Difference between revisions of "Oo-regexp-export.php"
| Line 7: | Line 7: | ||
$script = $GLOBALS['xwScript']; | $script = $GLOBALS['xwScript']; | ||
$bToken = ':-:'.'bullet'.':-:'; | $bToken = ':-:'.'bullet'.':-:'; | ||
| + | $hToken = ':-:'.'heading'.':-:'; | ||
| − | # | + | # ;text (pre) |
| − | + | $text = preg_replace("/\\n;(.+?)\\n/m", "\n\n$hToken\\1\n\n", $text); | |
| − | $text = preg_replace( | ||
| − | |||
| − | |||
# *text (pre) | # *text (pre) | ||
| Line 43: | Line 41: | ||
$xml .= "<text:p text:style-name=\"Text body\">$text</text:p>\n"; | $xml .= "<text:p text:style-name=\"Text body\">$text</text:p>\n"; | ||
} | } | ||
| + | |||
| + | # ;text (post) | ||
| + | $xml = preg_replace("/(?<=>)$hToken(.+)(?=<)/", "<text:span text:style-name=\"T2\">\\1</text:span>", $xml); | ||
# *text (post) | # *text (post) | ||
| Line 62: | Line 63: | ||
$error = false; | $error = false; | ||
| − | # Create a unique working folder | + | # Create a unique working folder to build zip in |
mkdir($tmp = "$htdocs/wiki/tmp/".uniqid('export')); | mkdir($tmp = "$htdocs/wiki/tmp/".uniqid('export')); | ||
| − | + | $sxw = "$tmp/sxw"; | |
# Zip compression library - www.phpconcept.net/pclzip/ | # Zip compression library - www.phpconcept.net/pclzip/ | ||
| Line 70: | Line 71: | ||
require_once("$htdocs/wiki/xmlwiki/lib/pclzip/pclzip.lib.php"); | require_once("$htdocs/wiki/xmlwiki/lib/pclzip/pclzip.lib.php"); | ||
| − | # Get OO content template | + | # Get OO content and styles template articles |
| − | $xml = xwArticleContent('oo-doc- | + | $content = xwArticleContent('oo-doc-content.xml'); |
| + | $styles = xwArticleContent('oo-doc-styles.xml'); | ||
# Insert our document info into existing tokens | # Insert our document info into existing tokens | ||
| − | $ | + | $content = str_replace('Title', str_replace('_',' ',$title), $content); |
| − | $ | + | $content = str_replace('Version', 'Exported by '.$GLOBALS['xwUserName'].' on '.date('d M Y').' at '.date('H:i:s'), $content); |
| − | + | $styles = str_replace('My-Header', "$script/$title", $styles); | |
| − | # Build | + | # Convert problem characters (should make them work instead!) |
| + | $article = str_replace('’',"'",$article); | ||
| + | $article = str_replace('…','...',$article); | ||
| + | $article = str_replace('–','-',$article); | ||
| + | $article = preg_replace('/“|�/','"',$article); | ||
| + | |||
| + | # Build main document body | ||
# - the match splits the article into sections at headings, extracting heading, level & text | # - the match splits the article into sections at headings, extracting heading, level & text | ||
| − | $body = ''; $last = 0; | + | $body = ''; $last = 0; $toclist = array(); |
if (preg_match_all('/<h1><tmp num="(#*)" title="(.*?)" anchor="(.*?)".><.h1>/s', $article, $sections, PREG_SET_ORDER | PREG_OFFSET_CAPTURE)) { | if (preg_match_all('/<h1><tmp num="(#*)" title="(.*?)" anchor="(.*?)".><.h1>/s', $article, $sections, PREG_SET_ORDER | PREG_OFFSET_CAPTURE)) { | ||
foreach ($sections as $m) { | foreach ($sections as $m) { | ||
| Line 87: | Line 95: | ||
if ($i = $m[0][1] - $last) $body .= ooParse(substr($article, $last, $i)); | if ($i = $m[0][1] - $last) $body .= ooParse(substr($article, $last, $i)); | ||
$style = $level > 1 ? "Heading $level" : 'P11'; | $style = $level > 1 ? "Heading $level" : 'P11'; | ||
| + | if ($level && $level < 3) $toclist[] = array('P'.($level+7), $anchor); | ||
if ($anchor) $body .= "<text:h text:style-name=\"$style\" text:level=\"$level\">$anchor</text:h>\n"; | if ($anchor) $body .= "<text:h text:style-name=\"$style\" text:level=\"$level\">$anchor</text:h>\n"; | ||
$last = $m[0][1] + strlen($m[0][0]); | $last = $m[0][1] + strlen($m[0][0]); | ||
| Line 95: | Line 104: | ||
# Insert the constructed body into the xml template | # Insert the constructed body into the xml template | ||
| − | $ | + | $content = str_replace('</office:body>', "\n$body\n</office:body>", $content); |
| − | # Dump the | + | # Insert TOC msg into template |
| − | $fh = fopen($ | + | $toc = "<text:p text:style-name=\"P8\">Contents is out of date, please update fields from tools menu\n<text:tab-stop/>???</text:p>\n"; |
| − | fwrite($fh, $xml); | + | $content = str_replace('</text:index-body>', "$toc\n</text:index-body>", $content); |
| + | |||
| + | # Dump the content & styles templates to files in the working folder | ||
| + | $fh = fopen($file1 = "$tmp/content.xml",'w+'); | ||
| + | fwrite($fh, $content); | ||
| + | fclose($fh); | ||
| + | $fh = fopen($file2 = "$tmp/styles.xml",'w+'); | ||
| + | fwrite($fh, $styles); | ||
fclose($fh); | fclose($fh); | ||
| − | # Add | + | # Add the files to the sxw archive... |
$zip = new PclZip($sxw); | $zip = new PclZip($sxw); | ||
| − | if ($zip-> | + | if ($zip->create( array($file1, $file2), PCLZIP_OPT_REMOVE_ALL_PATH )) { |
# ...and send as raw data back to browser as title.sxw | # ...and send as raw data back to browser as title.sxw | ||
ob_end_clean(); | ob_end_clean(); | ||
| Line 111: | Line 127: | ||
@readfile($sxw); | @readfile($sxw); | ||
# Clean up and die | # Clean up and die | ||
| − | @unlink( | + | @unlink($file1); |
| − | @unlink( | + | @unlink($file2); |
| + | @unlink($sxw); | ||
@rmdir($tmp); | @rmdir($tmp); | ||
die; | die; | ||
| Line 119: | Line 136: | ||
# Error, clean up files | # Error, clean up files | ||
| − | + | xwMessage($error,'red'); | |
| − | + | @unlink($file1); | |
| − | + | @unlink($file2); | |
| − | + | @unlink($sxw); | |
| + | @rmdir($tmp); | ||
?> | ?> | ||
Revision as of 21:27, 15 August 2005
<?
- Export article to OO format
- - only does document.php transform and its own wiki markup
# Main wiki->oo parser function ooParse($text) { $script = $GLOBALS['xwScript']; $bToken = ':-:'.'bullet'.':-:'; $hToken = ':-:'.'heading'.':-:';
# ;text (pre) $text = preg_replace("/\\n;(.+?)\\n/m", "\n\n$hToken\\1\n\n", $text);
# *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) $xml = preg_replace("/(?<=>)$hToken(.+)(?=<)/", "<text:span text:style-name=\"T2\">\\1</text:span>", $xml);
# *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(' ', ' ', $xml);
return $xml; }
$htdocs = $GLOBALS['_SERVER']['DOCUMENT_ROOT']; $error = false;
- Create a unique working folder to build zip in
mkdir($tmp = "$htdocs/wiki/tmp/".uniqid('export')); $sxw = "$tmp/sxw";
- Zip compression library - www.phpconcept.net/pclzip/
define('PCLZIP_TEMPORARY_DIR', "$tmp/"); require_once("$htdocs/wiki/xmlwiki/lib/pclzip/pclzip.lib.php");
- Get OO content and styles template articles
$content = xwArticleContent('oo-doc-content.xml'); $styles = xwArticleContent('oo-doc-styles.xml');
- Insert our document info into existing tokens
$content = str_replace('Title', str_replace('_',' ',$title), $content); $content = str_replace('Version', 'Exported by '.$GLOBALS['xwUserName'].' on '.date('d M Y').' at '.date('H:i:s'), $content); $styles = str_replace('My-Header', "$script/$title", $styles);
- Convert problem characters (should make them work instead!)
$article = str_replace('’',"'",$article); $article = str_replace('…','...',$article); $article = str_replace('–','-',$article); $article = preg_replace('/“|�/','"',$article);
- Build main document body
- - the match splits the article into sections at headings, extracting heading, level & text
$body = ; $last = 0; $toclist = array();
if (preg_match_all('/
