText* input, CharsetMatch *results) const {
30    bool hasBOM = false;
31    int32_t numValid = 0;
32    int32_t numInvalid = 0;
33    const uint8_t *inputBytes = input->fRawInput;
34    int32_t i;
35    int32_t trailBytes = 0;
36    int32_t confidence;
37
38    if (input->fRawLength >= 3 &&
39        inputBytes[0] == 0xEF && inputBytes[1] == 0xBB && inputBytes[2] == 0xBF) {
40            hasBOM = true;
41    }
42
43    // Scan for multi-byte sequences
44    for (i=0; i < input->fRawLength; i += 1) {
45        int32_t b = inputBytes[i];
46
47        if ((b & 0x80) == 0) {
48            continue;   // ASCII
49        }
50
51        // Hi bit on char found.  Figure out how long the sequence should be
52        if ((b & 0x0E0) == 0x0C0) {
53            trailBytes = 1;
54        } else if ((b & 0x0F0) == 0x0E0) {
55            trailBytes = 2;
56        } else if ((b & 0x0F8) == 0xF0) {
57            trailBytes = 3;
58        } else {
59            numInvalid += 1;
60            continue;
61        }
62
63        // Verify that we've got the right number of trail bytes in the sequence
64        for (;;) {
65            i += 1;
66
67            if (i >= input->fRawLength) {
68                break;
69            }
70
71            b = inputBytes[i];
72
73            if ((b & 0xC0) != 0x080) {
74                numInvalid += 1;
75                break;
76            }
77
78            if (--trailBytes == 0) {
79                numValid += 1;
80                break;
81            }
82        }
83
84    }
85
86    // Cook up some sort of confidence score, based on presence of a BOM
87    //    and the existence of valid and/or invalid multi-byte sequences.
88    confidence = 0;
89    if (hasBOM && numInvalid == 0) {
90        confidence = 100;
91    } else if (hasBOM && numValid > numInvalid*10) {
92        confidence = 80;
93    } else if (numValid > 3 && numInvalid == 0) {
94        confidence = 100;
95    } else if (numValid > 0 && numInvalid == 0) {
96        confidence = 80;
97    } else if (numValid == 0 && numInvalid == 0) {
98        // Plain ASCII. Confidence must be > 10, it's more likely than UTF-16, which
99        //              accepts ASCII with confidence = 10.
100        confidence = 15;
101    } else if (numValid > numInvalid*10) {
102        // Probably corrupt utf-8 data.  Valid sequences aren't likely by chance.
103        confidence = 25;
104    }
105
106    results->set(input, this, confidence);
107    return (confidence > 0);
108}
109
110U_NAMESPACE_END
111#endif
112