xref: /php-src/ext/mbstring/tests/mb_substr.phpt (revision e8141973)
1--TEST--
2mb_substr()
3--EXTENSIONS--
4mbstring
5--FILE--
6<?php
7ini_set('include_path','.');
8include_once('common.inc');
9
10// EUC-JP
11$euc_jp = mb_convert_encoding('0123この文字列は日本語です。EUC-JPを使っています。日本語は面倒臭い。', 'EUC-JP', 'UTF-8');
12// SJIS
13$sjis = mb_convert_encoding('日本語テキストです。0123456789。', 'SJIS', 'UTF-8');
14// ISO-2022-JP
15$iso2022jp = "\x1B\$B\x21\x21!r\x1B(BABC";
16// GB-18030
17$gb18030 = mb_convert_encoding('密码用户名密码名称名称', 'GB18030', 'UTF-8');
18// HZ
19$hz = "The next sentence is in GB.~{<:Ky2;S{#,NpJ)l6HK!#~}Bye.";
20// UTF-8
21$utf8 = "Greek: Σὲ γνωρίζω ἀπὸ τὴν κόψη Russian: Зарегистрируйтесь";
22// UTF-32
23$utf32 = mb_convert_encoding($utf8, 'UTF-32', 'UTF-8');
24// UTF-7
25$utf7 = mb_convert_encoding($utf8, 'UTF-7', 'UTF-8');
26
27echo "EUC-JP:\n";
28print "1: ". bin2hex(mb_substr($euc_jp, 10,  10, 'EUC-JP')) . "\n";
29print "2: ". bin2hex(mb_substr($euc_jp,  0, 100, 'EUC-JP')) . "\n";
30
31$str = mb_substr($euc_jp, 100, 10, 'EUC-JP');
32print ($str === "") ? "3 OK\n" : "BAD: " . bin2hex($str) . "\n";
33
34$str = mb_substr($euc_jp, -100, 10, 'EUC-JP');
35print ($str !== "") ? "4 OK: " . bin2hex($str) . "\n" : "BAD: " . bin2hex($str) . "\n";
36
37echo "SJIS:\n";
38print "1: " . bin2hex(mb_substr($sjis, 0, 3, 'SJIS')) . "\n";
39print "2: " . bin2hex(mb_substr($sjis, -1, null, 'SJIS')) . "\n";
40print "3: " . bin2hex(mb_substr($sjis, -5, 3, 'SJIS')) . "\n";
41print "4: " . bin2hex(mb_substr($sjis, 1, null, 'SJIS')) . "\n";
42print "5:" . bin2hex(mb_substr($sjis, 10, 0, 'SJIS')) . "\n";
43echo "-- Testing illegal SJIS byte 0x80 --\n";
44print bin2hex(mb_substr("\x80abc\x80\xA1", 3, 2, 'SJIS')) . "\n";
45print bin2hex(mb_substr("\x80abc\x80\xA1", 0, 3, 'SJIS')) . "\n";
46
47echo "SJIS-2004:\n";
48print bin2hex(mb_substr("\x80abc\x80\xA1", 3, 2, 'SJIS-2004')) . "\n";
49print bin2hex(mb_substr("\x80abc\x80\xA1", 0, 3, 'SJIS-2004')) . "\n";
50
51echo "MacJapanese:\n";
52print bin2hex(mb_substr("\x80abc\x80\xA1", 3, 2, 'MacJapanese')) . "\n";
53print bin2hex(mb_substr("\x80abc\x80\xA1", 0, 3, 'MacJapanese')) . "\n";
54
55echo "SJIS-Mobile#DOCOMO:\n";
56print bin2hex(mb_substr("\x80abc\x80\xA1", 3, 2, 'SJIS-Mobile#DOCOMO')) . "\n";
57print bin2hex(mb_substr("\x80abc\x80\xA1", 0, 3, 'SJIS-Mobile#DOCOMO')) . "\n";
58
59echo "SJIS-Mobile#KDDI:\n";
60print bin2hex(mb_substr("\x80abc\x80\xA1", 3, 2, 'SJIS-Mobile#KDDI')) . "\n";
61print bin2hex(mb_substr("\x80abc\x80\xA1", 0, 3, 'SJIS-Mobile#KDDI')) . "\n";
62
63echo "SJIS-Mobile#SoftBank:\n";
64print bin2hex(mb_substr("\x80abc\x80\xA1", 3, 2, 'SJIS-Mobile#SoftBank')) . "\n";
65print bin2hex(mb_substr("\x80abc\x80\xA1", 0, 3, 'SJIS-Mobile#SoftBank')) . "\n";
66
67echo "-- Testing MacJapanese characters which map to 3-5 codepoints each --\n";
68
69/* There are many characters in MacJapanese which map to sequences of several codepoints */
70print bin2hex(mb_substr("abc\x85\xAB\x85\xAC\x85\xAD", 0, 3, 'MacJapanese')) . "\n";
71print bin2hex(mb_substr("abc\x85\xAB\x85\xAC\x85\xAD", 3, 2, 'MacJapanese')) . "\n";
72print bin2hex(mb_substr("abc\x85\xAB\x85\xAC\x85\xAD", -2, 1, 'MacJapanese')) . "\n";
73print bin2hex(mb_substr("abc\x85\xBF\x85\xC0\x85\xC1", 0, 3, 'MacJapanese')) . "\n";
74print bin2hex(mb_substr("abc\x85\xBF\x85\xC0\x85\xC1", 3, 2, 'MacJapanese')) . "\n";
75print bin2hex(mb_substr("abc\x85\xBF\x85\xC0\x85\xC1", -2, 1, 'MacJapanese')) . "\n";
76
77echo "ISO-2022-JP:\n";
78print "1: " . bin2hex(mb_substr($iso2022jp, 0, 3, 'ISO-2022-JP')) . "\n";
79print "2: " . bin2hex(mb_substr($iso2022jp, -1, null, 'ISO-2022-JP')) . "\n";
80print "3: " . bin2hex(mb_substr($iso2022jp, -6, 3, 'ISO-2022-JP')) . "\n";
81print "4: " . bin2hex(mb_substr($iso2022jp, -3, 2, 'ISO-2022-JP')) . "\n";
82print "5: " . bin2hex(mb_substr($iso2022jp, 1, null, 'ISO-2022-JP')) . "\n";
83print "6:" . bin2hex(mb_substr($iso2022jp, 10, 0, 'ISO-2022-JP')) . "\n";
84print "7:" . bin2hex(mb_substr($iso2022jp, 100, 10, 'ISO-2022-JP')) . "\n";
85
86echo "GB-18030:\n";
87print "1: " . bin2hex(mb_substr($gb18030, 0, 3, 'GB-18030')) . "\n";
88print "2: " . bin2hex(mb_substr($gb18030, -1, null, 'GB-18030')) . "\n";
89print "3: " . bin2hex(mb_substr($gb18030, -5, 3, 'GB-18030')) . "\n";
90print "4: " . bin2hex(mb_substr($gb18030, 1, null, 'GB-18030')) . "\n";
91print "5:" . bin2hex(mb_substr($gb18030, 10, 0, 'GB-18030')) . "\n";
92
93echo "HZ:\n";
94print "1: " . mb_substr($hz, 0, 3, 'HZ') . "\n";
95print "2: " . mb_substr($hz, -1, null, 'HZ') . "\n";
96print "3: " . mb_substr($hz, -5, 3, 'HZ') . "\n";
97print "4: " . mb_substr($hz, 1, null, 'HZ') . "\n";
98print "5:" . mb_substr($hz, 10, 0, 'HZ') . "\n";
99
100echo "UTF-8:\n";
101print "1: " . mb_substr($utf8, 0, 3, 'UTF-8') . "\n";
102print "2: " . mb_substr($utf8, -1, null, 'UTF-8') . "\n";
103print "3: " . mb_substr($utf8, -5, 3, 'UTF-8') . "\n";
104print "4: " . mb_substr($utf8, 1, null, 'UTF-8') . "\n";
105print "5:" . mb_substr($utf8, 10, 0, 'UTF-8') . "\n";
106
107echo "UTF-32:\n";
108print "1: " . mb_convert_encoding(mb_substr($utf32, 0, 3, 'UTF-32'), 'UTF-8', 'UTF-32') . "\n";
109print "2: " . mb_convert_encoding(mb_substr($utf32, -1, null, 'UTF-32'), 'UTF-8', 'UTF-32') . "\n";
110print "3: " . mb_convert_encoding(mb_substr($utf32, -5, 3, 'UTF-32'), 'UTF-8', 'UTF-32') . "\n";
111print "4: " . mb_convert_encoding(mb_substr($utf32, 1, null, 'UTF-32'), 'UTF-8', 'UTF-32') . "\n";
112print "5:" . mb_convert_encoding(mb_substr($utf32, 10, 0, 'UTF-32'), 'UTF-8', 'UTF-32') . "\n";
113
114echo "UTF-7:\n";
115print "1: " . mb_convert_encoding(mb_substr($utf7, 0, 3, 'UTF-7'), 'UTF-8', 'UTF-7') . "\n";
116print "2: " . mb_convert_encoding(mb_substr($utf7, -1, null, 'UTF-7'), 'UTF-8', 'UTF-7') . "\n";
117print "3: " . mb_convert_encoding(mb_substr($utf7, -5, 3, 'UTF-7'), 'UTF-8', 'UTF-7') . "\n";
118print "4: " . mb_convert_encoding(mb_substr($utf7, 1, null, 'UTF-7'), 'UTF-8', 'UTF-7') . "\n";
119print "5:" . mb_convert_encoding(mb_substr($utf7, 10, 0, 'UTF-7'), 'UTF-8', 'UTF-7') . "\n";
120
121echo "Testing agreement with mb_strpos on invalid UTF-8 string:\n";
122/* Stefan Schiller pointed out that on invalid UTF-8 strings, character indices returned
123 * by mb_strpos would not extract the desired part of the string when passed to mb_substr.
124 * This is the test case which he provided: */
125$data = "\xF0AAA<b>";
126$pos = mb_strpos($data, "<", 0, "UTF-8");
127$out = mb_substr($data, 0, $pos, "UTF-8");
128print $out . "\n";
129
130echo "Regression:\n";
131/* During development, one >= comparison in mb_get_substr was wrongly written as >
132 * This was caught by libFuzzer */
133$str = "\xbd\xbd\xbd\xbd\xbd\xbd\xbd\xbe\xbd\xbd\xbd\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x89\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x00\x00\x00\x00\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8b\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8f\x8b\x8b\x8b\xbd\xbd\xbd\xbd\xbd\xbd\xbd\xbe\x01:O\xaa\xd3";
134echo bin2hex(mb_substr($str, 0, 128, "JIS")), "\n";
135
136/* Alex messed up when reimplementing mb_substr and, in cases where `from` is non-zero and
137 * the number of characters to extract is more than 128, miscalculated where to end the substring
138 * Thanks to Maurício Fauth for finding the issue */
139var_dump(mb_substr('Lorem ipsum dolor sit amet, consectetur adipiscing elit. Vestibulum dapibus feugiat ex non cursus. Pellentesque vestibulum tellus sit lectus.', 19, -1));
140
141?>
142--EXPECT--
143EUC-JP:
1441: c6fccbdcb8eca4c7a4b9a1a34555432d
1452: 30313233a4b3a4cecab8bbfacef3a4cfc6fccbdcb8eca4c7a4b9a1a34555432d4a50a4f2bbc8a4c3a4c6a4a4a4dea4b9a1a3c6fccbdcb8eca4cfccccc5ddbdada4a4a1a3
1463 OK
1474 OK: 30313233a4b3a4cecab8bbfacef3a4cf
148SJIS:
1491: 93fa967b8cea
1502: 8142
1513: 825582568257
1524: 967b8cea8365834c8358836782c582b781423031323334825482558256825782588142
1535:
154-- Testing illegal SJIS byte 0x80 --
155633f
1563f6162
157SJIS-2004:
158633f
1593f6162
160MacJapanese:
1616380
162806162
163SJIS-Mobile#DOCOMO:
164633f
1653f6162
166SJIS-Mobile#KDDI:
167633f
1683f6162
169SJIS-Mobile#SoftBank:
170633f
1713f6162
172-- Testing MacJapanese characters which map to 3-5 codepoints each --
173616263
1743f3f
17558
176616263
1773f3f
17878
179ISO-2022-JP:
1801: 1b2442212121721b284241
1812: 43
1823: 1b2442212121721b284241
1834: 4142
1845: 1b244221721b2842414243
1856:
1867:
187GB-18030:
1881: c3dcc2ebd3c3
1892: b3c6
1903: c2ebc3fbb3c6
1914: c2ebd3c3bba7c3fbc3dcc2ebc3fbb3c6c3fbb3c6
1925:
193HZ:
1941: The
1952: .
1963: ~{!#~}By
1974: he next sentence is in GB.~{<:Ky2;S{#,NpJ)l6HK!#~}Bye.
1985:
199UTF-8:
2001: Gre
2012: ь
2023: йте
2034: reek: Σὲ γνωρίζω ἀπὸ τὴν κόψη Russian: Зарегистрируйтесь
2045:
205UTF-32:
2061: Gre
2072: ь
2083: йте
2094: reek: Σὲ γνωρίζω ἀπὸ τὴν κόψη Russian: Зарегистрируйтесь
2105:
211UTF-7:
2121: Gre
2132: ь
2143: йте
2154: reek: Σὲ γνωρίζω ἀπὸ τὴν κόψη Russian: Зарегистрируйтесь
2165:
217Testing agreement with mb_strpos on invalid UTF-8 string:
218?AAA
219Regression:
2201b28493d3d3d3d3d3d3d3e3d3d3d1b28423f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f000000003f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f3f1b28493d3d3d3d3d3d3d3e1b2842013a4f1b28492a1b2842
221string(121) "it amet, consectetur adipiscing elit. Vestibulum dapibus feugiat ex non cursus. Pellentesque vestibulum tellus sit lectus"
222