Difference between revisions of "Extension:PdfBook"

From Organic Design wiki
(action hook must return true since no die;)
(0.0.5 - generate books from queries or explicit lists if not a cat)
Line 7: Line 7:
 
if (!defined('MEDIAWIKI')) die('Not an entry point.');
 
if (!defined('MEDIAWIKI')) die('Not an entry point.');
  
define('PDFBOOK_VERSION','0.0.4, 2007-09-25');
+
define('PDFBOOK_VERSION','0.0.5, 2007-09-25');
  
$wgPdfBookMagic               = "book";
+
$wgPdfBookMagic         = "book";
$wgExtensionFunctions[]       = 'wfSetupPdfBook';
+
$wgExtensionFunctions[] = 'wfSetupPdfBook';
 
$wgHooks['LanguageGetMagic'][] = 'wfPdfBookLanguageGetMagic';
 
$wgHooks['LanguageGetMagic'][] = 'wfPdfBookLanguageGetMagic';
  
 
$wgExtensionCredits['parserhook'][] = array(
 
$wgExtensionCredits['parserhook'][] = array(
'name'       => 'Pdf Book',
+
'name' => 'Pdf Book',
 
'author'      => '[http://www.organicdesign.co.nz/nad User:Nad]',
 
'author'      => '[http://www.organicdesign.co.nz/nad User:Nad]',
 
'description' => 'Composes a book from articles in a category and exports as a PDF book',
 
'description' => 'Composes a book from articles in a category and exports as a PDF book',
'url'         => 'http://www.mediawiki.org/wiki/Extension:Pdf_Book',
+
'url' => 'http://www.mediawiki.org/wiki/Extension:Pdf_Book',
 
'version'    => PDFBOOK_VERSION
 
'version'    => PDFBOOK_VERSION
 
);
 
);
Line 74: Line 74:
 
$top    = $this->setProperty('TopMargin',  '1cm');
 
$top    = $this->setProperty('TopMargin',  '1cm');
 
$bottom  = $this->setProperty('BottomMargin','1cm');
 
$bottom  = $this->setProperty('BottomMargin','1cm');
$font    = $this->setProperty('Font',       'Arial');
+
$font    = $this->setProperty('Font', 'Arial');
 
$size    = $this->setProperty('FontSize',    '8');
 
$size    = $this->setProperty('FontSize',    '8');
 
$link    = $this->setProperty('LinkColour',  '217A28');
 
$link    = $this->setProperty('LinkColour',  '217A28');
Line 81: Line 81:
 
if (!is_array($exclude)) $exclude = split('\\s*,\\s*',$exclude);
 
if (!is_array($exclude)) $exclude = split('\\s*,\\s*',$exclude);
 
   
 
   
# Select all article titles in category
+
# Select articles from members if a category or links in content if not
$title = $article->getTitle();
+
$articles = array();
$book  = $title->getText();
+
$title   = $article->getTitle();
$cat    = $title->getDBkey();
+
$opt      = ParserOptions::newFromUser($wgUser);
$db    = &wfGetDB(DB_SLAVE);
+
if ($title->getNamespace() == NS_CATEGORY) {
$cl    = $db->tableName('categorylinks');
+
$cat    = $title->getDBkey();
$result = $db->query("SELECT cl_from FROM $cl WHERE cl_to = '$cat' ORDER BY cl_sortkey");
+
$db    = &wfGetDB(DB_SLAVE);
if ($result instanceof ResultWrapper) $result = $result->result;
+
$cl    = $db->tableName('categorylinks');
 +
$result = $db->query("SELECT cl_from FROM $cl WHERE cl_to = '$cat' ORDER BY cl_sortkey");
 +
if ($result instanceof ResultWrapper) $result = $result->result;
 +
while ($row = $db->fetchRow($result)) $articles[] = Title::newFromID($row[0]);
 +
}
 +
else {
 +
$text = $article->fetchContent();
 +
$text = $wgParser->preprocess($text,$title,$opt);
 +
if (preg_match_all('/^\\*\\s*\\[{2}\\s*([^\\|\\]]+)\\s*.*?\\]{2}/m',$text,$links))
 +
foreach ($links[1] as $link) $articles[] = Title::newFromText($link);
 +
}
  
 
# Format the article's as a single HTML document with absolute URL's
 
# Format the article's as a single HTML document with absolute URL's
$html = '';
+
$book   = $title->getText();
 +
$html   = '';
 
$wgArticlePath = $wgServer.$wgArticlePath;
 
$wgArticlePath = $wgServer.$wgArticlePath;
 
$wgScriptPath  = $wgServer.$wgScriptPath;
 
$wgScriptPath  = $wgServer.$wgScriptPath;
 
$wgUploadPath  = $wgServer.$wgUploadPath;
 
$wgUploadPath  = $wgServer.$wgUploadPath;
 
$wgScript      = $wgServer.$wgScript;
 
$wgScript      = $wgServer.$wgScript;
while ($row = $db->fetchRow($result)) {
+
foreach ($articles as $title) {
$title = Title::newFromID($row[0]);
 
 
$ttext = $title->getPrefixedText();
 
$ttext = $title->getPrefixedText();
 
if (!in_array($ttext,$exclude)) {
 
if (!in_array($ttext,$exclude)) {
$article       = new Article($title);
+
$article = new Article($title);
$text         = $article->fetchContent();
+
$text   = $article->fetchContent();
$opt          = ParserOptions::newFromUser($wgUser);
+
$text   = preg_replace('/<!--([^@]+?)-->/s','@@'.'@@$1@@'.'@@',$text); # preserve HTML comments
$text = preg_replace('/<!--([^@]+?)-->/s','@@'.'@@$1@@'.'@@',$text); # preserve HTML comments
 
 
$opt->setEditSection(false);    # remove section-edit links
 
$opt->setEditSection(false);    # remove section-edit links
 
$wgOut->setHTMLTitle($ttext);  # use this so DISPLAYTITLE magic works
 
$wgOut->setHTMLTitle($ttext);  # use this so DISPLAYTITLE magic works
$out           = $wgParser->parse($text,$title,$opt,true,true);
+
$out     = $wgParser->parse($text,$title,$opt,true,true);
$ttext         = $wgOut->getHTMLTitle();
+
$ttext   = $wgOut->getHTMLTitle();
$text         = $out->getText();
+
$text   = $out->getText();
$text         = preg_replace('/@{4}([^@]+?)@{4}/s','<!--$1-->',$text); # HTML comments hack
+
$text   = preg_replace('/@{4}([^@]+?)@{4}/s','<!--$1-->',$text); # HTML comments hack
$text         = preg_replace('/<table/','<table border borderwidth=2 cellpadding=3 cellspacing=0',$text);
+
$text   = preg_replace('/<table/','<table border borderwidth=2 cellpadding=3 cellspacing=0',$text);
$ttext         = basename($ttext);
+
$ttext   = basename($ttext);
$html         .= utf8_decode("<h1>$ttext</h1>$text\n");
+
$html   .= utf8_decode("<h1>$ttext</h1>$text\n");
 
}
 
}
 
}
 
}
Line 136: Line 145:
 
}
 
}
  
return true;
+
return false;
 
}
 
}
  

Revision as of 04:30, 25 September 2007

<?php

  1. Extension:PdfBook
Info.svg These are the MediaWiki extensions we're using and/or developing. Please refer to the information on the mediawiki.org wiki for installation and usage details. Extensions here which have no corresponding mediawiki article are either not ready for use or have been superseded. You can also browse our extension code in our local Subversion repository or our GitHub mirror.

Template:PhpCategory:Extensions created with Template:Extension

  1. - Licenced under LGPL (http://www.gnu.org/copyleft/lesser.html)
  2. - Author: User:Nad
  3. - Started: 2007-08-08

if (!defined('MEDIAWIKI')) die('Not an entry point.');

define('PDFBOOK_VERSION','0.0.5, 2007-09-25');

$wgPdfBookMagic = "book"; $wgExtensionFunctions[] = 'wfSetupPdfBook'; $wgHooks['LanguageGetMagic'][] = 'wfPdfBookLanguageGetMagic';

$wgExtensionCredits['parserhook'][] = array( 'name' => 'Pdf Book', 'author' => 'User:Nad', 'description' => 'Composes a book from articles in a category and exports as a PDF book', 'url' => 'http://www.mediawiki.org/wiki/Extension:Pdf_Book', 'version' => PDFBOOK_VERSION );

class PdfBook {

# Constructor function PdfBook() { global $wgHooks,$wgParser,$wgPdfBookMagic; $wgParser->setFunctionHook($wgPdfBookMagic,array($this,'magicBook')); $wgHooks['UnknownAction'][] = $this;

# Add a new pdf log type global $wgLogTypes,$wgLogNames,$wgLogHeaders,$wgLogActions; $wgLogTypes[] = 'pdf'; $wgLogNames ['pdf'] = 'pdflogpage'; $wgLogHeaders['pdf'] = 'pdflogpagetext'; $wgLogActions['pdf/book'] = 'pdflogentry'; }

# Expand the book-magic function magicBook(&$parser) {

# Populate $argv with both named and numeric parameters $argv = array(); foreach (func_get_args() as $arg) if (!is_object($arg)) { if (preg_match('/^(.+?)\\s*=\\s*(.+)$/',$arg,$match)) $argv[$match[1]] = $match[2]; else $argv[] = $arg; }

# Return result with available parser flags return array( $text, found => true, nowiki => false, noparse => false, noargs => false, isHTML => false ); }

function onUnknownAction($action,$article) { global $wgOut,$wgUser,$wgTitle,$wgParser; global $wgServer,$wgArticlePath,$wgScriptPath,$wgUploadPath,$wgUploadDirectory,$wgScript;

if ($action == 'pdfbook') {

# Log the export $msg = $wgUser->getUserPage()->getPrefixedText().' exported as a PDF book'; $log = new LogPage('pdf',false); $log->addEntry('book',$wgTitle,$msg);

# Initialise PDF variables $layout = '--firstpage toc'; $left = $this->setProperty('LeftMargin', '1cm'); $right = $this->setProperty('RightMargin', '1cm'); $top = $this->setProperty('TopMargin', '1cm'); $bottom = $this->setProperty('BottomMargin','1cm'); $font = $this->setProperty('Font', 'Arial'); $size = $this->setProperty('FontSize', '8'); $link = $this->setProperty('LinkColour', '217A28'); $levels = $this->setProperty('TocLevels', '2'); $exclude = $this->setProperty('Exclude', array()); if (!is_array($exclude)) $exclude = split('\\s*,\\s*',$exclude);

# Select articles from members if a category or links in content if not $articles = array(); $title = $article->getTitle(); $opt = ParserOptions::newFromUser($wgUser); if ($title->getNamespace() == NS_CATEGORY) { $cat = $title->getDBkey(); $db = &wfGetDB(DB_SLAVE); $cl = $db->tableName('categorylinks'); $result = $db->query("SELECT cl_from FROM $cl WHERE cl_to = '$cat' ORDER BY cl_sortkey"); if ($result instanceof ResultWrapper) $result = $result->result; while ($row = $db->fetchRow($result)) $articles[] = Title::newFromID($row[0]); } else { $text = $article->fetchContent(); $text = $wgParser->preprocess($text,$title,$opt); if (preg_match_all('/^\\*\\s*\\[{2}\\s*([^\\|\\]]+)\\s*.*?\\]{2}/m',$text,$links)) foreach ($links[1] as $link) $articles[] = Title::newFromText($link); }

# Format the article's as a single HTML document with absolute URL's $book = $title->getText(); $html = ; $wgArticlePath = $wgServer.$wgArticlePath; $wgScriptPath = $wgServer.$wgScriptPath; $wgUploadPath = $wgServer.$wgUploadPath; $wgScript = $wgServer.$wgScript; foreach ($articles as $title) { $ttext = $title->getPrefixedText(); if (!in_array($ttext,$exclude)) { $article = new Article($title); $text = $article->fetchContent(); $text = preg_replace('//s','@@'.'@@$1@@'.'@@',$text); # preserve HTML comments $opt->setEditSection(false); # remove section-edit links $wgOut->setHTMLTitle($ttext); # use this so DISPLAYTITLE magic works $out = $wgParser->parse($text,$title,$opt,true,true); $ttext = $wgOut->getHTMLTitle(); $text = $out->getText(); $text = preg_replace('/@{4}([^@]+?)@{4}/s',,$text); # HTML comments hack $text = preg_replace('/<table/','<table border borderwidth=2 cellpadding=3 cellspacing=0',$text); $ttext = basename($ttext);

$html .= utf8_decode("

$ttext

$text\n");

} }

# Write the HTML to a tmp file $file = "$wgUploadDirectory/".uniqid('pdf-book'); $fh = fopen($file,'w+'); fwrite($fh,$html); fclose($fh);

# Send the file to the client via htmldoc converter $wgOut->disable(); header("Content-Type: application/pdf"); header("Content-Disposition: attachment; filename=\"$book.pdf\""); $cmd = "--left $left --right $right --top $top --bottom $bottom"; $cmd .= " --header ... --footer .1. --headfootsize 8 --quiet --jpeg --color"; $cmd .= " --bodyfont $font --fontsize $size --linkstyle plain --linkcolor $links"; $cmd .= " --toclevels $levels --format pdf14 --numbered $layout"; $cmd = "htmldoc -t pdf --charset iso-8859-1 $cmd $file"; putenv("HTMLDOC_NOCGI=1"); passthru($cmd); @unlink($file); }

return false; }

# Return a property for htmldoc using global, request or passed default function setProperty($name,$default) { if (isset($_REQUEST["pdf$name"])) return $_REQUEST["pdf$name"]; if (isset($GLOBALS["wgPdfBook$name"])) return $GLOBALS["wgPdfBook$name"]; return $default; }

# Needed in some versions to prevent Special:Version from breaking function __toString() { return 'PdfBook'; }

	}
  1. Called from $wgExtensionFunctions array when initialising extensions

function wfSetupPdfBook() { global $wgPdfBook; $wgPdfBook = new PdfBook(); }

  1. Needed in MediaWiki >1.8.0 for magic word hooks to work properly

function wfPdfBookLanguageGetMagic(&$magicWords,$langCode = 0) { global $wgPdfBookMagic; $magicWords[$wgPdfBookMagic] = array(0,$wgPdfBookMagic); return true; } ?>