59177427 |
1 | <?php |
43fcef5c |
2 | |
35586184 |
3 | /** |
4 | * url_parser.php |
5 | * |
15e6162e |
6 | * Copyright (c) 1999-2002 The SquirrelMail Project Team |
35586184 |
7 | * Licensed under the GNU GPL. For full terms see the file COPYING. |
8 | * |
9 | * This code provides various string manipulation functions that are |
10 | * used by the rest of the Squirrelmail code. |
11 | * |
12 | * $Id$ |
13 | */ |
43fcef5c |
14 | |
35586184 |
15 | function replaceBlock (&$in, $replace, $start, $end) { |
16 | $begin = substr($in,0,$start); |
17 | $end = substr($in,$end,strlen($in)-$end); |
18 | $in = $begin.$replace.$end; |
19 | } |
43fcef5c |
20 | |
01d27858 |
21 | /* Having this defined in just one spot could help when changes need |
22 | * to be made to the pattern |
23 | * Make sure that the expression is evaluated case insensitively |
24 | * |
25 | * Here's pretty sophisticated IP matching: |
26 | * $IPMatch = '(2[0-5][0-9]|1?[0-9]{1,2})'; |
27 | * $IPMatch = '\[?' . $IPMatch . '(\.' . $IPMatch . '){3}\]?'; |
28 | */ |
29 | /* Here's enough: */ |
30 | global $IP_RegExp_Match, $Host_RegExp_Match, $Email_RegExp_Match; |
31 | $IP_RegExp_Match = '\\[?[0-9]{1,3}(\\.[0-9]{1,3}){3}\\]?'; |
32 | $Host_RegExp_Match = '(' . $IP_RegExp_Match . |
33 | '|[0-9a-z]([-.]?[0-9a-z])*\\.[a-z][a-z]+)'; |
34 | $Email_RegExp_Match = '[0-9a-z]([-_.+]?[0-9a-z])*(%' . $Host_RegExp_Match . |
35 | ')?@' . $Host_RegExp_Match; |
8f7163e7 |
36 | |
01d27858 |
37 | function parseEmail (&$body) { |
38 | global $color, $Email_RegExp_Match; |
39 | $Size = strlen($body); |
9eea179c |
40 | |
01d27858 |
41 | /* |
42 | * This is here in case we ever decide to use highlighting of searched |
43 | * text. this does it for email addresses |
44 | * |
45 | * if ($what && ($where == "BODY" || $where == "TEXT")) { |
46 | * eregi ($Email_RegExp_Match, $body, $regs); |
47 | * $oldaddr = $regs[0]; |
48 | * if ($oldaddr) { |
49 | * $newaddr = eregi_replace ($what, "<b><font color=\"$color[2]\">$what</font></font></b>", $oldaddr); |
50 | * $body = str_replace ($oldaddr, "<a href=\"../src/compose.php?send_to=$oldaddr\">$newaddr</a>", $body); |
51 | * } |
52 | * } else { |
53 | * $body = eregi_replace ($Email_RegExp_Match, "<a href=\"../src/compose.php?send_to=\\0\">\\0</a>", $body); |
54 | * } |
55 | */ |
56 | |
57 | if( eregi($Email_RegExp_Match, $body, $regs) ) { |
58 | $body = str_replace($regs[0], '<a href="../src/compose.php?send_to='. |
59 | urlencode($regs[0]).'">'.$regs[0].'</a>', $body); |
60 | } |
61 | |
62 | /* If there are any changes, it'll just get bigger. */ |
63 | if ($Size != strlen($body)) { |
64 | return 1; |
65 | } |
66 | return 0; |
67 | } |
43fcef5c |
68 | |
43fcef5c |
69 | |
01d27858 |
70 | /* We don't want to re-initialize this stuff for every line. Save work |
71 | * and just do it once here. |
72 | */ |
73 | global $url_parser_url_tokens; |
74 | $url_parser_url_tokens = array( |
75 | 'http://', |
76 | 'https://', |
77 | 'ftp://', |
78 | 'telnet:', // Special case -- doesn't need the slashes |
79 | 'gopher://', |
80 | 'news://'); |
8f7163e7 |
81 | |
01d27858 |
82 | global $url_parser_poss_ends; |
83 | $url_parser_poss_ends = array(' ', "\n", "\r", '<', '>', ".\r", ".\n", |
84 | '. ', ' ', ')', '(', '"', '<', '>', '.<', |
85 | ']', '[', '{', '}', "\240", ', ', '. ', ",\n", ",\r"); |
8f7163e7 |
86 | |
20a60f89 |
87 | |
01d27858 |
88 | function parseUrl (&$body) { |
89 | global $url_parser_poss_ends, $url_parser_url_tokens;; |
90 | $start = 0; |
91 | $target_pos = strlen($body); |
8f7163e7 |
92 | |
01d27858 |
93 | while ($start != $target_pos) { |
8f7163e7 |
94 | $target_token = ''; |
95 | |
01d27858 |
96 | /* Find the first token to replace */ |
97 | foreach ($url_parser_url_tokens as $the_token) { |
98 | $pos = strpos(strtolower($body), $the_token, $start); |
99 | if (is_int($pos) && $pos < $target_pos) { |
100 | $target_pos = $pos; |
101 | $target_token = $the_token; |
102 | } |
8f7163e7 |
103 | } |
104 | |
01d27858 |
105 | /* Look for email addresses between $start and $target_pos */ |
8f7163e7 |
106 | $check_str = substr($body, $start, $target_pos); |
8f7163e7 |
107 | |
01d27858 |
108 | if (parseEmail($check_str)) { |
109 | replaceBlock($body, $check_str, $start, $target_pos); |
110 | $target_pos = strlen($check_str) + $start; |
8f7163e7 |
111 | } |
e2ef6f4b |
112 | |
01d27858 |
113 | /* If there was a token to replace, replace it */ |
114 | if ($target_token != '') { |
115 | /* Find the end of the URL */ |
116 | $end=strlen($body); |
117 | foreach ($url_parser_poss_ends as $key => $val) { |
118 | $enda = strpos($body,$val,$target_pos); |
119 | if (is_int($enda) && $enda < $end) { |
120 | $end = $enda; |
121 | } |
122 | } |
123 | |
124 | /* Extract URL */ |
125 | $url = substr($body, $target_pos, $end-$target_pos); |
8f7163e7 |
126 | |
01d27858 |
127 | /* Needed since lines are not passed with \n or \r */ |
128 | while ( ereg("[,\.]$", $url) ) { |
129 | $url = substr( $url, 0, -1 ); |
130 | $end--; |
131 | } |
1b3324b3 |
132 | |
01d27858 |
133 | /* Replace URL with HyperLinked Url, requires 1 char in link */ |
134 | if ($url != '' && $url != $target_token) { |
135 | $url_str = "<a href=\"$url\" target=\"_blank\">$url</a>"; |
136 | replaceBlock($body,$url_str,$target_pos,$end); |
137 | $target_pos += strlen($url_str); |
138 | } |
139 | else { |
140 | // Not quite a valid link, skip ahead to next chance |
141 | $target_pos += strlen($target_token); |
142 | } |
8f7163e7 |
143 | } |
144 | |
01d27858 |
145 | /* Move forward */ |
8f7163e7 |
146 | $start = $target_pos; |
147 | $target_pos = strlen($body); |
01d27858 |
148 | } |
149 | } |
43fcef5c |
150 | ?> |