From ff8e5c090a534b4cf658804ca3fa36229f5f8876 Mon Sep 17 00:00:00 2001
From: "stephane.chazelas" <stephane.chazelas@emerson.com>
Date: Fri, 4 Jan 2008 15:53:01 +0100
Subject: [PATCH] parenthesis_escaper_fix
darcs-hash:20080104145301-3d17c-6ba186de4a2661f7d4f6b080e4cf74fed9ae1904.gz
---
inc/parser/lexer.php | 99 ++++++++++++++++++++------------------------
1 file changed, 44 insertions(+), 55 deletions(-)
diff --git a/inc/parser/lexer.php b/inc/parser/lexer.php
index 0887dde93..88deb7fe9 100644
--- a/inc/parser/lexer.php
+++ b/inc/parser/lexer.php
@@ -138,60 +138,49 @@ class Doku_LexerParallelRegex {
$cnt = count($this->_patterns);
for ($i = 0; $i < $cnt; $i++) {
- // Replace lookaheads / lookbehinds with marker
- $m = "\1\1";
- $pattern = preg_replace(
- array (
- '/\(\?(i|m|s|x|U)\)/U',
- '/\(\?(\-[i|m|s|x|U])\)/U',
- '/\(\?\=(.*)\)/sU',
- '/\(\?\!(.*)\)/sU',
- '/\(\?\<\=(.*)\)/sU',
- '/\(\?\<\!(.*)\)/sU',
- '/\(\?\:(.*)\)/sU',
- ),
- array (
- $m.'SO:\\1'.$m,
- $m.'SOR:\\1'.$m,
- $m.'LA:IS:\\1'.$m,
- $m.'LA:NOT:\\1'.$m,
- $m.'LB:IS:\\1'.$m,
- $m.'LB:NOT:\\1'.$m,
- $m.'GRP:\\1'.$m,
- ),
- $this->_patterns[$i]
- );
- // Quote the rest
- $pattern = str_replace(
- array('/', '(', ')'),
- array('\/', '\(', '\)'),
- $pattern
- );
-
- // Restore lookaheads / lookbehinds
- $pattern = preg_replace(
- array (
- '/'.$m.'SO:(.{1})'.$m.'/',
- '/'.$m.'SOR:(.{2})'.$m.'/',
- '/'.$m.'LA:IS:(.*)'.$m.'/sU',
- '/'.$m.'LA:NOT:(.*)'.$m.'/sU',
- '/'.$m.'LB:IS:(.*)'.$m.'/sU',
- '/'.$m.'LB:NOT:(.*)'.$m.'/sU',
- '/'.$m.'GRP:(.*)'.$m.'/sU',
- ),
- array (
- '(?\\1)',
- '(?\\1)',
- '(?=\\1)',
- '(?!\\1)',
- '(?<=\\1)',
- '(?<!\\1)',
- '(?:\\1)',
- ),
- $pattern
- );
-
- $this->_patterns[$i] = '('.$pattern.')';
+ /*
+ * decompose the input pattern into "(", "(?", ")",
+ * "[...]", "[]..]", "[^]..]", "[...[:...:]..]", "\x"...
+ * elements.
+ */
+ preg_match_all('/\\\\.|' .
+ '\(\?|' .
+ '[()]|' .
+ '\[\^?\]?(?:\\\\.|\[:[^]]*:\]|[^]\\\\])*\]|' .
+ '[^[()\\\\]+/', $this->_patterns[$i], $elts);
+
+ $pattern = "";
+ $level = 0;
+
+ foreach ($elts[0] as $elt) {
+ /*
+ * for "(", ")" remember the nesting level, add "\"
+ * only to the non-"(?" ones.
+ */
+
+ switch($elt) {
+ case '(':
+ $pattern .= '\(';
+ break;
+ case ')':
+ if ($level > 0)
+ $level--; /* closing (? */
+ else
+ $pattern .= '\\';
+ $pattern .= ')';
+ break;
+ case '(?':
+ $level++;
+ $pattern .= '(?';
+ break;
+ default:
+ if (substr($elt, 0, 1) == '\\')
+ $pattern .= $elt;
+ else
+ $pattern .= str_replace('/', '\/', $elt);
+ }
+ }
+ $this->_patterns[$i] = "($pattern)";
}
$this->_regex = "/" . implode("|", $this->_patterns) . "/" . $this->_getPerlMatchingFlags();
}
@@ -591,4 +580,4 @@ function Doku_Lexer_Escape($str) {
return preg_replace($chars, $escaped, $str);
}
-//Setup VIM: ex: et ts=4 enc=utf-8 :
+//Setup VIM: ex: et ts=4 sw=4 enc=utf-8 :
--
GitLab