1
|
|
|
<?php |
2
|
|
|
/* |
3
|
|
|
* YOURLS |
4
|
|
|
* Function library for anything related to formatting / validating / sanitizing |
5
|
|
|
*/ |
6
|
|
|
|
7
|
|
|
/** |
8
|
|
|
* Convert an integer (1337) to a string (3jk). |
9
|
|
|
* |
10
|
|
|
*/ |
11
|
|
|
function yourls_int2string( $num, $chars = null ) { |
12
|
|
|
if( $chars == null ) |
13
|
|
|
$chars = yourls_get_shorturl_charset(); |
14
|
|
|
$string = ''; |
15
|
|
|
$len = strlen( $chars ); |
16
|
|
|
while( $num >= $len ) { |
17
|
|
|
$mod = bcmod( $num, $len ); |
18
|
|
|
$num = bcdiv( $num, $len ); |
19
|
|
|
$string = $chars[ $mod ] . $string; |
20
|
|
|
} |
21
|
|
|
$string = $chars[ intval( $num ) ] . $string; |
22
|
|
|
|
23
|
|
|
return yourls_apply_filter( 'int2string', $string, $num, $chars ); |
24
|
|
|
} |
25
|
|
|
|
26
|
|
|
/** |
27
|
|
|
* Convert a string (3jk) to an integer (1337) |
28
|
|
|
* |
29
|
|
|
*/ |
30
|
|
|
function yourls_string2int( $string, $chars = null ) { |
31
|
|
|
if( $chars == null ) |
32
|
|
|
$chars = yourls_get_shorturl_charset(); |
33
|
|
|
$integer = 0; |
34
|
|
|
$string = strrev( $string ); |
35
|
|
|
$baselen = strlen( $chars ); |
36
|
|
|
$inputlen = strlen( $string ); |
37
|
|
|
for ($i = 0; $i < $inputlen; $i++) { |
38
|
|
|
$index = strpos( $chars, $string[$i] ); |
39
|
|
|
$integer = bcadd( $integer, bcmul( $index, bcpow( $baselen, $i ) ) ); |
40
|
|
|
} |
41
|
|
|
|
42
|
|
|
return yourls_apply_filter( 'string2int', $integer, $string, $chars ); |
43
|
|
|
} |
44
|
|
|
|
45
|
|
|
/** |
46
|
|
|
* Return a unique(ish) hash for a string to be used as a valid HTML id |
47
|
|
|
* |
48
|
|
|
*/ |
49
|
|
|
function yourls_string2htmlid( $string ) { |
50
|
|
|
return yourls_apply_filter( 'string2htmlid', 'y'.abs( crc32( $string ) ) ); |
51
|
|
|
} |
52
|
|
|
|
53
|
|
|
/** |
54
|
|
|
* Make sure a link keyword (ie "1fv" as in "http://sho.rt/1fv") is valid. |
55
|
|
|
* |
56
|
|
|
*/ |
57
|
|
|
function yourls_sanitize_string( $string ) { |
58
|
|
|
// make a regexp pattern with the shorturl charset, and remove everything but this |
59
|
|
|
$pattern = yourls_make_regexp_pattern( yourls_get_shorturl_charset() ); |
60
|
|
|
$valid = substr( preg_replace( '![^'.$pattern.']!', '', $string ), 0, 199 ); |
61
|
|
|
|
62
|
|
|
return yourls_apply_filter( 'sanitize_string', $valid, $string ); |
63
|
|
|
} |
64
|
|
|
|
65
|
|
|
/** |
66
|
|
|
* Alias function. I was always getting it wrong. |
67
|
|
|
* |
68
|
|
|
*/ |
69
|
|
|
function yourls_sanitize_keyword( $keyword ) { |
70
|
|
|
return yourls_sanitize_string( $keyword ); |
71
|
|
|
} |
72
|
|
|
|
73
|
|
|
/** |
74
|
|
|
* Sanitize a page title. No HTML per W3C http://www.w3.org/TR/html401/struct/global.html#h-7.4.2 |
75
|
|
|
* |
76
|
|
|
* |
77
|
|
|
* @since 1.5 |
78
|
|
|
* @param string $unsafe_title Title, potentially unsafe |
79
|
|
|
* @param string $fallback Optional fallback if after sanitization nothing remains |
80
|
|
|
* @return string Safe title |
81
|
|
|
*/ |
82
|
|
|
function yourls_sanitize_title( $unsafe_title, $fallback = '' ) { |
83
|
|
|
$title = $unsafe_title; |
84
|
|
|
$title = strip_tags( $title ); |
85
|
|
|
$title = preg_replace( "/\s+/", ' ', trim( $title ) ); |
86
|
|
|
|
87
|
|
|
if ( '' === $title || false === $title ) { |
88
|
|
|
$title = $fallback; |
89
|
|
|
} |
90
|
|
|
|
91
|
|
|
return yourls_apply_filter( 'sanitize_title', $title, $unsafe_title, $fallback ); |
92
|
|
|
} |
93
|
|
|
|
94
|
|
|
/** |
95
|
|
|
* A few sanity checks on the URL. Used for redirection or DB. For display purpose, see yourls_esc_url() |
96
|
|
|
* |
97
|
|
|
* @param string $unsafe_url unsafe URL |
98
|
|
|
* @param array $protocols Optional allowed protocols, default to global $yourls_allowedprotocols |
99
|
|
|
* @return string Safe URL |
100
|
|
|
*/ |
101
|
|
|
function yourls_sanitize_url( $unsafe_url, $protocols = array() ) { |
102
|
|
|
$url = yourls_esc_url( $unsafe_url, 'redirection', $protocols ); |
103
|
|
|
return yourls_apply_filter( 'sanitize_url', $url, $unsafe_url ); |
104
|
|
|
} |
105
|
|
|
|
106
|
|
|
/** |
107
|
|
|
* Perform a replacement while a string is found, eg $subject = '%0%0%0DDD', $search ='%0D' -> $result ='' |
108
|
|
|
* |
109
|
|
|
* Stolen from WP's _deep_replace |
110
|
|
|
* |
111
|
|
|
*/ |
112
|
|
|
function yourls_deep_replace( $search, $subject ){ |
113
|
|
|
$found = true; |
114
|
|
|
while($found) { |
115
|
|
|
$found = false; |
116
|
|
|
foreach( (array) $search as $val ) { |
117
|
|
|
while( strpos( $subject, $val ) !== false ) { |
118
|
|
|
$found = true; |
119
|
|
|
$subject = str_replace( $val, '', $subject ); |
120
|
|
|
} |
121
|
|
|
} |
122
|
|
|
} |
123
|
|
|
|
124
|
|
|
return $subject; |
125
|
|
|
} |
126
|
|
|
|
127
|
|
|
/** |
128
|
|
|
* Make sure an integer is a valid integer (PHP's intval() limits to too small numbers) |
129
|
|
|
* |
130
|
|
|
*/ |
131
|
|
|
function yourls_sanitize_int( $in ) { |
132
|
|
|
return ( substr( preg_replace( '/[^0-9]/', '', strval( $in ) ), 0, 20 ) ); |
133
|
|
|
} |
134
|
|
|
|
135
|
|
|
/** |
136
|
|
|
* Escape a string or an array of strings before DB usage. ALWAYS escape before using in a SQL query. Thanks. |
137
|
|
|
* |
138
|
|
|
* @param string|array $data string or array of strings to be escaped |
139
|
|
|
* @return string|array escaped data |
140
|
|
|
*/ |
141
|
|
|
function yourls_escape( $data ) { |
142
|
|
|
if( is_array( $data ) ) { |
143
|
|
|
foreach( $data as $k => $v ) { |
144
|
|
|
if( is_array( $v ) ) { |
145
|
|
|
$data[ $k ] = yourls_escape( $v ); |
146
|
|
|
} else { |
147
|
|
|
$data[ $k ] = yourls_escape_real( $v ); |
148
|
|
|
} |
149
|
|
|
} |
150
|
|
|
} else { |
151
|
|
|
$data = yourls_escape_real( $data ); |
152
|
|
|
} |
153
|
|
|
|
154
|
|
|
return $data; |
155
|
|
|
} |
156
|
|
|
|
157
|
|
|
/** |
158
|
|
|
* "Real" escape. This function should NOT be called directly. Use yourls_escape() instead. |
159
|
|
|
* |
160
|
|
|
* This function uses a "real" escape if possible, using PDO, MySQL or MySQLi functions, |
161
|
|
|
* with a fallback to a "simple" addslashes |
162
|
|
|
* If you're implementing a custom DB engine or a custom cache system, you can define an |
163
|
|
|
* escape function using filter 'custom_escape_real' |
164
|
|
|
* |
165
|
|
|
* @since 1.7 |
166
|
|
|
* @param string $a string to be escaped |
|
|
|
|
167
|
|
|
* @return string escaped string |
168
|
|
|
*/ |
169
|
|
|
function yourls_escape_real( $string ) { |
170
|
|
|
global $ydb; |
171
|
|
|
if( isset( $ydb ) && ( $ydb instanceof ezSQLcore ) ) |
172
|
|
|
return $ydb->escape( $string ); |
173
|
|
|
|
174
|
|
|
// YOURLS DB classes have been bypassed by a custom DB engine or a custom cache layer |
175
|
|
|
return yourls_apply_filter( 'custom_escape_real', addslashes( $string ), $string ); |
176
|
|
|
} |
177
|
|
|
|
178
|
|
|
/** |
179
|
|
|
* Sanitize an IP address |
180
|
|
|
* |
181
|
|
|
*/ |
182
|
|
|
function yourls_sanitize_ip( $ip ) { |
183
|
|
|
return preg_replace( '/[^0-9a-fA-F:., ]/', '', $ip ); |
184
|
|
|
} |
185
|
|
|
|
186
|
|
|
/** |
187
|
|
|
* Make sure a date is m(m)/d(d)/yyyy, return false otherwise |
188
|
|
|
* |
189
|
|
|
*/ |
190
|
|
|
function yourls_sanitize_date( $date ) { |
191
|
|
|
if( !preg_match( '!^\d{1,2}/\d{1,2}/\d{4}$!' , $date ) ) { |
192
|
|
|
return false; |
193
|
|
|
} |
194
|
|
|
return $date; |
195
|
|
|
} |
196
|
|
|
|
197
|
|
|
/** |
198
|
|
|
* Sanitize a date for SQL search. Return false if malformed input. |
199
|
|
|
* |
200
|
|
|
*/ |
201
|
|
|
function yourls_sanitize_date_for_sql( $date ) { |
202
|
|
|
if( !yourls_sanitize_date( $date ) ) |
203
|
|
|
return false; |
204
|
|
|
return date( 'Y-m-d', strtotime( $date ) ); |
205
|
|
|
} |
206
|
|
|
|
207
|
|
|
/** |
208
|
|
|
* Return trimmed string |
209
|
|
|
* |
210
|
|
|
*/ |
211
|
|
|
function yourls_trim_long_string( $string, $length = 60, $append = '[...]' ) { |
212
|
|
|
$newstring = $string; |
213
|
|
|
if ( mb_strlen( $newstring ) > $length ) { |
214
|
|
|
$newstring = mb_substr( $newstring, 0, $length - mb_strlen( $append ), 'UTF-8' ) . $append; |
215
|
|
|
} |
216
|
|
|
return yourls_apply_filter( 'trim_long_string', $newstring, $string, $length, $append ); |
217
|
|
|
} |
218
|
|
|
|
219
|
|
|
/** |
220
|
|
|
* Sanitize a version number (1.4.1-whatever-RC1 -> 1.4.1) |
221
|
|
|
* |
222
|
|
|
* @since 1.4.1 |
223
|
|
|
* @param string $ver Version number |
224
|
|
|
* @return string Sanitized version number |
225
|
|
|
*/ |
226
|
|
|
function yourls_sanitize_version( $ver ) { |
227
|
|
|
preg_match( '/(^[0-9.]+).*$/', $ver, $matches ); |
228
|
|
|
return isset( $matches[1] ) ? trim( $matches[1], '.' ) : ''; |
229
|
|
|
} |
230
|
|
|
|
231
|
|
|
/** |
232
|
|
|
* Sanitize a filename (no Win32 stuff) |
233
|
|
|
* |
234
|
|
|
*/ |
235
|
|
|
function yourls_sanitize_filename( $file ) { |
236
|
|
|
$file = str_replace( '\\', '/', $file ); // sanitize for Win32 installs |
237
|
|
|
$file = preg_replace( '|/+|' ,'/', $file ); // remove any duplicate slash |
238
|
|
|
return $file; |
239
|
|
|
} |
240
|
|
|
|
241
|
|
|
/** |
242
|
|
|
* Check if a string seems to be UTF-8. Stolen from WP. |
243
|
|
|
* |
244
|
|
|
*/ |
245
|
|
|
function yourls_seems_utf8( $str ) { |
246
|
|
|
$length = strlen( $str ); |
247
|
|
|
for ( $i=0; $i < $length; $i++ ) { |
248
|
|
|
$c = ord( $str[ $i ] ); |
249
|
|
|
if ( $c < 0x80 ) $n = 0; # 0bbbbbbb |
250
|
|
|
elseif (($c & 0xE0) == 0xC0) $n=1; # 110bbbbb |
251
|
|
|
elseif (($c & 0xF0) == 0xE0) $n=2; # 1110bbbb |
252
|
|
|
elseif (($c & 0xF8) == 0xF0) $n=3; # 11110bbb |
253
|
|
|
elseif (($c & 0xFC) == 0xF8) $n=4; # 111110bb |
254
|
|
|
elseif (($c & 0xFE) == 0xFC) $n=5; # 1111110b |
255
|
|
|
else return false; # Does not match any model |
256
|
|
|
for ($j=0; $j<$n; $j++) { # n bytes matching 10bbbbbb follow ? |
257
|
|
|
if ((++$i == $length) || ((ord($str[$i]) & 0xC0) != 0x80)) |
258
|
|
|
return false; |
259
|
|
|
} |
260
|
|
|
} |
261
|
|
|
return true; |
262
|
|
|
} |
263
|
|
|
|
264
|
|
|
|
265
|
|
|
/** |
266
|
|
|
* Check for PCRE /u modifier support. Stolen from WP. |
267
|
|
|
* |
268
|
|
|
* Just in case "PCRE is not compiled with PCRE_UTF8" which seems to happen |
269
|
|
|
* on some distros even for PHP 5.3 |
270
|
|
|
* |
271
|
|
|
* @since 1.7.1 |
272
|
|
|
* |
273
|
|
|
* @return bool whether there's /u support or not |
274
|
|
|
*/ |
275
|
|
|
function yourls_supports_pcre_u() { |
276
|
|
|
static $utf8_pcre; |
277
|
|
|
if( !isset( $utf8_pcre ) ) { |
278
|
|
|
$utf8_pcre = (bool) @preg_match( '/^./u', 'a' ); |
279
|
|
|
} |
280
|
|
|
return $utf8_pcre; |
281
|
|
|
} |
282
|
|
|
|
283
|
|
|
/** |
284
|
|
|
* Checks for invalid UTF8 in a string. Stolen from WP |
285
|
|
|
* |
286
|
|
|
* @since 1.6 |
287
|
|
|
* |
288
|
|
|
* @param string $string The text which is to be checked. |
289
|
|
|
* @param boolean $strip Optional. Whether to attempt to strip out invalid UTF8. Default is false. |
290
|
|
|
* @return string The checked text. |
291
|
|
|
*/ |
292
|
|
|
function yourls_check_invalid_utf8( $string, $strip = false ) { |
293
|
|
|
$string = (string) $string; |
294
|
|
|
|
295
|
|
|
if ( 0 === strlen( $string ) ) { |
296
|
|
|
return ''; |
297
|
|
|
} |
298
|
|
|
|
299
|
|
|
// We can't demand utf8 in the PCRE installation, so just return the string in those cases |
300
|
|
|
if ( ! yourls_supports_pcre_u() ) { |
301
|
|
|
return $string; |
302
|
|
|
} |
303
|
|
|
|
304
|
|
|
// preg_match fails when it encounters invalid UTF8 in $string |
305
|
|
|
if ( 1 === @preg_match( '/^./us', $string ) ) { |
306
|
|
|
return $string; |
307
|
|
|
} |
308
|
|
|
|
309
|
|
|
// Attempt to strip the bad chars if requested (not recommended) |
310
|
|
|
if ( $strip && function_exists( 'iconv' ) ) { |
311
|
|
|
return iconv( 'utf-8', 'utf-8', $string ); |
312
|
|
|
} |
313
|
|
|
|
314
|
|
|
return ''; |
315
|
|
|
} |
316
|
|
|
|
317
|
|
|
/** |
318
|
|
|
* Converts a number of special characters into their HTML entities. Stolen from WP. |
319
|
|
|
* |
320
|
|
|
* Specifically deals with: &, <, >, ", and '. |
321
|
|
|
* |
322
|
|
|
* $quote_style can be set to ENT_COMPAT to encode " to |
323
|
|
|
* ", or ENT_QUOTES to do both. Default is ENT_NOQUOTES where no quotes are encoded. |
324
|
|
|
* |
325
|
|
|
* @since 1.6 |
326
|
|
|
* |
327
|
|
|
* @param string $string The text which is to be encoded. |
328
|
|
|
* @param mixed $quote_style Optional. Converts double quotes if set to ENT_COMPAT, both single and double if set to ENT_QUOTES or none if set to ENT_NOQUOTES. Also compatible with old values; converting single quotes if set to 'single', double if set to 'double' or both if otherwise set. Default is ENT_NOQUOTES. |
329
|
|
|
* @param boolean $double_encode Optional. Whether to encode existing html entities. Default is false. |
330
|
|
|
* @return string The encoded text with HTML entities. |
331
|
|
|
*/ |
332
|
|
|
function yourls_specialchars( $string, $quote_style = ENT_NOQUOTES, $double_encode = false ) { |
333
|
|
|
$string = (string) $string; |
334
|
|
|
|
335
|
|
|
if ( 0 === strlen( $string ) ) |
336
|
|
|
return ''; |
337
|
|
|
|
338
|
|
|
// Don't bother if there are no specialchars - saves some processing |
339
|
|
|
if ( ! preg_match( '/[&<>"\']/', $string ) ) |
340
|
|
|
return $string; |
341
|
|
|
|
342
|
|
|
// Account for the previous behaviour of the function when the $quote_style is not an accepted value |
343
|
|
View Code Duplication |
if ( empty( $quote_style ) ) |
|
|
|
|
344
|
|
|
$quote_style = ENT_NOQUOTES; |
345
|
|
|
elseif ( ! in_array( $quote_style, array( 0, 2, 3, 'single', 'double' ), true ) ) |
346
|
|
|
$quote_style = ENT_QUOTES; |
347
|
|
|
|
348
|
|
|
$charset = 'UTF-8'; |
349
|
|
|
|
350
|
|
|
$_quote_style = $quote_style; |
351
|
|
|
|
352
|
|
|
if ( $quote_style === 'double' ) { |
353
|
|
|
$quote_style = ENT_COMPAT; |
354
|
|
|
$_quote_style = ENT_COMPAT; |
355
|
|
|
} elseif ( $quote_style === 'single' ) { |
356
|
|
|
$quote_style = ENT_NOQUOTES; |
357
|
|
|
} |
358
|
|
|
|
359
|
|
|
// Handle double encoding ourselves |
360
|
|
|
if ( $double_encode ) { |
361
|
|
|
$string = @htmlspecialchars( $string, $quote_style, $charset ); |
362
|
|
|
} else { |
363
|
|
|
// Decode & into & |
364
|
|
|
$string = yourls_specialchars_decode( $string, $_quote_style ); |
365
|
|
|
|
366
|
|
|
// Guarantee every &entity; is valid or re-encode the & |
367
|
|
|
$string = yourls_kses_normalize_entities( $string ); |
368
|
|
|
|
369
|
|
|
// Now re-encode everything except &entity; |
370
|
|
|
$string = preg_split( '/(&#?x?[0-9a-z]+;)/i', $string, -1, PREG_SPLIT_DELIM_CAPTURE ); |
371
|
|
|
|
372
|
|
|
for ( $i = 0; $i < count( $string ); $i += 2 ) |
|
|
|
|
373
|
|
|
$string[$i] = @htmlspecialchars( $string[$i], $quote_style, $charset ); |
374
|
|
|
|
375
|
|
|
$string = implode( '', $string ); |
376
|
|
|
} |
377
|
|
|
|
378
|
|
|
// Backwards compatibility |
379
|
|
|
if ( 'single' === $_quote_style ) |
380
|
|
|
$string = str_replace( "'", ''', $string ); |
381
|
|
|
|
382
|
|
|
return $string; |
383
|
|
|
} |
384
|
|
|
|
385
|
|
|
/** |
386
|
|
|
* Converts a number of HTML entities into their special characters. Stolen from WP. |
387
|
|
|
* |
388
|
|
|
* Specifically deals with: &, <, >, ", and '. |
389
|
|
|
* |
390
|
|
|
* $quote_style can be set to ENT_COMPAT to decode " entities, |
391
|
|
|
* or ENT_QUOTES to do both " and '. Default is ENT_NOQUOTES where no quotes are decoded. |
392
|
|
|
* |
393
|
|
|
* @since 1.6 |
394
|
|
|
* |
395
|
|
|
* @param string $string The text which is to be decoded. |
396
|
|
|
* @param mixed $quote_style Optional. Converts double quotes if set to ENT_COMPAT, both single and double if set to ENT_QUOTES or none if set to ENT_NOQUOTES. Also compatible with old _wp_specialchars() values; converting single quotes if set to 'single', double if set to 'double' or both if otherwise set. Default is ENT_NOQUOTES. |
397
|
|
|
* @return string The decoded text without HTML entities. |
398
|
|
|
*/ |
399
|
|
|
function yourls_specialchars_decode( $string, $quote_style = ENT_NOQUOTES ) { |
400
|
|
|
$string = (string) $string; |
401
|
|
|
|
402
|
|
|
if ( 0 === strlen( $string ) ) { |
403
|
|
|
return ''; |
404
|
|
|
} |
405
|
|
|
|
406
|
|
|
// Don't bother if there are no entities - saves a lot of processing |
407
|
|
|
if ( strpos( $string, '&' ) === false ) { |
408
|
|
|
return $string; |
409
|
|
|
} |
410
|
|
|
|
411
|
|
|
// Match the previous behaviour of _wp_specialchars() when the $quote_style is not an accepted value |
412
|
|
View Code Duplication |
if ( empty( $quote_style ) ) { |
|
|
|
|
413
|
|
|
$quote_style = ENT_NOQUOTES; |
414
|
|
|
} elseif ( !in_array( $quote_style, array( 0, 2, 3, 'single', 'double' ), true ) ) { |
415
|
|
|
$quote_style = ENT_QUOTES; |
416
|
|
|
} |
417
|
|
|
|
418
|
|
|
// More complete than get_html_translation_table( HTML_SPECIALCHARS ) |
419
|
|
|
$single = array( ''' => '\'', ''' => '\'' ); |
420
|
|
|
$single_preg = array( '/�*39;/' => ''', '/�*27;/i' => ''' ); |
421
|
|
|
$double = array( '"' => '"', '"' => '"', '"' => '"' ); |
422
|
|
|
$double_preg = array( '/�*34;/' => '"', '/�*22;/i' => '"' ); |
423
|
|
|
$others = array( '<' => '<', '<' => '<', '>' => '>', '>' => '>', '&' => '&', '&' => '&', '&' => '&' ); |
424
|
|
|
$others_preg = array( '/�*60;/' => '<', '/�*62;/' => '>', '/�*38;/' => '&', '/�*26;/i' => '&' ); |
425
|
|
|
|
426
|
|
|
if ( $quote_style === ENT_QUOTES ) { |
427
|
|
|
$translation = array_merge( $single, $double, $others ); |
428
|
|
|
$translation_preg = array_merge( $single_preg, $double_preg, $others_preg ); |
429
|
|
|
} elseif ( $quote_style === ENT_COMPAT || $quote_style === 'double' ) { |
430
|
|
|
$translation = array_merge( $double, $others ); |
431
|
|
|
$translation_preg = array_merge( $double_preg, $others_preg ); |
432
|
|
|
} elseif ( $quote_style === 'single' ) { |
433
|
|
|
$translation = array_merge( $single, $others ); |
434
|
|
|
$translation_preg = array_merge( $single_preg, $others_preg ); |
435
|
|
|
} elseif ( $quote_style === ENT_NOQUOTES ) { |
436
|
|
|
$translation = $others; |
437
|
|
|
$translation_preg = $others_preg; |
438
|
|
|
} |
439
|
|
|
|
440
|
|
|
// Remove zero padding on numeric entities |
441
|
|
|
$string = preg_replace( array_keys( $translation_preg ), array_values( $translation_preg ), $string ); |
|
|
|
|
442
|
|
|
|
443
|
|
|
// Replace characters according to translation table |
444
|
|
|
return strtr( $string, $translation ); |
|
|
|
|
445
|
|
|
} |
446
|
|
|
|
447
|
|
|
|
448
|
|
|
/** |
449
|
|
|
* Escaping for HTML blocks. Stolen from WP |
450
|
|
|
* |
451
|
|
|
* @since 1.6 |
452
|
|
|
* |
453
|
|
|
* @param string $text |
454
|
|
|
* @return string |
455
|
|
|
*/ |
456
|
|
|
function yourls_esc_html( $text ) { |
457
|
|
|
$safe_text = yourls_check_invalid_utf8( $text ); |
458
|
|
|
$safe_text = yourls_specialchars( $safe_text, ENT_QUOTES ); |
459
|
|
|
return yourls_apply_filter( 'esc_html', $safe_text, $text ); |
460
|
|
|
} |
461
|
|
|
|
462
|
|
|
/** |
463
|
|
|
* Escaping for HTML attributes. Stolen from WP |
464
|
|
|
* |
465
|
|
|
* @since 1.6 |
466
|
|
|
* |
467
|
|
|
* @param string $text |
468
|
|
|
* @return string |
469
|
|
|
*/ |
470
|
|
|
function yourls_esc_attr( $text ) { |
471
|
|
|
$safe_text = yourls_check_invalid_utf8( $text ); |
472
|
|
|
$safe_text = yourls_specialchars( $safe_text, ENT_QUOTES ); |
473
|
|
|
return yourls_apply_filter( 'esc_attr', $safe_text, $text ); |
474
|
|
|
} |
475
|
|
|
|
476
|
|
|
/** |
477
|
|
|
* Checks and cleans a URL before printing it. Stolen from WP. |
478
|
|
|
* |
479
|
|
|
* A number of characters are removed from the URL. If the URL is for displaying |
480
|
|
|
* (the default behaviour) ampersands are also replaced. |
481
|
|
|
* |
482
|
|
|
* @since 1.6 |
483
|
|
|
* |
484
|
|
|
* @param string $url The URL to be cleaned. |
485
|
|
|
* @param string $context 'display' or something else. Use yourls_sanitize_url() for database or redirection usage. |
486
|
|
|
* @param array $protocols Optional. Array of allowed protocols, defaults to global $yourls_allowedprotocols |
487
|
|
|
* @return string The cleaned $url |
488
|
|
|
*/ |
489
|
|
|
function yourls_esc_url( $url, $context = 'display', $protocols = array() ) { |
490
|
|
|
// trim first -- see #1931 |
491
|
|
|
$url = trim( $url ); |
492
|
|
|
|
493
|
|
|
// make sure there's only one 'http://' at the beginning (prevents pasting a URL right after the default 'http://') |
494
|
|
|
$url = str_replace( |
495
|
|
|
array( 'http://http://', 'http://https://' ), |
496
|
|
|
array( 'http://', 'https://' ), |
497
|
|
|
$url |
498
|
|
|
); |
499
|
|
|
|
500
|
|
|
if ( '' == $url ) |
501
|
|
|
return $url; |
502
|
|
|
|
503
|
|
|
// make sure there's a protocol, add http:// if not |
504
|
|
|
if ( ! yourls_get_protocol( $url ) ) |
505
|
|
|
$url = 'http://'.$url; |
506
|
|
|
|
507
|
|
|
$original_url = $url; |
508
|
|
|
|
509
|
|
|
// force scheme and domain to lowercase - see issues 591 and 1630 |
510
|
|
|
$url = yourls_lowercase_scheme_domain( $url ); |
511
|
|
|
|
512
|
|
|
$url = preg_replace( '|[^a-z0-9-~+_.?#=!&;,/:%@$\|*\'()\[\]\\x80-\\xff]|i', '', $url ); |
513
|
|
|
// Previous regexp in YOURLS was '|[^a-z0-9-~+_.?\[\]\^#=!&;,/:%@$\|*`\'<>"()\\x80-\\xff\{\}]|i' |
514
|
|
|
// TODO: check if that was it too destructive |
515
|
|
|
$strip = array( '%0d', '%0a', '%0D', '%0A' ); |
516
|
|
|
$url = yourls_deep_replace( $strip, $url ); |
517
|
|
|
$url = str_replace( ';//', '://', $url ); |
518
|
|
|
|
519
|
|
|
// Replace ampersands and single quotes only when displaying. |
520
|
|
|
if ( 'display' == $context ) { |
521
|
|
|
$url = yourls_kses_normalize_entities( $url ); |
522
|
|
|
$url = str_replace( '&', '&', $url ); |
523
|
|
|
$url = str_replace( "'", ''', $url ); |
524
|
|
|
} |
525
|
|
|
|
526
|
|
|
if ( ! is_array( $protocols ) or ! $protocols ) { |
527
|
|
|
global $yourls_allowedprotocols; |
528
|
|
|
$protocols = yourls_apply_filter( 'esc_url_protocols', $yourls_allowedprotocols ); |
529
|
|
|
// Note: $yourls_allowedprotocols is also globally filterable in functions-kses.php/yourls_kses_init() |
530
|
|
|
} |
531
|
|
|
|
532
|
|
|
if ( !yourls_is_allowed_protocol( $url, $protocols ) ) |
533
|
|
|
return ''; |
534
|
|
|
|
535
|
|
|
// I didn't use KSES function kses_bad_protocol() because it doesn't work the way I liked (returns //blah from illegal://blah) |
536
|
|
|
|
537
|
|
|
return yourls_apply_filter( 'esc_url', $url, $original_url, $context ); |
538
|
|
|
} |
539
|
|
|
|
540
|
|
|
|
541
|
|
|
/** |
542
|
|
|
* Lowercase scheme and domain of an URI - see issues 591, 1630, 1889 |
543
|
|
|
* |
544
|
|
|
* This function is trickier than what seems to be needed at first |
545
|
|
|
* |
546
|
|
|
* First, we need to handle several URI types: http://example.com, mailto:[email protected], facetime:[email protected], and so on, see |
547
|
|
|
* yourls_kses_allowed_protocols() in functions-kses.php |
548
|
|
|
* The general rule is that the scheme ("stuff://" or "stuff:") is case insensitive and should be lowercase. But then, depending on the |
549
|
|
|
* scheme, parts of what follows the scheme may or may not be case sensitive. |
550
|
|
|
* |
551
|
|
|
* Second, simply using parse_url() and its opposite http_build_url() (see functions-compat.php) is a pretty unsafe process: |
552
|
|
|
* - parse_url() can easily trip up on malformed or weird URLs |
553
|
|
|
* - exploding a URL with parse_url(), lowercasing some stuff, and glueing things back with http_build_url() does not handle well |
554
|
|
|
* "stuff:"-like URI [1] and can result in URLs ending modified [2][3]. We don't want to *validate* URI, we just want to lowercase |
555
|
|
|
* what is supposed to be lowercased. |
556
|
|
|
* |
557
|
|
|
* So, to be conservative, this functions: |
558
|
|
|
* - lowercases the scheme |
559
|
|
|
* - does not lowercase anything else on "stuff:" URI |
560
|
|
|
* - tries to lowercase only scheme and domain of "stuff://" URI |
561
|
|
|
* |
562
|
|
|
* [1] http_build_url(parse_url("mailto:ozh")) == "mailto:///ozh" |
563
|
|
|
* [2] http_build_url(parse_url("http://blah#omg")) == "http://blah/#omg" |
564
|
|
|
* [3] http_build_url(parse_url("http://blah?#")) == "http://blah/" |
565
|
|
|
* |
566
|
|
|
* @since 1.7.1 |
567
|
|
|
* @param string $url URL |
568
|
|
|
* @return string URL with lowercase scheme and protocol |
569
|
|
|
*/ |
570
|
|
|
function yourls_lowercase_scheme_domain( $url ) { |
571
|
|
|
$scheme = yourls_get_protocol( $url ); |
572
|
|
|
|
573
|
|
|
if( '' == $scheme ) { |
574
|
|
|
// Scheme not found, malformed URL? Something else? Not sure. |
575
|
|
|
return $url; |
576
|
|
|
} |
577
|
|
|
|
578
|
|
|
// Case 1 : scheme like "stuff://" (eg "http://example.com/" or "ssh://[email protected]") |
579
|
|
|
if( substr( $scheme, -2, 2 ) == '//' ) { |
580
|
|
|
|
581
|
|
|
$parts = parse_url( $url ); |
582
|
|
|
|
583
|
|
|
// Most likely malformed stuff, could not parse : we'll just lowercase the scheme and leave the rest untouched |
584
|
|
|
if( false == $parts ) { |
585
|
|
|
$url = str_replace( $scheme, strtolower( $scheme ), $url ); |
586
|
|
|
|
587
|
|
|
// URL seems parsable, let's do the best we can |
588
|
|
|
} else { |
589
|
|
|
|
590
|
|
|
$lower = array(); |
591
|
|
|
|
592
|
|
|
$lower['scheme'] = strtolower( $parts['scheme'] ); |
593
|
|
|
|
594
|
|
|
if( isset( $parts['host'] ) ) { |
595
|
|
|
$lower['host'] = strtolower( $parts['host'] ); |
596
|
|
|
} else { |
597
|
|
|
$parts['host'] = '***'; |
598
|
|
|
} |
599
|
|
|
|
600
|
|
|
// We're not going to glue back things that could be modified in the process |
601
|
|
|
unset( $parts['path'] ); |
602
|
|
|
unset( $parts['query'] ); |
603
|
|
|
unset( $parts['fragment'] ); |
604
|
|
|
|
605
|
|
|
// original beginning of the URL and its lowercase-where-needed counterpart |
606
|
|
|
// We trim the / after the domain to avoid avoid "http://example.com" being reconstructed as "http://example.com/" |
607
|
|
|
$partial_original_url = trim( http_build_url( $parts ), '/' ); |
608
|
|
|
$partial_lower_original_url = trim( http_build_url( $parts, $lower ), '/' ); |
609
|
|
|
|
610
|
|
|
$url = str_replace( $partial_original_url , $partial_lower_original_url, $url ); |
611
|
|
|
|
612
|
|
|
} |
613
|
|
|
|
614
|
|
|
// Case 2 : scheme like "stuff:" (eg "mailto:[email protected]" or "bitcoin:15p1o8vnWqNkJBJGgwafNgR1GCCd6EGtQR?amount=1&label=Ozh") |
615
|
|
|
// In this case, we only lowercase the scheme, because depending on it, things after should or should not be lowercased |
616
|
|
|
} else { |
617
|
|
|
|
618
|
|
|
$url = str_replace( $scheme, strtolower( $scheme ), $url ); |
619
|
|
|
|
620
|
|
|
} |
621
|
|
|
|
622
|
|
|
return $url; |
623
|
|
|
} |
624
|
|
|
|
625
|
|
|
|
626
|
|
|
/** |
627
|
|
|
* Escape single quotes, htmlspecialchar " < > &, and fix line endings. Stolen from WP. |
628
|
|
|
* |
629
|
|
|
* Escapes text strings for echoing in JS. It is intended to be used for inline JS |
630
|
|
|
* (in a tag attribute, for example onclick="..."). Note that the strings have to |
631
|
|
|
* be in single quotes. The filter 'js_escape' is also applied here. |
632
|
|
|
* |
633
|
|
|
* @since 1.6 |
634
|
|
|
* |
635
|
|
|
* @param string $text The text to be escaped. |
636
|
|
|
* @return string Escaped text. |
637
|
|
|
*/ |
638
|
|
|
function yourls_esc_js( $text ) { |
639
|
|
|
$safe_text = yourls_check_invalid_utf8( $text ); |
640
|
|
|
$safe_text = yourls_specialchars( $safe_text, ENT_COMPAT ); |
641
|
|
|
$safe_text = preg_replace( '/&#(x)?0*(?(1)27|39);?/i', "'", stripslashes( $safe_text ) ); |
642
|
|
|
$safe_text = str_replace( "\r", '', $safe_text ); |
643
|
|
|
$safe_text = str_replace( "\n", '\\n', addslashes( $safe_text ) ); |
644
|
|
|
return yourls_apply_filter( 'esc_js', $safe_text, $text ); |
645
|
|
|
} |
646
|
|
|
|
647
|
|
|
/** |
648
|
|
|
* Escaping for textarea values. Stolen from WP. |
649
|
|
|
* |
650
|
|
|
* @since 1.6 |
651
|
|
|
* |
652
|
|
|
* @param string $text |
653
|
|
|
* @return string |
654
|
|
|
*/ |
655
|
|
|
function yourls_esc_textarea( $text ) { |
656
|
|
|
$safe_text = htmlspecialchars( $text, ENT_QUOTES ); |
657
|
|
|
return yourls_apply_filter( 'esc_textarea', $safe_text, $text ); |
658
|
|
|
} |
659
|
|
|
|
660
|
|
|
|
661
|
|
|
/** |
662
|
|
|
* PHP emulation of JS's encodeURI |
663
|
|
|
* |
664
|
|
|
* @link https://developer.mozilla.org/en/JavaScript/Reference/Global_Objects/encodeURI |
665
|
|
|
* @param $url |
666
|
|
|
* @return string |
667
|
|
|
*/ |
668
|
|
|
function yourls_encodeURI( $url ) { |
669
|
|
|
// Decode URL all the way |
670
|
|
|
$result = yourls_rawurldecode_while_encoded( $url ); |
671
|
|
|
// Encode once |
672
|
|
|
$result = strtr( rawurlencode( $result ), array ( |
673
|
|
|
'%3B' => ';', '%2C' => ',', '%2F' => '/', '%3F' => '?', '%3A' => ':', '%40' => '@', |
674
|
|
|
'%26' => '&', '%3D' => '=', '%2B' => '+', '%24' => '$', '%21' => '!', '%2A' => '*', |
675
|
|
|
'%27' => '\'', '%28' => '(', '%29' => ')', '%23' => '#', |
676
|
|
|
) ); |
677
|
|
|
// @TODO: |
678
|
|
|
// Known limit: this will most likely break IDN URLs such as http://www.acad�mie-fran�aise.fr/ |
679
|
|
|
// To fully support IDN URLs, advocate use of a plugin. |
680
|
|
|
return yourls_apply_filter( 'encodeURI', $result, $url ); |
681
|
|
|
} |
682
|
|
|
|
683
|
|
|
/** |
684
|
|
|
* Adds backslashes before letters and before a number at the start of a string. Stolen from WP. |
685
|
|
|
* |
686
|
|
|
* @since 1.6 |
687
|
|
|
* |
688
|
|
|
* @param string $string Value to which backslashes will be added. |
689
|
|
|
* @return string String with backslashes inserted. |
690
|
|
|
*/ |
691
|
|
|
function yourls_backslashit($string) { |
692
|
|
|
$string = preg_replace('/^([0-9])/', '\\\\\\\\\1', $string); |
693
|
|
|
$string = preg_replace('/([a-z])/i', '\\\\\1', $string); |
694
|
|
|
return $string; |
695
|
|
|
} |
696
|
|
|
|
697
|
|
|
/** |
698
|
|
|
* Check if a string seems to be urlencoded |
699
|
|
|
* |
700
|
|
|
* We use rawurlencode instead of urlencode to avoid messing with '+' |
701
|
|
|
* |
702
|
|
|
* @since 1.7 |
703
|
|
|
* @param string $string |
704
|
|
|
* @return bool |
705
|
|
|
*/ |
706
|
|
|
function yourls_is_rawurlencoded( $string ) { |
707
|
|
|
return rawurldecode( $string ) != $string; |
708
|
|
|
} |
709
|
|
|
|
710
|
|
|
/** |
711
|
|
|
* rawurldecode a string till it's not encoded anymore |
712
|
|
|
* |
713
|
|
|
* Deals with multiple encoding (eg "%2521" => "%21" => "!"). |
714
|
|
|
* See https://github.com/YOURLS/YOURLS/issues/1303 |
715
|
|
|
* |
716
|
|
|
* @since 1.7 |
717
|
|
|
* @param string $string |
718
|
|
|
* @return string |
719
|
|
|
*/ |
720
|
|
|
function yourls_rawurldecode_while_encoded( $string ) { |
721
|
|
|
$string = rawurldecode( $string ); |
722
|
|
|
if( yourls_is_rawurlencoded( $string ) ) { |
723
|
|
|
$string = yourls_rawurldecode_while_encoded( $string ); |
724
|
|
|
} |
725
|
|
|
return $string; |
726
|
|
|
} |
727
|
|
|
|
728
|
|
|
/** |
729
|
|
|
* Converts readable Javascript code into a valid bookmarklet link |
730
|
|
|
* |
731
|
|
|
* Uses https://github.com/ozh/bookmarkletgen |
732
|
|
|
* |
733
|
|
|
* @since 1.7.1 |
734
|
|
|
* @param string $code Javascript code |
735
|
|
|
* @return string Bookmarklet link |
736
|
|
|
*/ |
737
|
|
|
function yourls_make_bookmarklet( $code ) { |
738
|
|
|
if ( !class_exists( 'BookmarkletGen', false ) ) { |
739
|
|
|
require_once YOURLS_INC . '/BookmarkletGen/BookmarkletGen.php'; |
740
|
|
|
} |
741
|
|
|
|
742
|
|
|
$book = new BookmarkletGen; |
743
|
|
|
return $book->crunch( $code ); |
744
|
|
|
} |
745
|
|
|
|
This check looks for PHPDoc comments describing methods or function parameters that do not exist on the corresponding method or function.
Consider the following example. The parameter
$italy
is not defined by the methodfinale(...)
.The most likely cause is that the parameter was removed, but the annotation was not.