Oo-regexp-export.php

From Organic Design wiki
Revision as of 09:29, 10 August 2005 by Nad (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

<?

  1. Export article to OO format
  2. - only does document.php transform and its own wiki markup

# Main wiki->oo parser function ooParse($text) { $script = $GLOBALS['xwScript']; $bToken = ':-:'.'bullet'.':-:';

# Convert problem quotes and dashes $text = str_replace('–','-',$text); $text = preg_replace('/“|�/','"',$text);

# TODO ; :

# *text (pre) $text = preg_replace("/\\n(\\*+) *(.+?)$/m", "$bToken\\1\\2$bToken", $text);

$xml = ; foreach (explode("\n\n", trim(htmlentities($text))) as $text) {

# text $text = preg_replace("/(.+?)/s", "<text:span text:style-name=\"T2\">\\1</text:span>", $text);

# text $text = preg_replace("/(.+?)/s", "<text:span text:style-name=\"T1\">\\1</text:span>", $text);

# anchor $text = preg_replace("/\\[\\[Wikipedia:([^\\|\\]]+)\\|([^\\]]*)]]/", "<text:a xlink:type=\"simple\" xlink:href=\"http://en.wikipedia.org/wiki/\\1\">\\2</text:a>", $text);

# anchor $text = preg_replace("/\\[\\[([^\\|\\]]+)\\|([^\\]]*)]]/", "<text:a xlink:type=\"simple\" xlink:href=\"$script/\\1\">\\2</text:a>", $text);

# article $text = preg_replace("/\\[\\[([^\\]]+)]]/", "<text:a xlink:type=\"simple\" xlink:href=\"$script/\\1\">\\1</text:a>", $text);

# [url anchor] $text = preg_replace("/\\[([^ \\]]+) +([^\\]]*)]/", "<text:a xlink:type=\"simple\" xlink:href=\"\\1\">\\2</text:a>", $text);

# [url] $text = preg_replace("/\\[([^\\]]+)]/", "<text:a xlink:type=\"simple\" xlink:href=\"\\1\">\\1</text:a>", $text);

$xml .= "<text:p text:style-name=\"Text body\">$text</text:p>\n"; }

# *text (post) $xml= preg_replace("/$bToken\\*+(.*?)$bToken/m", "<text:list-item><text:p text:style-name=\"P12\">\\1</text:p></text:list-item>", $xml); $xml= preg_replace("/(?<!<\\/text:list-item>)(<text:list-item>)/", "</text:p>\n<text:unordered-list text:style-name=\"L1\">\\1", $xml); $xml= preg_replace("/(<\\/text:list-item>)(?!<text:list-item>)/", "\\1</text:unordered-list>\n<text:p text:style-name=\"Text body\">", $xml);

#


$xml = str_replace('<hr></text:p>', "</text:p>\n<text:p text:style-name=\"P13\" />\n", $xml); $xml = str_replace('<hr>', "</text:p>\n<text:p text:style-name=\"P13\" />\n<text:p text:style-name=\"Text body\">", $xml);

#   $xml = str_replace('&nbsp;', ' ', $xml);

return $xml; }

$htdocs = $GLOBALS['_SERVER']['DOCUMENT_ROOT']; $error = false;

  1. Create a unique working folder and copy OO doc file-template to new file in it

mkdir($tmp = "$htdocs/wiki/tmp/".uniqid('export')); copy("$htdocs/wiki/xmlwiki/doc-template.sxw", $sxw = "$tmp/sxw");

  1. Zip compression library - www.phpconcept.net/pclzip/

define('PCLZIP_TEMPORARY_DIR', "$tmp/"); require_once("$htdocs/wiki/xmlwiki/lib/pclzip/pclzip.lib.php");

  1. Get OO content template article

$xml = xwArticleContent('oo-doc-template.xml');

  1. Insert our document info into existing tokens

$xml = str_replace('Title', str_replace('_',' ',$title), $xml); $xml = str_replace('Version', 'Exported by '.$GLOBALS['xwUserName'].' on '.date('d M Y').' at '.date('H:i:s'), $xml);

  1. $xml = str_replace('My-Header', "$script/$title", $xml); # this is in styles
  1. Build mian document body
  2. - the match splits the article into sections at headings, extracting heading, level & text

$body = ; $last = 0;

if (preg_match_all('/

<tmp num="(#*)" title="(.*?)" anchor="(.*?)".><.h1>/s', $article, $sections, PREG_SET_ORDER | PREG_OFFSET_CAPTURE)) { foreach ($sections as $m) { $level = strlen($m[1][0]); $anchor = htmlentities($m[3][0]); if ($i = $m[0][1] - $last) $body .= ooParse(substr($article, $last, $i)); $style = $level > 1 ? "Heading $level" : 'P11'; if ($anchor) $body .= "<text:h text:style-name=\"$style\" text:level=\"$level\">$anchor</text:h>\n"; $last = $m[0][1] + strlen($m[0][0]); } }
  1. Handle last section (or all if there were no headings
if ($i = strlen($article) - $last) $body .= ooParse(substr($article, $last, $i));
  1. Insert the constructed body into the xml template
$xml = str_replace('</office:body>', "\n$body\n</office:body>", $xml);
  1. Dump the template to content.xml file in the working folder
$fh = fopen($file = "$tmp/content.xml",'w+'); fwrite($fh, $xml); fclose($fh);
  1. Add content.xml to the sxw archive...
$zip = new PclZip($sxw); if ($zip->add($file, PCLZIP_OPT_REMOVE_ALL_PATH)) { # ...and send as raw data back to browser as title.sxw ob_end_clean(); header('Content-type: application/zip'); header("Content-Disposition: attachment; filename=\"$title.sxw\""); @readfile($sxw); # Clean up and die @unlink("$tmp/content.xml"); @unlink("$tmp/sxw"); @rmdir($tmp); die; } else $error = $zip->errorInfo(true);
  1. Error, clean up files
if ($error) xwMessage($error,'red'); if (!@unlink("$tmp/content.xml")) xwMessage("Couldn't remove $tmp/content.xml!",'red'); if (!@unlink("$tmp/sxw")) xwMessage("Couldn't remove $tmp/sxw!",'red'); if (!@rmdir($tmp)) xwMessage("Couldn't remove $tmp",'red'); ?>