CLI Program to turn website into static document set

A simple CLI PHP script to massage the output of dokuwiki plugin SiteExport into a set of HTML pages that form the basis of the new documentation set

1 files changed, 221 insertions, 0 deletions

diff --git a/scripts/fix-static-docs.php b/scripts/fix-static-docs.php
new file mode 100644
index 00000000..32133e02
--- /dev/null
+++ b/scripts/fix-static-docs.php
@@ -0,0 +1,221 @@
+<?php 
+/**
+* A Program to post-process the dokuwiki document export to clean it up
+* and fix the broken links
+*
+* @link http://adodb.org/dokuwiki/doku.php?id=v6:offline_docs_build
+* @author Mark Newnham
+* @since 02/13/2015
+*/
+
+/**
+* Recurses a directory and deletes files inside
+*
+* Copied from php.net
+*
+* @param string $dir  The driectory name
+* $return null
+*/
+function delTree($dir) { 
+    $files = array_diff(scandir($dir), array('.','..')); 
+    foreach ($files as $file) {
+      (is_dir("$dir/$file")) ? delTree("$dir/$file") : unlink("$dir/$file"); 
+    } 
+    return rmdir($dir); 
+} 
+
+/**
+* Initializes the listdiraux method with a starting directory point
+*
+* copied from php.net
+*
+* @param  string $dir	Starting directory
+* @return array			FQ list of files
+*/  
+function listdir($dir='.') { 
+    if (!is_dir($dir)) { 
+        return false; 
+    } 
+    $files = array(); 
+    listdiraux($dir, $files); 
+
+    return $files; 
+} 
+
+/**
+* Recurses a directory structure and creates a list of files
+*
+* @param  string 	$dir	Starting directory
+* @param  string[]  $files  By reference, the current file list
+* @return null
+*/  
+function listdiraux($dir, &$files) { 
+    $handle = opendir($dir); 
+    while (($file = readdir($handle)) !== false) { 
+        if ($file == '.' || $file == '..') { 
+            continue; 
+        } 
+        
+		if (preg_match('/v6$/',$file))
+			/*
+			* This is only v5 documentation
+			*/
+			continue;
+		
+		$filepath = $dir == '.' ? $file : $dir . '/' . $file; 
+        if (is_link($filepath)) 
+            continue; 
+        if (is_file($filepath)) 
+            $files[] = $filepath; 
+        else if (is_dir($filepath)) 
+            listdiraux($filepath, $files); 
+    } 
+    closedir($handle); 
+} 
+/*
+* Clean up the documentation directory from prior use
+*/
+if (is_dir('documentation'))
+	deltree('documentation');
+mkdir('documentation');
+
+$files = listdir('documentation-base'); 
+sort($files, SORT_LOCALE_STRING); 
+
+/*
+* Loop through files in documentation-base directory, creating a mirror
+* structure in documentation, and applying the post-process rules defined
+* below
+*/
+foreach ($files as $f) { 
+
+	$r = str_replace('documentation-base','documentation',$f);
+    $dList = explode ('/',$r);
+	$titleList = $dList;
+	/*
+	* Get rid of the initial directory
+	*/
+	array_shift($titleList);
+	
+	$depth = count($dList) -2;
+	$dSlash = '';
+	while(count($dList) > 1)
+	{
+		$dSlash .= array_shift($dList) . '/';
+		if (!is_dir($dSlash))
+			mkdir($dSlash);
+	}
+	if (!is_file($f))
+		continue;
+	if (substr($f,-4) <> 'html')
+	{
+		/*
+		* An image or something else, copy unmodified
+		*/
+		copy ($f,$r);
+		continue;
+	}
+		
+	$prepend = str_repeat('../',$depth);
+	
+	$doc = new DOMDocument();
+	@$doc->loadHTMLFile($f);
+	
+	/*
+	* Remove Page Tools Group
+	*/
+	$xpath = new DOMXPath($doc);
+		
+	/*
+	* Remove Top Menu Tools Group, and add a link to the ADOdb site
+	*/
+	$nodes = $xpath->query("//div[@class='tools group']");
+	foreach($nodes as $node) {
+		$pn = $node->parentNode;
+		$pn->removeChild($node);
+		$newChild = $doc->createElement('div','');
+		$newDiv = $pn->appendChild($newChild);
+		$newDiv->setAttribute('style','text-align:right');
+		$newChild = $doc->createElement('a','ADOdb Web Site');
+		$newA = $newDiv->appendChild($newChild);
+		$newA->setAttribute('href','http://adodb.org');
+    }
+	
+	/*
+	* Remove Trace
+	*/
+	$nodes = $xpath->query("//div[@class='breadcrumbs']");
+	foreach($nodes as $node) {
+		$node->parentNode->removeChild($node);
+    }
+	
+	/*
+	* Remove Side Menu Tools Group
+	*/
+	$nodes = $xpath->query("//div[@id='dokuwiki__pagetools']");
+	foreach($nodes as $node) {
+		$node->parentNode->removeChild($node);
+    }
+	
+	/*
+	* Fix main links
+	*/
+	$nodes = $xpath->query("//a[@class='wikilink1']");
+	foreach($nodes as $node) {
+		$n =  $node->getAttribute('title');
+		$p = $prepend . str_replace(':','/',$n) . '.html';
+		$node->setAttribute('href', $p);
+    }
+	
+	/*
+	* Fix In Page links
+	*/
+	$nodes = $xpath->query("//a[@class='wikilink2']");
+	foreach($nodes as $node) {
+		$n =  $node->getAttribute('title');
+		$p = $prepend . str_replace(':','/',$n) . '.html';
+		$node->setAttribute('href', $p);
+    }
+	
+	/*
+	* Make Graphic point to first page. This will break if the image size
+	* ever changes.
+	*/
+	$corePage = $prepend . '/index.html';
+	$nodes = $xpath->query("//img[@width='176']");
+	foreach($nodes as $node) {
+		$node->parentNode->setAttribute('href', $corePage);
+    }
+	
+	/*
+	* Change title of page
+	*/
+	$nodes = $xpath->query("//title");
+	foreach($nodes as $node) {
+		
+		
+		$docTitle = implode(':',$titleList);
+		$docTitle = str_replace('.html','',$docTitle);
+		$pn = $node->parentNode; 
+		$pn->removeChild($node);
+		$newChild = $doc->createElement('title',$docTitle);
+		$pn->appendChild($newChild);
+		
+    }
+		
+	$doc->saveHTMLFile($r);
+	
+    echo  $r, "\n"; 
+}
+/*
+* Now remove the original index and replace it with the hardcopy documentation one
+*/
+unlink ('documentation/index.html');
+rename('documentation/adodb_index.html','documentation/index.html');
+
+/*
+* We could add in an auto zip and upload here, but this is a good place to
+* stop and check the output
+*/
+
+?>