1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
|
4
|
|
|
namespace Enjoys\AssetsCollector\Content; |
5
|
|
|
|
6
|
|
|
|
7
|
|
|
class UrlConverter |
8
|
|
|
{ |
9
|
|
|
|
10
|
|
|
public function relativeToAbsolute(string $baseUrl, string $relativeUrl) |
11
|
|
|
{ |
12
|
|
|
// If relative URL has a scheme, clean path and return. |
13
|
|
|
$r = $this->split_url($relativeUrl); |
14
|
|
|
if ($r === false) { |
|
|
|
|
15
|
|
|
return false; |
16
|
|
|
} |
17
|
|
|
if (!empty($r['scheme'])) { |
18
|
|
|
if (!empty($r['path']) && $r['path'][0] == '/') { |
19
|
|
|
$r['path'] = $this->url_remove_dot_segments($r['path']); |
20
|
|
|
} |
21
|
|
|
return $this->join_url($r); |
22
|
|
|
} |
23
|
|
|
|
24
|
|
|
// Make sure the base URL is absolute. |
25
|
|
|
$b = $this->split_url($baseUrl); |
26
|
|
|
if ($b === false || empty($b['scheme']) || empty($b['host'])) { |
27
|
|
|
return false; |
28
|
|
|
} |
29
|
|
|
$r['scheme'] = $b['scheme']; |
30
|
|
|
|
31
|
|
|
// If relative URL has an authority, clean path and return. |
32
|
|
|
if (isset($r['host'])) { |
33
|
|
|
if (!empty($r['path'])) { |
34
|
|
|
$r['path'] = $this->url_remove_dot_segments($r['path']); |
35
|
|
|
} |
36
|
|
|
return $this->join_url($r); |
37
|
|
|
} |
38
|
|
|
unset($r['port']); |
39
|
|
|
unset($r['user']); |
40
|
|
|
unset($r['pass']); |
41
|
|
|
|
42
|
|
|
// Copy base authority. |
43
|
|
|
$r['host'] = $b['host']; |
44
|
|
|
if (isset($b['port'])) { |
45
|
|
|
$r['port'] = $b['port']; |
46
|
|
|
} |
47
|
|
|
if (isset($b['user'])) { |
48
|
|
|
$r['user'] = $b['user']; |
49
|
|
|
} |
50
|
|
|
if (isset($b['pass'])) { |
51
|
|
|
$r['pass'] = $b['pass']; |
52
|
|
|
} |
53
|
|
|
|
54
|
|
|
// If relative URL has no path, use base path |
55
|
|
|
if (empty($r['path'])) { |
56
|
|
|
if (!empty($b['path'])) { |
57
|
|
|
$r['path'] = $b['path']; |
58
|
|
|
} |
59
|
|
|
if (!isset($r['query']) && isset($b['query'])) { |
60
|
|
|
$r['query'] = $b['query']; |
61
|
|
|
} |
62
|
|
|
return $this->join_url($r); |
63
|
|
|
} |
64
|
|
|
|
65
|
|
|
// If relative URL path doesn't start with /, merge with base path |
66
|
|
|
if ($r['path'][0] != '/') { |
67
|
|
|
$base = \mb_strrchr($b['path'], '/', true, 'UTF-8'); |
68
|
|
|
if ($base === false) { |
69
|
|
|
$base = ''; |
70
|
|
|
} |
71
|
|
|
$r['path'] = $base . '/' . $r['path']; |
72
|
|
|
} |
73
|
|
|
$r['path'] = $this->url_remove_dot_segments($r['path']); |
74
|
|
|
return $this->join_url($r); |
75
|
|
|
} |
76
|
|
|
|
77
|
|
|
|
78
|
|
|
private function url_remove_dot_segments($path) |
79
|
|
|
{ |
80
|
|
|
// multi-byte character explode |
81
|
|
|
$inSegs = preg_split('!/!u', $path); |
82
|
|
|
$outSegs = array(); |
83
|
|
|
foreach ($inSegs as $seg) { |
84
|
|
|
if ($seg == '' || $seg == '.') { |
85
|
|
|
continue; |
86
|
|
|
} |
87
|
|
|
if ($seg == '..') { |
88
|
|
|
array_pop($outSegs); |
89
|
|
|
} else { |
90
|
|
|
array_push($outSegs, $seg); |
91
|
|
|
} |
92
|
|
|
} |
93
|
|
|
$outPath = implode('/', $outSegs); |
94
|
|
|
if ($path[0] == '/') { |
95
|
|
|
$outPath = '/' . $outPath; |
96
|
|
|
} |
97
|
|
|
// compare last multi-byte character against '/' |
98
|
|
|
if ($outPath != '/' && |
99
|
|
|
(\mb_strlen($path) - 1) == \mb_strrpos($path, '/')) { |
100
|
|
|
$outPath .= '/'; |
101
|
|
|
} |
102
|
|
|
return $outPath; |
103
|
|
|
} |
104
|
|
|
|
105
|
|
|
private function join_url($parts, $encode = true) |
106
|
|
|
{ |
107
|
|
|
if ($encode) { |
108
|
|
|
if (isset($parts['user'])) { |
109
|
|
|
$parts['user'] = rawurlencode($parts['user']); |
110
|
|
|
} |
111
|
|
|
if (isset($parts['pass'])) { |
112
|
|
|
$parts['pass'] = rawurlencode($parts['pass']); |
113
|
|
|
} |
114
|
|
|
if (isset($parts['host']) && |
115
|
|
|
!preg_match('!^(\[[\da-f.:]+\]])|([\da-f.:]+)$!ui', $parts['host'])) { |
116
|
|
|
$parts['host'] = rawurlencode($parts['host']); |
117
|
|
|
} |
118
|
|
|
if (!empty($parts['path'])) { |
119
|
|
|
$parts['path'] = preg_replace( |
120
|
|
|
'!%2F!ui', |
121
|
|
|
'/', |
122
|
|
|
rawurlencode($parts['path']) |
123
|
|
|
); |
124
|
|
|
} |
125
|
|
|
if (isset($parts['query'])) { |
126
|
|
|
$parts['query'] = rawurlencode($parts['query']); |
127
|
|
|
} |
128
|
|
|
if (isset($parts['fragment'])) { |
129
|
|
|
$parts['fragment'] = rawurlencode($parts['fragment']); |
130
|
|
|
} |
131
|
|
|
} |
132
|
|
|
|
133
|
|
|
$url = ''; |
134
|
|
|
if (!empty($parts['scheme'])) { |
135
|
|
|
$url .= $parts['scheme'] . ':'; |
136
|
|
|
} |
137
|
|
|
if (isset($parts['host'])) { |
138
|
|
|
$url .= '//'; |
139
|
|
|
if (isset($parts['user'])) { |
140
|
|
|
$url .= $parts['user']; |
141
|
|
|
if (isset($parts['pass'])) { |
142
|
|
|
$url .= ':' . $parts['pass']; |
143
|
|
|
} |
144
|
|
|
$url .= '@'; |
145
|
|
|
} |
146
|
|
|
if (preg_match('!^[\da-f]*:[\da-f.:]+$!ui', $parts['host'])) { |
147
|
|
|
$url .= '[' . $parts['host'] . ']'; |
148
|
|
|
} // IPv6 |
149
|
|
|
else { |
150
|
|
|
$url .= $parts['host']; |
151
|
|
|
} // IPv4 or name |
152
|
|
|
if (isset($parts['port'])) { |
153
|
|
|
$url .= ':' . $parts['port']; |
154
|
|
|
} |
155
|
|
|
if (!empty($parts['path']) && $parts['path'][0] != '/') { |
156
|
|
|
$url .= '/'; |
157
|
|
|
} |
158
|
|
|
} |
159
|
|
|
if (!empty($parts['path'])) { |
160
|
|
|
$url .= $parts['path']; |
161
|
|
|
} |
162
|
|
|
if (isset($parts['query'])) { |
163
|
|
|
$url .= '?' . $parts['query']; |
164
|
|
|
} |
165
|
|
|
if (isset($parts['fragment'])) { |
166
|
|
|
$url .= '#' . $parts['fragment']; |
167
|
|
|
} |
168
|
|
|
return $url; |
169
|
|
|
} |
170
|
|
|
|
171
|
|
|
private function split_url($url, $decode = true) |
172
|
|
|
{ |
173
|
|
|
// Character sets from RFC3986. |
174
|
|
|
$xunressub = 'a-zA-Z\d\-._~\!$&\'()*+,;='; |
175
|
|
|
$xpchar = $xunressub . ':@%'; |
176
|
|
|
|
177
|
|
|
// Scheme from RFC3986. |
178
|
|
|
$xscheme = '([a-zA-Z][a-zA-Z\d+\-.]*)'; |
179
|
|
|
|
180
|
|
|
// User info (user + password) from RFC3986. |
181
|
|
|
$xuserinfo = '(([' . $xunressub . '%]*)' . |
182
|
|
|
'(:([' . $xunressub . ':%]*))?)'; |
183
|
|
|
|
184
|
|
|
// IPv4 from RFC3986 (without digit constraints). |
185
|
|
|
$xipv4 = '(\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3})'; |
186
|
|
|
|
187
|
|
|
// IPv6 from RFC2732 (without digit and grouping constraints). |
188
|
|
|
$xipv6 = '(\[([a-fA-F\d.:]+)\])'; |
189
|
|
|
|
190
|
|
|
// Host name from RFC1035. Technically, must start with a letter. |
191
|
|
|
// Relax that restriction to better parse URL structure, then |
192
|
|
|
// leave host name validation to application. |
193
|
|
|
$xhost_name = '([a-zA-Z\d\-.%]+)'; |
194
|
|
|
|
195
|
|
|
// Authority from RFC3986. Skip IP future. |
196
|
|
|
$xhost = '(' . $xhost_name . '|' . $xipv4 . '|' . $xipv6 . ')'; |
197
|
|
|
$xport = '(\d*)'; |
198
|
|
|
$xauthority = '((' . $xuserinfo . '@)?' . $xhost . |
199
|
|
|
'?(:' . $xport . ')?)'; |
200
|
|
|
|
201
|
|
|
// Path from RFC3986. Blend absolute & relative for efficiency. |
202
|
|
|
$xslash_seg = '(/[' . $xpchar . ']*)'; |
203
|
|
|
$xpath_authabs = '((//' . $xauthority . ')((/[' . $xpchar . ']*)*))'; |
204
|
|
|
$xpath_rel = '([' . $xpchar . ']+' . $xslash_seg . '*)'; |
205
|
|
|
$xpath_abs = '(/(' . $xpath_rel . ')?)'; |
206
|
|
|
$xapath = '(' . $xpath_authabs . '|' . $xpath_abs . |
207
|
|
|
'|' . $xpath_rel . ')'; |
208
|
|
|
|
209
|
|
|
// Query and fragment from RFC3986. |
210
|
|
|
$xqueryfrag = '([' . $xpchar . '/?' . ']*)'; |
211
|
|
|
|
212
|
|
|
// URL. |
213
|
|
|
$xurl = '^(' . $xscheme . ':)?' . $xapath . '?' . |
214
|
|
|
'(\?' . $xqueryfrag . ')?(#' . $xqueryfrag . ')?$'; |
215
|
|
|
|
216
|
|
|
|
217
|
|
|
// Split the URL into components. |
218
|
|
|
if (!preg_match('!' . $xurl . '!', $url, $m)) { |
219
|
|
|
return false; |
220
|
|
|
} |
221
|
|
|
|
222
|
|
|
if (!empty($m[2])) { |
223
|
|
|
$parts['scheme'] = strtolower($m[2]); |
|
|
|
|
224
|
|
|
} |
225
|
|
|
|
226
|
|
|
if (!empty($m[7])) { |
227
|
|
|
if (isset($m[9])) { |
228
|
|
|
$parts['user'] = $m[9]; |
229
|
|
|
} else { |
230
|
|
|
$parts['user'] = ''; |
231
|
|
|
} |
232
|
|
|
} |
233
|
|
|
if (!empty($m[10])) { |
234
|
|
|
$parts['pass'] = $m[11]; |
235
|
|
|
} |
236
|
|
|
|
237
|
|
|
if (!empty($m[13])) { |
238
|
|
|
$h = $parts['host'] = $m[13]; |
239
|
|
|
} else { |
240
|
|
|
if (!empty($m[14])) { |
241
|
|
|
$parts['host'] = $m[14]; |
242
|
|
|
} else { |
243
|
|
|
if (!empty($m[16])) { |
244
|
|
|
$parts['host'] = $m[16]; |
245
|
|
|
} else { |
246
|
|
|
if (!empty($m[5])) { |
247
|
|
|
$parts['host'] = ''; |
248
|
|
|
} |
249
|
|
|
} |
250
|
|
|
} |
251
|
|
|
} |
252
|
|
|
if (!empty($m[17])) { |
253
|
|
|
$parts['port'] = $m[18]; |
254
|
|
|
} |
255
|
|
|
|
256
|
|
|
if (!empty($m[19])) { |
257
|
|
|
$parts['path'] = $m[19]; |
258
|
|
|
} else { |
259
|
|
|
if (!empty($m[21])) { |
260
|
|
|
$parts['path'] = $m[21]; |
261
|
|
|
} else { |
262
|
|
|
if (!empty($m[25])) { |
263
|
|
|
$parts['path'] = $m[25]; |
264
|
|
|
} |
265
|
|
|
} |
266
|
|
|
} |
267
|
|
|
|
268
|
|
|
if (!empty($m[27])) { |
269
|
|
|
$parts['query'] = $m[28]; |
270
|
|
|
} |
271
|
|
|
if (!empty($m[29])) { |
272
|
|
|
$parts['fragment'] = $m[30]; |
273
|
|
|
} |
274
|
|
|
|
275
|
|
|
if (!$decode) { |
276
|
|
|
return $parts; |
|
|
|
|
277
|
|
|
} |
278
|
|
|
if (!empty($parts['user'])) { |
279
|
|
|
$parts['user'] = rawurldecode($parts['user']); |
280
|
|
|
} |
281
|
|
|
if (!empty($parts['pass'])) { |
282
|
|
|
$parts['pass'] = rawurldecode($parts['pass']); |
283
|
|
|
} |
284
|
|
|
if (!empty($parts['path'])) { |
285
|
|
|
$parts['path'] = rawurldecode($parts['path']); |
286
|
|
|
} |
287
|
|
|
if (isset($h)) { |
288
|
|
|
$parts['host'] = rawurldecode($parts['host']); |
289
|
|
|
} |
290
|
|
|
if (!empty($parts['query'])) { |
291
|
|
|
$parts['query'] = rawurldecode($parts['query']); |
292
|
|
|
} |
293
|
|
|
if (!empty($parts['fragment'])) { |
294
|
|
|
$parts['fragment'] = rawurldecode($parts['fragment']); |
295
|
|
|
} |
296
|
|
|
return $parts; |
297
|
|
|
} |
298
|
|
|
} |