X-Git-Url: https://vcs.fsf.org/?p=squirrelmail.git;a=blobdiff_plain;f=functions%2Furl_parser.php;h=c9ee48627853ec8ef5a68a920e200da3f6ff45db;hp=bcb5f7b91d113677d7ab93a6e7f086f127fe313a;hb=7350889b45590aee831ba1150ee94efb95b03992;hpb=3806fa52f70c5cb6f923957f7a1c2e0e867d0fdf diff --git a/functions/url_parser.php b/functions/url_parser.php index bcb5f7b9..c9ee4862 100644 --- a/functions/url_parser.php +++ b/functions/url_parser.php @@ -1,54 +1,154 @@ \\0", $body); - return $body; - } - - function parseUrl ($body) { - #Possible ways a URL could finish. - - $poss_ends=array(" ", "\n", "\r", "<", ">", ".\r", ".\n", ". ", " ", ")", "("); - $done=False; - while (!$done) { - #Look for when a URL starts - $where = strpos($body,"http:",$start); - if ($where) { - # Find the end of that URL - reset($poss_ends); $end=0; - while (list($key, $val) = each($poss_ends)) { - $enda = strpos($body,$val,$where); - if ($end == 0) $end = $enda; - if ($enda < $end and $enda != 0) $end = $enda; - } - #Extract URL - $url = substr($body,$where,$end-$where); - #Replace URL with HyperLinked Url - if ($url != "") { - $url_str = "$url"; - # $body = str_replace($url,$url_str,$body); - $body = replaceBlock($body,$url_str,$where,$end); - $start = strpos($body,"",$where); - } else { - $start = $where + 7; - } - } else { - $done=true; + // Having this defined in just one spot could help when changes need + // to be made to the pattern + // Make sure that the expression is evaluated case insensitively + // + // Here's pretty sophisticated IP matching: + // $IPMatch = '(2[0-5][0-9]|1?[0-9]{1,2})'; + // $IPMatch = '\[?' . $IPMatch . '(\.' . $IPMatch . '){3}\]?'; + // + // Here's enough: + global $IP_RegExp_Match, $Host_RegExp_Match, $Email_RegExp_Match; + $IP_RegExp_Match = '\\[?[0-9]{1,3}(\\.[0-9]{1,3}){3}\\]?'; + $Host_RegExp_Match = '(' . $IP_RegExp_Match . + '|[0-9a-z]([-.]?[0-9a-z])*\\.[a-z][a-z]+)'; + $Email_RegExp_Match = '[0-9a-z]([-_.]?[0-9a-z])*(%' . $Host_RegExp_Match . + ')?@' . $Host_RegExp_Match; + + function parseEmail (&$body) { + global $color, $Email_RegExp_Match; + $Size = strlen($body); + + /* + This is here in case we ever decide to use highlighting of searched + text. this does it for email addresses + + if ($what && ($where == "BODY" || $where == "TEXT")) { + eregi ($Email_RegExp_Match, $body, $regs); + $oldaddr = $regs[0]; + if ($oldaddr) { + $newaddr = eregi_replace ($what, "$what", $oldaddr); + $body = str_replace ($oldaddr, "$newaddr", $body); } + } else { + $body = eregi_replace ($Email_RegExp_Match, "\\0", $body); } - - return $body; + */ + + $body = eregi_replace ($Email_RegExp_Match, "\\0", $body); + + // If there are any changes, it'll just get bigger. + if ($Size != strlen($body)) + return 1; + return 0; } -?> + // We don't want to re-initialize this stuff for every line. Save work + // and just do it once here. + global $url_parser_url_tokens; + $url_parser_url_tokens = array( + 'http://', + 'https://', + 'ftp://', + 'telnet:', // Special case -- doesn't need the slashes + 'gopher://', + 'news://'); + + global $url_parser_poss_ends; + $url_parser_poss_ends = array(' ', "\n", "\r", '<', '>', ".\r", ".\n", + '. ', ' ', ')', '(', '"', '<', '>', '.<', + ']', '[', '{', '}', "\240", ', ', '. ', ",\n", ",\r"); + + + function parseUrl (&$body) + { + global $url_parser_poss_ends, $url_parser_url_tokens;; + $start = 0; + $target_pos = strlen($body); + + while ($start != $target_pos) + { + $target_token = ''; + + // Find the first token to replace + foreach ($url_parser_url_tokens as $the_token) + { + $pos = strpos(strtolower($body), $the_token, $start); + if (is_int($pos) && $pos < $target_pos) + { + $target_pos = $pos; + $target_token = $the_token; + } + } + + // Look for email addresses between $start and $target_pos + $check_str = substr($body, $start, $target_pos); + + if (parseEmail($check_str)) + { + replaceBlock($body, $check_str, $start, $target_pos); + $target_pos = strlen($check_str) + $start; + } + + // If there was a token to replace, replace it + if ($target_token != '') + { + // Find the end of the URL + $end=strlen($body); + foreach ($url_parser_poss_ends as $key => $val) + { + $enda = strpos($body,$val,$target_pos); + if (is_int($enda) && $enda < $end) + $end = $enda; + } + + // Extract URL + $url = substr($body, $target_pos, $end-$target_pos); + + // Needed since lines are not passed with \n or \r + while ( ereg("[,\.]$", $url) ) { + $url = substr( $url, 0, -1 ); + $end--; + } + + // Replace URL with HyperLinked Url, requires 1 char in link + if ($url != '' && $url != $target_token) + { + $url_str = "$url"; + replaceBlock($body,$url_str,$target_pos,$end); + $target_pos += strlen($url_str); + } + else + { + // Not quite a valid link, skip ahead to next chance + $target_pos += strlen($target_token); + } + } + + // Move forward + $start = $target_pos; + $target_pos = strlen($body); + } + } + +?>