1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
|
<?php
// Common code for tests which focus on conversion and verification of text
// in some specific encoding
// Read a file with one character and its equivalent Unicode codepoint on each
// line, delimited by tabs
function readConversionTable($path, &$from, &$to, $utf32 = false) {
$from = array();
$to = array();
$fp = fopen($path, 'r+');
while ($line = fgets($fp, 256)) {
if ($line[0] == '#')
continue;
if (sscanf($line, "0x%x\t0x%x", $char, $codepoint) == 2) {
$codepoint = $utf32 ? pack('N', $codepoint) : pack('n', $codepoint);
if ($char <= 0xFF)
$char = chr($char); // hex codes must not have leading zero bytes
else if ($char <= 0xFFFF)
$char = pack('n', $char);
else if ($char <= 0xFFFFFF)
$char = chr($char >> 16) . pack('n', $char & 0xFFFF);
else
$char = pack('N', $char);
$from[$char] = $codepoint;
$to[$codepoint] = $char;
}
}
}
function dbgPrint($str) {
$result = '';
if (mb_check_encoding($str, 'ASCII'))
$result .= '"' . $str . '" ';
return $result . "(" . bin2hex($str) . ")";
}
function identifyValidString($goodString, $encoding) {
$result = mb_check_encoding($goodString, $encoding);
if (!$result)
die("mb_check_encoding failed on good $encoding string: " . dbgPrint($goodString));
}
function identifyInvalidString($badString, $encoding) {
$result = mb_check_encoding($badString, $encoding);
if ($result)
die("mb_check_encoding passed on bad $encoding string: " . dbgPrint($badString));
}
function testConversion($fromString, $toString, $fromEncoding, $toEncoding) {
$result = mb_convert_encoding($fromString, $toEncoding, $fromEncoding);
if ($result !== $toString)
die("mb_convert_encoding not working on $fromEncoding input: " . dbgPrint($fromString) . "\nExpected $toEncoding: " . dbgPrint($toString) . "\nActually got: " . dbgPrint($result));
}
function testValidConversion($fromString, $toString, $fromEncoding, $toEncoding) {
$illegalChars = mb_get_info('illegal_chars');
testConversion($fromString, $toString, $fromEncoding, $toEncoding);
if (mb_get_info('illegal_chars') !== $illegalChars)
die("mb_convert_encoding incremented illegal_chars on valid $fromEncoding string: " . dbgPrint($fromString) . " when converting to $toEncoding");
}
function convertValidString($fromString, $toString, $fromEncoding, $toEncoding, $bothWays = true) {
testValidConversion($fromString, $toString, $fromEncoding, $toEncoding);
if ($bothWays)
testValidConversion($toString, $fromString, $toEncoding, $fromEncoding);
}
function convertInvalidString($fromString, $toString, $fromEncoding, $toEncoding) {
$illegalChars = mb_get_info('illegal_chars');
testConversion($fromString, $toString, $fromEncoding, $toEncoding);
if (mb_get_info('illegal_chars') <= $illegalChars)
die("mb_convert_encoding did not increment illegal_chars on invalid $fromEncoding string: " . dbgPrint($fromString) . " when converting to $toEncoding");
}
function testValidString($fromString, $toString, $fromEncoding, $toEncoding, $bothWays = true) {
identifyValidString($fromString, $fromEncoding);
convertValidString($fromString, $toString, $fromEncoding, $toEncoding, $bothWays);
}
function testInvalidString($fromString, $toString, $fromEncoding, $toEncoding) {
identifyInvalidString($fromString, $fromEncoding);
convertInvalidString($fromString, $toString, $fromEncoding, $toEncoding);
}
// Only for encodings where valid characters can be concatenated together in any
// way, without any escape sequences
function testAllValidChars($charMap, $fromEncoding, $toEncoding, $bothWays = true) {
$goodChars = array_keys($charMap);
shuffle($goodChars);
while (!empty($goodChars)) {
$length = min(rand(5,10), count($goodChars));
$fromString = $toString = '';
while ($length--) {
$goodChar = array_pop($goodChars);
$fromString .= $goodChar;
$toString .= $charMap[$goodChar];
}
testValidString($fromString, $toString, $fromEncoding, $toEncoding, $bothWays);
}
}
function testAllInvalidChars($badChars, $charMap, $fromEncoding, $toEncoding, $replacement) {
$badChars = array_keys($badChars);
$goodChars = array();
while (!empty($badChars)) {
if (empty($goodChars)) {
$goodChars = array_keys($charMap);
shuffle($goodChars);
}
$goodChar = array_pop($goodChars);
$fromString = array_pop($badChars) . $goodChar;
$toString = $replacement . $charMap[$goodChar];
testInvalidString($fromString, $toString, $fromEncoding, $toEncoding);
}
}
function convertAllInvalidChars($badChars, $charMap, $fromEncoding, $toEncoding, $replacement) {
$badChars = array_keys($badChars);
$goodChars = array();
while (!empty($badChars)) {
if (empty($goodChars)) {
$goodChars = array_keys($charMap);
shuffle($goodChars);
}
$goodChar = array_pop($goodChars);
$fromString = array_pop($badChars) . $goodChar;
$toString = $replacement . $charMap[$goodChar];
convertInvalidString($fromString, $toString, $fromEncoding, $toEncoding);
}
}
function testTruncatedChars($truncated, $fromEncoding, $toEncoding, $replacement) {
$truncatedChars = array_keys($truncated);
foreach ($truncatedChars as $truncatedChar) {
testInvalidString($truncatedChar, $replacement, $fromEncoding, $toEncoding);
}
}
// For variable-width encodings, where we have an exhaustive list of
// all valid characters of any width
//
// `$startBytes` maps from first-byte values to the corresponding character length
// (For encodings where the first byte can tell you the length of a multi-byte
// character)
// Note that `$startBytes` can be partial!
function findInvalidChars($valid, &$invalid, &$truncated, $startBytes = array()) {
$invalid = array();
$truncated = array();
$prefixes = array(); /* All sequences which are not (but can start) a valid character */
foreach ($valid as $char => $unicode) {
for ($len = 1; $len < strlen($char); $len++)
$prefixes[substr($char, 0, $len)] = true;
}
$varLength = function($prefix) use($valid, $prefixes, &$invalid, &$truncated, &$varLength) {
for ($byte = 0; $byte < 256; $byte++) {
$str = $prefix . chr($byte);
if (!isset($valid[$str])) {
if (isset($prefixes[$str])) {
$truncated[$str] = true;
$varLength($str);
} else {
$invalid[$str] = true;
}
}
}
};
$fixedLength = function($prefix, $remaining) use($valid, $prefixes, &$invalid, &$truncated, &$fixedLength) {
if ($remaining == 0) {
if (!isset($valid[$prefix]))
$invalid[$prefix] = true;
} else if ($remaining == 1) {
$truncated[$prefix] = true;
for ($i = 0; $i < 256; $i++) {
$str = $prefix . chr($i);
if (!isset($valid[$str]))
$invalid[$str] = true;
}
} else {
$truncated[$prefix] = true;
for ($i = 0; $i < 256; $i++)
$fixedLength($prefix . chr($i), $remaining - 1);
}
};
for ($byte = 0; $byte < 256; $byte++) {
if (isset($startBytes[$byte])) {
$fixedLength(chr($byte), $startBytes[$byte] - 1);
} else {
$str = chr($byte);
if (!isset($valid[$str])) {
if (isset($prefixes[$str])) {
$truncated[$str] = true;
$varLength($str);
} else {
$invalid[$str] = true;
}
}
}
}
}
function testEncodingFromUTF16ConversionTable($path, $encoding, $replacement = '%', $startBytes = array()) {
srand(1000); // Make results consistent
mb_substitute_character(0x25); // '%'
readConversionTable($path, $toUnicode, $fromUnicode);
findInvalidChars($toUnicode, $invalid, $truncated, $startBytes);
testAllValidChars($toUnicode, $encoding, 'UTF-16BE');
testAllInvalidChars($invalid, $toUnicode, $encoding, 'UTF-16BE', "\x00%");
testTruncatedChars($truncated, $encoding, 'UTF-16BE', "\x00%");
echo "Tested $encoding -> UTF-16BE\n";
findInvalidChars($fromUnicode, $invalid, $unused, array_fill_keys(range(0,0xFF), 2));
convertAllInvalidChars($invalid, $fromUnicode, 'UTF-16BE', $encoding, $replacement);
echo "Tested UTF-16BE -> $encoding\n";
}
?>
|
