Last active
May 24, 2016 07:46
-
-
Save musen/f7d357df202e38c235b32f52b7131ff2 to your computer and use it in GitHub Desktop.
Extracts and returns table of contents (toc) and bookmarked content from any html.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
<?php | |
function create_toc( $content ) { | |
preg_match_all( '/<h([1-6])(.*)>([^<]+)<\/h[1-6]>/i', $content, $matches, PREG_SET_ORDER ); | |
global $anchors; | |
$anchors = array(); | |
$toc = '<ol class="toc">'."\n"; | |
$i = 0; | |
foreach ( $matches as $heading ) { | |
if ($i == 0) | |
$startlvl = $heading[1]; | |
$lvl = $heading[1]; | |
$ret = preg_match( '/id=[\'|"](.*)?[\'|"]/i', stripslashes($heading[2]), $anchor ); | |
if ( $ret && $anchor[1] != '' ) { | |
$anchor = stripslashes( $anchor[1] ); | |
$add_id = false; | |
} else { | |
$anchor = preg_replace( '/\s+/', '-', preg_replace('/[^a-z\s]/', '', strtolower( $heading[3] ) ) ); | |
$add_id = true; | |
} | |
if ( !in_array( $anchor, $anchors ) ) { | |
$anchors[] = $anchor; | |
} else { | |
$orig_anchor = $anchor; | |
$i = 2; | |
while ( in_array( $anchor, $anchors ) ) { | |
$anchor = $orig_anchor.'-'.$i; | |
$i++; | |
} | |
$anchors[] = $anchor; | |
} | |
if ( $add_id ) { | |
$content = substr_replace( $content, '<h'.$lvl.' id="'.$anchor.'"'.$heading[2].'>'.$heading[3].'</h'.$lvl.'>', strpos( $content, $heading[0] ), strlen( $heading[0] ) ); | |
} | |
$ret = preg_match( '/title=[\'|"](.*)?[\'|"]/i', stripslashes( $heading[2] ), $title ); | |
if ( $ret && $title[1] != '' ) | |
$title = stripslashes( $title[1] ); | |
else | |
$title = $heading[3]; | |
$title = trim( strip_tags( $title ) ); | |
if ($i > 0) { | |
if ($prevlvl < $lvl) { | |
$toc .= "\n"."<ol>"."\n"; | |
} else if ($prevlvl > $lvl) { | |
$toc .= '</li>'."\n"; | |
while ($prevlvl > $lvl) { | |
$toc .= "</ol>"."\n".'</li>'."\n"; | |
$prevlvl--; | |
} | |
} else { | |
$toc .= '</li>'."\n"; | |
} | |
} | |
$j = 0; | |
$toc .= '<li><a href="#'.$anchor.'">'.$title.'</a>'; | |
$prevlvl = $lvl; | |
$i++; | |
} | |
unset( $anchors ); | |
while ( $lvl > $startlvl ) { | |
$toc .= "\n</ol>"; | |
$lvl--; | |
} | |
$toc .= '</li>'."\n"; | |
$toc .= '</ol>'."\n"; | |
return array( | |
'toc' => $toc, | |
'content' => $content | |
); | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment