first commit

htmlpurifier-4.10.0/maintenance/.htaccess

@@ -0,0 +1 @@
+Deny from all

htmlpurifier-4.10.0/maintenance/add-vimline.php

@@ -0,0 +1,130 @@
+#!/usr/bin/php
+<?php
+
+chdir(dirname(__FILE__));
+require_once 'common.php';
+assertCli();
+
+/**
+ * @file
+ * Adds vimline to files
+ */
+
+chdir(dirname(__FILE__) . '/..');
+$FS = new FSTools();
+
+$vimline = 'vim: et sw=4 sts=4';
+
+$files = $FS->globr('.', '*');
+foreach ($files as $file) {
+    if (
+        !is_file($file) ||
+        prefix_is('./docs/doxygen', $file) ||
+        prefix_is('./library/standalone', $file) ||
+        prefix_is('./docs/specimens', $file) ||
+        postfix_is('.ser', $file) ||
+        postfix_is('.tgz', $file) ||
+        postfix_is('.patch', $file) ||
+        postfix_is('.dtd', $file) ||
+        postfix_is('.ent', $file) ||
+        postfix_is('.png', $file) ||
+        postfix_is('.ico', $file) ||
+        // wontfix
+        postfix_is('.vtest', $file) ||
+        postfix_is('.svg', $file) ||
+        postfix_is('.phpt', $file) ||
+        postfix_is('VERSION', $file) ||
+        postfix_is('WHATSNEW', $file) ||
+        postfix_is('configdoc/usage.xml', $file) ||
+        postfix_is('library/HTMLPurifier.includes.php', $file) ||
+        postfix_is('library/HTMLPurifier.safe-includes.php', $file) ||
+        postfix_is('smoketests/xssAttacks.xml', $file) ||
+        // phpt files
+        postfix_is('.diff', $file) ||
+        postfix_is('.exp', $file) ||
+        postfix_is('.log', $file) ||
+        postfix_is('.out', $file) ||
+
+        $file == './library/HTMLPurifier/Lexer/PH5P.php' ||
+        $file == './maintenance/PH5P.php'
+    ) continue;
+    $ext = strrchr($file, '.');
+    if (
+        postfix_is('README', $file) ||
+        postfix_is('LICENSE', $file) ||
+        postfix_is('CREDITS', $file) ||
+        postfix_is('INSTALL', $file) ||
+        postfix_is('NEWS', $file) ||
+        postfix_is('TODO', $file) ||
+        postfix_is('WYSIWYG', $file) ||
+        postfix_is('Changelog', $file)
+    ) $ext = '.txt';
+    if (postfix_is('Doxyfile', $file)) $ext = 'Doxyfile';
+    if (postfix_is('.php.in', $file)) $ext = '.php';
+    $no_nl = false;
+    switch ($ext) {
+        case '.php':
+        case '.inc':
+        case '.js':
+            $line = '// %s';
+            break;
+        case '.html':
+        case '.xsl':
+        case '.xml':
+        case '.htc':
+            $line = "<!-- %s\n-->";
+            break;
+        case '.htmlt':
+            $no_nl = true;
+            $line = '--# %s';
+            break;
+        case '.ini':
+            $line = '; %s';
+            break;
+        case '.css':
+            $line = '/* %s */';
+            break;
+        case '.bat':
+            $line = 'rem %s';
+            break;
+        case '.txt':
+        case '.utf8':
+            if (
+                prefix_is('./library/HTMLPurifier/ConfigSchema', $file) ||
+                prefix_is('./smoketests/test-schema', $file) ||
+                prefix_is('./tests/HTMLPurifier/StringHashParser', $file)
+            ) {
+                $no_nl = true;
+                $line = '--# %s';
+            } else {
+                $line = '    %s';
+            }
+            break;
+        case 'Doxyfile':
+            $line = '# %s';
+            break;
+        default:
+            throw new Exception('Unknown file: ' . $file);
+    }
+
+    echo "$file\n";
+    $contents = file_get_contents($file);
+
+    $regex = '~' . str_replace('%s', 'vim: .+', preg_quote($line, '~')) .  '~m';
+    $contents = preg_replace($regex, '', $contents);
+
+    $contents = rtrim($contents);
+
+    if (strpos($contents, "\r\n") !== false) $nl = "\r\n";
+    elseif (strpos($contents, "\n") !== false) $nl = "\n";
+    elseif (strpos($contents, "\r") !== false) $nl = "\r";
+    else $nl = PHP_EOL;
+
+    if (!$no_nl) $contents .= $nl;
+    $contents .= $nl . str_replace('%s', $vimline, $line) . $nl;
+
+    file_put_contents($file, $contents);
+
+}
+
+// vim: et sw=4 sts=4

htmlpurifier-4.10.0/maintenance/common.php

@@ -0,0 +1,25 @@
+<?php
+
+function assertCli()
+{
+    if (php_sapi_name() != 'cli' && !getenv('PHP_IS_CLI')) {
+        echo 'Script cannot be called from web-browser (if you are indeed calling via cli,
+set environment variable PHP_IS_CLI to work around this).';
+        exit(1);
+    }
+}
+
+function prefix_is($comp, $subject)
+{
+    return strncmp($comp, $subject, strlen($comp)) === 0;
+}
+
+function postfix_is($comp, $subject)
+{
+    return strlen($subject) < $comp ? false : substr($subject, -strlen($comp)) === $comp;
+}
+
+// Load useful stuff like FSTools
+require_once dirname(__FILE__) . '/../extras/HTMLPurifierExtras.auto.php';
+
+// vim: et sw=4 sts=4

htmlpurifier-4.10.0/maintenance/compile-doxygen.sh

@@ -0,0 +1,11 @@
+#!/bin/bash
+cd ..
+mkdir docs/doxygen
+rm -Rf docs/doxygen/*
+doxygen 1>docs/doxygen/info.log 2>docs/doxygen/errors.log
+if [ "$?" != 0 ]; then
+    cat docs/doxygen/errors.log
+    exit
+fi
+cd docs
+tar czf doxygen.tgz doxygen

htmlpurifier-4.10.0/maintenance/config-scanner.php

@@ -0,0 +1,155 @@
+#!/usr/bin/php
+<?php
+
+chdir(dirname(__FILE__));
+require_once 'common.php';
+require_once '../library/HTMLPurifier.auto.php';
+assertCli();
+
+if (version_compare(PHP_VERSION, '5.2.2', '<')) {
+    echo "This script requires PHP 5.2.2 or later, for tokenizer line numbers.";
+    exit(1);
+}
+
+/**
+ * @file
+ * Scans HTML Purifier source code for $config tokens and records the
+ * directive being used; configdoc can use this info later.
+ *
+ * Currently, this just dumps all the info onto the console. Eventually, it
+ * will create an XML file that our XSLT transform can use.
+ */
+
+$FS = new FSTools();
+chdir(dirname(__FILE__) . '/../library/');
+$raw_files = $FS->globr('.', '*.php');
+$files = array();
+foreach ($raw_files as $file) {
+    $file = substr($file, 2); // rm leading './'
+    if (strncmp('standalone/', $file, 11) === 0) continue; // rm generated files
+    if (substr_count($file, '.') > 1) continue; // rm meta files
+    $files[] = $file;
+}
+
+/**
+ * Moves the $i cursor to the next non-whitespace token
+ */
+function consumeWhitespace($tokens, &$i)
+{
+    do {$i++;} while (is_array($tokens[$i]) && $tokens[$i][0] === T_WHITESPACE);
+}
+
+/**
+ * Tests whether or not a token is a particular type. There are three run-cases:
+ *      - ($token, $expect_token): tests if the token is $expect_token type;
+ *      - ($token, $expect_value): tests if the token is the string $expect_value;
+ *      - ($token, $expect_token, $expect_value): tests if token is $expect_token type, and
+ *        its string representation is $expect_value
+ */
+function testToken($token, $value_or_token, $value = null)
+{
+    if (is_null($value)) {
+        if (is_int($value_or_token)) return is_array($token) && $token[0] === $value_or_token;
+        else return $token === $value_or_token;
+    } else {
+        return is_array($token) && $token[0] === $value_or_token && $token[1] === $value;
+    }
+}
+
+$counter = 0;
+$full_counter = 0;
+$tracker = array();
+
+foreach ($files as $file) {
+    $tokens = token_get_all(file_get_contents($file));
+    $file = str_replace('\\', '/', $file);
+    for ($i = 0, $c = count($tokens); $i < $c; $i++) {
+        $ok = false;
+        // Match $config
+        if (!$ok && testToken($tokens[$i], T_VARIABLE, '$config')) $ok = true;
+        // Match $this->config
+        while (!$ok && testToken($tokens[$i], T_VARIABLE, '$this')) {
+            consumeWhitespace($tokens, $i);
+            if (!testToken($tokens[$i], T_OBJECT_OPERATOR)) break;
+            consumeWhitespace($tokens, $i);
+            if (testToken($tokens[$i], T_STRING, 'config')) $ok = true;
+            break;
+        }
+        if (!$ok) continue;
+
+        $ok = false;
+        for($i++; $i < $c; $i++) {
+            if ($tokens[$i] === ',' || $tokens[$i] === ')' || $tokens[$i] === ';') {
+                break;
+            }
+            if (is_string($tokens[$i])) continue;
+            if ($tokens[$i][0] === T_OBJECT_OPERATOR) {
+                $ok = true;
+                break;
+            }
+        }
+        if (!$ok) continue;
+
+        $line = $tokens[$i][2];
+
+        consumeWhitespace($tokens, $i);
+        if (!testToken($tokens[$i], T_STRING, 'get')) continue;
+
+        consumeWhitespace($tokens, $i);
+        if (!testToken($tokens[$i], '(')) continue;
+
+        $full_counter++;
+
+        $matched = false;
+        do {
+
+            // What we currently don't match are batch retrievals, and
+            // wildcard retrievals. This data might be useful in the future,
+            // which is why we have a do {} while loop that doesn't actually
+            // do anything.
+
+            consumeWhitespace($tokens, $i);
+            if (!testToken($tokens[$i], T_CONSTANT_ENCAPSED_STRING)) continue;
+            $id = substr($tokens[$i][1], 1, -1);
+
+            $counter++;
+            $matched = true;
+
+            if (!isset($tracker[$id])) $tracker[$id] = array();
+            if (!isset($tracker[$id][$file])) $tracker[$id][$file] = array();
+            $tracker[$id][$file][] = $line;
+
+        } while (0);
+
+        //echo "$file:$line uses $namespace.$directive\n";
+    }
+}
+
+echo "\n$counter/$full_counter instances of \$config or \$this->config found in source code.\n";
+
+echo "Generating XML... ";
+
+$xw = new XMLWriter();
+$xw->openURI('../configdoc/usage.xml');
+$xw->setIndent(true);
+$xw->startDocument('1.0', 'UTF-8');
+$xw->startElement('usage');
+foreach ($tracker as $id => $files) {
+    $xw->startElement('directive');
+    $xw->writeAttribute('id', $id);
+    foreach ($files as $file => $lines) {
+        $xw->startElement('file');
+        $xw->writeAttribute('name', $file);
+        foreach ($lines as $line) {
+            $xw->writeElement('line', $line);
+        }
+        $xw->endElement();
+    }
+    $xw->endElement();
+}
+$xw->endElement();
+$xw->flush();
+
+echo "done!\n";
+
+// vim: et sw=4 sts=4

htmlpurifier-4.10.0/maintenance/flush-definition-cache.php

@@ -0,0 +1,42 @@
+#!/usr/bin/php
+<?php
+
+chdir(dirname(__FILE__));
+require_once 'common.php';
+assertCli();
+
+/**
+ * @file
+ * Flushes the definition serial cache. This file should be
+ * called if changes to any subclasses of HTMLPurifier_Definition
+ * or related classes (such as HTMLPurifier_HTMLModule) are made. This
+ * may also be necessary if you've modified a customized version.
+ *
+ * @param Accepts one argument, cache type to flush; otherwise flushes all
+ *      the caches.
+ */
+
+echo "Flushing cache... \n";
+
+require_once(dirname(__FILE__) . '/../library/HTMLPurifier.auto.php');
+
+$config = HTMLPurifier_Config::createDefault();
+
+$names = array('HTML', 'CSS', 'URI', 'Test');
+if (isset($argv[1])) {
+    if (in_array($argv[1], $names)) {
+        $names = array($argv[1]);
+    } else {
+        throw new Exception("Cache parameter {$argv[1]} is not a valid cache");
+    }
+}
+
+foreach ($names as $name) {
+    echo " - Flushing $name\n";
+    $cache = new HTMLPurifier_DefinitionCache_Serializer($name);
+    $cache->flush($config);
+}
+
+echo "Cache flushed successfully.\n";
+
+// vim: et sw=4 sts=4

htmlpurifier-4.10.0/maintenance/flush.php

@@ -0,0 +1,30 @@
+#!/usr/bin/php
+<?php
+
+chdir(dirname(__FILE__));
+require_once 'common.php';
+assertCli();
+
+/**
+ * @file
+ * Runs all generation/flush cache scripts to ensure that somewhat volatile
+ * generated files are up-to-date.
+ */
+
+function e($cmd)
+{
+    echo "\$ $cmd\n";
+    passthru($cmd, $status);
+    echo "\n";
+    if ($status) exit($status);
+}
+
+$php = empty($_SERVER['argv'][1]) ? 'php' : $_SERVER['argv'][1];
+
+e($php . ' generate-includes.php');
+e($php . ' generate-schema-cache.php');
+e($php . ' flush-definition-cache.php');
+e($php . ' generate-standalone.php');
+e($php . ' config-scanner.php');
+
+// vim: et sw=4 sts=4

htmlpurifier-4.10.0/maintenance/generate-entity-file.php

@@ -0,0 +1,75 @@
+#!/usr/bin/php
+<?php
+
+chdir(dirname(__FILE__));
+require_once 'common.php';
+assertCli();
+
+/**
+ * @file
+ * Parses *.ent files into an entity lookup table, and then serializes and
+ * writes the whole kaboodle to a file. The resulting file is cached so
+ * that this script does not need to be run. This script should rarely,
+ * if ever, be run, since HTML's entities are fairly immutable.
+ */
+
+// here's where the entity files are located, assuming working directory
+// is the same as the location of this PHP file. Needs trailing slash.
+$entity_dir = '../docs/entities/';
+
+// defines the output file for the serialized content.
+$output_file = '../library/HTMLPurifier/EntityLookup/entities.ser';
+
+// courtesy of a PHP manual comment
+function unichr($dec)
+{
+    if ($dec < 128) {
+        $utf  = chr($dec);
+    } elseif ($dec < 2048) {
+        $utf  = chr(192 + (($dec - ($dec % 64)) / 64));
+        $utf .= chr(128 + ($dec % 64));
+    } else {
+        $utf  = chr(224 + (($dec - ($dec % 4096)) / 4096));
+        $utf .= chr(128 + ((($dec % 4096) - ($dec % 64)) / 64));
+        $utf .= chr(128 + ($dec % 64));
+    }
+    return $utf;
+}
+
+if ( !is_dir($entity_dir) ) exit("Fatal Error: Can't find entity directory.\n");
+if ( file_exists($output_file) ) exit("Fatal Error: output file already exists.\n");
+
+$dh = @opendir($entity_dir);
+if ( !$dh ) exit("Fatal Error: Cannot read entity directory.\n");
+
+$entity_files = array();
+while (($file = readdir($dh)) !== false) {
+    if (@$file[0] === '.') continue;
+    if (substr(strrchr($file, "."), 1) !== 'ent') continue;
+    $entity_files[] = $file;
+}
+closedir($dh);
+
+if ( !$entity_files ) exit("Fatal Error: No entity files to parse.\n");
+
+$entity_table = array();
+$regexp = '/<!ENTITY\s+([A-Za-z0-9]+)\s+"&#(?:38;#)?([0-9]+);">/';
+
+foreach ( $entity_files as $file ) {
+    $contents = file_get_contents($entity_dir . $file);
+    $matches = array();
+    preg_match_all($regexp, $contents, $matches, PREG_SET_ORDER);
+    foreach ($matches as $match) {
+        $entity_table[$match[1]] = unichr($match[2]);
+    }
+}
+
+$output = serialize($entity_table);
+
+$fh = fopen($output_file, 'w');
+fwrite($fh, $output);
+fclose($fh);
+
+echo "Completed successfully.";
+
+// vim: et sw=4 sts=4

htmlpurifier-4.10.0/maintenance/generate-includes.php

@@ -0,0 +1,192 @@
+#!/usr/bin/php
+<?php
+
+chdir(dirname(__FILE__));
+require_once 'common.php';
+require_once '../tests/path2class.func.php';
+require_once '../library/HTMLPurifier/Bootstrap.php';
+assertCli();
+
+/**
+ * @file
+ * Generates an include stub for users who do not want to use the autoloader.
+ * When new files are added to HTML Purifier's main codebase, this file should
+ * be called.
+ */
+
+chdir(dirname(__FILE__) . '/../library/');
+$FS = new FSTools();
+
+$exclude_dirs = array(
+    'HTMLPurifier/Language/',
+    'HTMLPurifier/ConfigSchema/',
+    'HTMLPurifier/Filter/',
+    'HTMLPurifier/Printer/',
+    /* These should be excluded, but need to have ConfigSchema support first
+
+    */
+);
+$exclude_files = array(
+    'HTMLPurifier/Lexer/PEARSax3.php',
+    'HTMLPurifier/Lexer/PH5P.php',
+    'HTMLPurifier/Printer.php',
+);
+
+// Determine what files need to be included:
+echo 'Scanning for files... ';
+$raw_files = $FS->globr('.', '*.php');
+if (!$raw_files) throw new Exception('Did not find any PHP source files');
+$files = array();
+foreach ($raw_files as $file) {
+    $file = substr($file, 2); // rm leading './'
+    if (strncmp('standalone/', $file, 11) === 0) continue; // rm generated files
+    if (substr_count($file, '.') > 1) continue; // rm meta files
+    $ok = true;
+    foreach ($exclude_dirs as $dir) {
+        if (strncmp($dir, $file, strlen($dir)) === 0) {
+            $ok = false;
+            break;
+        }
+    }
+    if (!$ok) continue; // rm excluded directories
+    if (in_array($file, $exclude_files)) continue; // rm excluded files
+    $files[] = $file;
+}
+echo "done!\n";
+
+// Reorder list so that dependencies are included first:
+
+/**
+ * Returns a lookup array of dependencies for a file.
+ *
+ * @note This function expects that format $name extends $parent on one line
+ *
+ * @param string $file
+ *      File to check dependencies of.
+ * @return array
+ *      Lookup array of files the file is dependent on, sorted accordingly.
+ */
+function get_dependency_lookup($file)
+{
+    static $cache = array();
+    if (isset($cache[$file])) return $cache[$file];
+    if (!file_exists($file)) {
+        echo "File doesn't exist: $file\n";
+        return array();
+    }
+    $fh = fopen($file, 'r');
+    $deps = array();
+    while (!feof($fh)) {
+        $line = fgets($fh);
+        if (strncmp('class', $line, 5) === 0) {
+            // The implementation here is fragile and will break if we attempt
+            // to use interfaces. Beware!
+            $arr = explode(' extends ', trim($line, ' {'."\n\r"), 2);
+            if (count($arr) < 2) break;
+            $parent = $arr[1];
+            $dep_file = HTMLPurifier_Bootstrap::getPath($parent);
+            if (!$dep_file) break;
+            $deps[$dep_file] = true;
+            break;
+        }
+    }
+    fclose($fh);
+    foreach (array_keys($deps) as $file) {
+        // Extra dependencies must come *before* base dependencies
+        $deps = get_dependency_lookup($file) + $deps;
+    }
+    $cache[$file] = $deps;
+    return $deps;
+}
+
+/**
+ * Sorts files based on dependencies. This function is lazy and will not
+ * group files with dependencies together; it will merely ensure that a file
+ * is never included before its dependencies are.
+ *
+ * @param $files
+ *      Files array to sort.
+ * @return
+ *      Sorted array ($files is not modified by reference!)
+ */
+function dep_sort($files)
+{
+    $ret = array();
+    $cache = array();
+    foreach ($files as $file) {
+        if (isset($cache[$file])) continue;
+        $deps = get_dependency_lookup($file);
+        foreach (array_keys($deps) as $dep) {
+            if (!isset($cache[$dep])) {
+                $ret[] = $dep;
+                $cache[$dep] = true;
+            }
+        }
+        $cache[$file] = true;
+        $ret[] = $file;
+    }
+    return $ret;
+}
+
+$files = dep_sort($files);
+
+// Build the actual include stub:
+
+$version = trim(file_get_contents('../VERSION'));
+
+// stub
+$php = "<?php
+
+/**
+ * @file
+ * This file was auto-generated by generate-includes.php and includes all of
+ * the core files required by HTML Purifier. Use this if performance is a
+ * primary concern and you are using an opcode cache. PLEASE DO NOT EDIT THIS
+ * FILE, changes will be overwritten the next time the script is run.
+ *
+ * @version $version
+ *
+ * @warning
+ *      You must *not* include any other HTML Purifier files before this file,
+ *      because 'require' not 'require_once' is used.
+ *
+ * @warning
+ *      This file requires that the include path contains the HTML Purifier
+ *      library directory; this is not auto-set.
+ */
+
+";
+
+foreach ($files as $file) {
+    $php .= "require '$file';" . PHP_EOL;
+}
+
+echo "Writing HTMLPurifier.includes.php... ";
+file_put_contents('HTMLPurifier.includes.php', $php);
+echo "done!\n";
+
+$php = "<?php
+
+/**
+ * @file
+ * This file was auto-generated by generate-includes.php and includes all of
+ * the core files required by HTML Purifier. This is a convenience stub that
+ * includes all files using dirname(__FILE__) and require_once. PLEASE DO NOT
+ * EDIT THIS FILE, changes will be overwritten the next time the script is run.
+ *
+ * Changes to include_path are not necessary.
+ */
+
+\$__dir = dirname(__FILE__);
+
+";
+
+foreach ($files as $file) {
+    $php .= "require_once \$__dir . '/$file';" . PHP_EOL;
+}
+
+echo "Writing HTMLPurifier.safe-includes.php... ";
+file_put_contents('HTMLPurifier.safe-includes.php', $php);
+echo "done!\n";
+
+// vim: et sw=4 sts=4

htmlpurifier-4.10.0/maintenance/generate-ph5p-patch.php

@@ -0,0 +1,22 @@
+<?php
+
+/**
+ * @file
+ * This file compares our version of PH5P with Jero's original version, and
+ * generates a patch of the differences. This script should be run whenever
+ * library/HTMLPurifier/Lexer/PH5P.php is modified.
+ */
+
+$orig = realpath(dirname(__FILE__) . '/PH5P.php');
+$new  = realpath(dirname(__FILE__) . '/../library/HTMLPurifier/Lexer/PH5P.php');
+$newt = dirname(__FILE__) . '/PH5P.new.php'; // temporary file
+
+// minor text-processing of new file to get into same format as original
+$new_src = file_get_contents($new);
+$new_src = '<?php' . PHP_EOL . substr($new_src, strpos($new_src, 'class HTML5 {'));
+
+file_put_contents($newt, $new_src);
+shell_exec("diff -u \"$orig\" \"$newt\" > PH5P.patch");
+unlink($newt);
+
+// vim: et sw=4 sts=4

htmlpurifier-4.10.0/maintenance/generate-schema-cache.php

@@ -0,0 +1,45 @@
+#!/usr/bin/php
+<?php
+
+require_once dirname(__FILE__) . '/common.php';
+require_once dirname(__FILE__) . '/../library/HTMLPurifier.auto.php';
+assertCli();
+
+/**
+ * @file
+ * Generates a schema cache file, saving it to
+ * library/HTMLPurifier/ConfigSchema/schema.ser.
+ *
+ * This should be run when new configuration options are added to
+ * HTML Purifier. A cached version is available via the repository
+ * so this does not normally have to be regenerated.
+ *
+ * If you have a directory containing custom configuration schema files,
+ * you can simple add a path to that directory as a parameter to
+ * this, and they will get included.
+ */
+
+$target = dirname(__FILE__) . '/../library/HTMLPurifier/ConfigSchema/schema.ser';
+
+$builder = new HTMLPurifier_ConfigSchema_InterchangeBuilder();
+$interchange = new HTMLPurifier_ConfigSchema_Interchange();
+
+$builder->buildDir($interchange);
+
+$loader = dirname(__FILE__) . '/../config-schema.php';
+if (file_exists($loader)) include $loader;
+foreach ($_SERVER['argv'] as $i => $dir) {
+    if ($i === 0) continue;
+    $builder->buildDir($interchange, realpath($dir));
+}
+
+$interchange->validate();
+
+$schema_builder = new HTMLPurifier_ConfigSchema_Builder_ConfigSchema();
+$schema = $schema_builder->build($interchange);
+
+echo "Saving schema... ";
+file_put_contents($target, serialize($schema));
+echo "done!\n";
+
+// vim: et sw=4 sts=4

htmlpurifier-4.10.0/maintenance/generate-standalone.php

@@ -0,0 +1,159 @@
+#!/usr/bin/php
+<?php
+
+chdir(dirname(__FILE__));
+require_once 'common.php';
+assertCli();
+
+/**
+ * @file
+ * Compiles all of HTML Purifier's library files into one big file
+ * named HTMLPurifier.standalone.php. This is usually called during the
+ * release process.
+ */
+
+/**
+ * Global hash that tracks already loaded includes
+ */
+$GLOBALS['loaded'] = array();
+
+/**
+ * Custom FSTools for this script that overloads some behavior
+ * @warning The overloading of copy() is not necessarily global for
+ *          this script. Watch out!
+ */
+class MergeLibraryFSTools extends FSTools
+{
+    public function copyable($entry)
+    {
+        // Skip hidden files
+        if ($entry[0] == '.') {
+            return false;
+        }
+        return true;
+    }
+    public function copy($source, $dest)
+    {
+        copy_and_remove_includes($source, $dest);
+    }
+}
+$FS = new MergeLibraryFSTools();
+
+/**
+ * Replaces the includes inside PHP source code with the corresponding
+ * source.
+ * @param string $text PHP source code to replace includes from
+ */
+function replace_includes($text)
+{
+    // also remove vim modelines
+    return preg_replace_callback(
+        "/require(?:_once)? ['\"]([^'\"]+)['\"];/",
+        'replace_includes_callback',
+        $text
+    );
+}
+
+/**
+ * Removes leading PHP tags from included files. Assumes that there is
+ * no trailing tag. Also removes vim modelines.
+ * @note This is safe for files that have internal <?php
+ * @param string $text Text to have leading PHP tag from
+ */
+function remove_php_tags($text)
+{
+    $text = preg_replace('#// vim:.+#', '', $text);
+    return substr($text, 5);
+}
+
+/**
+ * Copies the contents of a directory to the standalone directory
+ * @param string $dir Directory to copy
+ */
+function make_dir_standalone($dir)
+{
+    global $FS;
+    return $FS->copyr($dir, 'standalone/' . $dir);
+}
+
+/**
+ * Copies the contents of a file to the standalone directory
+ * @param string $file File to copy
+ */
+function make_file_standalone($file)
+{
+    global $FS;
+    $FS->mkdirr('standalone/' . dirname($file));
+    copy_and_remove_includes($file, 'standalone/' . $file);
+    return true;
+}
+
+/**
+ * Copies a file to another location recursively, if it is a PHP file
+ * remove includes
+ * @param string $file Original file
+ * @param string $sfile New location of file
+ */
+function copy_and_remove_includes($file, $sfile)
+{
+    $contents = file_get_contents($file);
+    if (strrchr($file, '.') === '.php') $contents = replace_includes($contents);
+    return file_put_contents($sfile, $contents);
+}
+
+/**
+ * @param $matches preg_replace_callback matches array, where index 1
+ *        is the filename to include
+ */
+function replace_includes_callback($matches)
+{
+    $file = $matches[1];
+    $preserve = array(
+      // PEAR (external)
+      'XML/HTMLSax3.php' => 1
+    );
+    if (isset($preserve[$file])) {
+        return $matches[0];
+    }
+    if (isset($GLOBALS['loaded'][$file])) return '';
+    $GLOBALS['loaded'][$file] = true;
+    return replace_includes(remove_php_tags(file_get_contents($file)));
+}
+
+echo 'Generating includes file... ';
+shell_exec('php generate-includes.php');
+echo "done!\n";
+
+chdir(dirname(__FILE__) . '/../library/');
+
+echo 'Creating full file...';
+$contents = replace_includes(file_get_contents('HTMLPurifier.includes.php'));
+$contents = str_replace(
+    // Note that bootstrap is now inside the standalone file
+    "define('HTMLPURIFIER_PREFIX', realpath(dirname(__FILE__) . '/..'));",
+    "define('HTMLPURIFIER_PREFIX', dirname(__FILE__) . '/standalone');
+    set_include_path(HTMLPURIFIER_PREFIX . PATH_SEPARATOR . get_include_path());",
+    $contents
+);
+file_put_contents('HTMLPurifier.standalone.php', $contents);
+echo ' done!' . PHP_EOL;
+
+echo 'Creating standalone directory...';
+$FS->rmdirr('standalone'); // ensure a clean copy
+
+// data files
+$FS->mkdirr('standalone/HTMLPurifier/DefinitionCache/Serializer');
+make_file_standalone('HTMLPurifier/EntityLookup/entities.ser');
+make_file_standalone('HTMLPurifier/ConfigSchema/schema.ser');
+
+// non-standard inclusion setup
+make_dir_standalone('HTMLPurifier/ConfigSchema');
+make_dir_standalone('HTMLPurifier/Language');
+make_dir_standalone('HTMLPurifier/Filter');
+make_dir_standalone('HTMLPurifier/Printer');
+make_file_standalone('HTMLPurifier/Printer.php');
+make_file_standalone('HTMLPurifier/Lexer/PH5P.php');
+
+echo ' done!' . PHP_EOL;
+
+// vim: et sw=4 sts=4

htmlpurifier-4.10.0/maintenance/merge-library.php

@@ -0,0 +1,11 @@
+#!/usr/bin/php
+<?php
+
+/**
+ * @file
+ * Deprecated in favor of generate-standalone.php.
+ */
+
+require dirname(__FILE__) . '/generate-standalone.php';
+
+// vim: et sw=4 sts=4

htmlpurifier-4.10.0/maintenance/old-extract-schema.php

@@ -0,0 +1,71 @@
+#!/usr/bin/php
+<?php
+
+chdir(dirname(__FILE__));
+require_once 'common.php';
+assertCli();
+
+echo "Please do not run this script. It is here for historical purposes only.";
+exit;
+
+/**
+ * @file
+ * Extracts all definitions inside a configuration schema
+ * (HTMLPurifier_ConfigSchema) and exports them as plain text files.
+ *
+ * @todo Extract version numbers.
+ */
+
+define('HTMLPURIFIER_SCHEMA_STRICT', true); // description data needs to be collected
+require_once dirname(__FILE__) . '/../library/HTMLPurifier.auto.php';
+
+// We need includes to ensure all HTMLPurifier_ConfigSchema calls are
+// performed.
+require_once 'HTMLPurifier.includes.php';
+
+// Also, these extra files will be necessary.
+require_once 'HTMLPurifier/Filter/ExtractStyleBlocks.php';
+
+/**
+ * Takes a hash and saves its contents to library/HTMLPurifier/ConfigSchema/
+ */
+function saveHash($hash)
+{
+    if ($hash === false) return;
+    $dir = realpath(dirname(__FILE__) . '/../library/HTMLPurifier/ConfigSchema');
+    $name = $hash['ID'] . '.txt';
+    $file = $dir . '/' . $name;
+    if (file_exists($file)) {
+        trigger_error("File already exists; skipped $name");
+        return;
+    }
+    $file = new FSTools_File($file);
+    $file->open('w');
+    $multiline = false;
+    foreach ($hash as $key => $value) {
+        $multiline = $multiline || (strpos($value, "\n") !== false);
+        if ($multiline) {
+            $file->put("--$key--" . PHP_EOL);
+            $file->put(str_replace("\n", PHP_EOL, $value) . PHP_EOL);
+        } else {
+            if ($key == 'ID') {
+                $file->put("$value" . PHP_EOL);
+            } else {
+                $file->put("$key: $value" . PHP_EOL);
+            }
+        }
+    }
+    $file->close();
+}
+
+$schema  = HTMLPurifier_ConfigSchema::instance();
+$adapter = new HTMLPurifier_ConfigSchema_StringHashReverseAdapter($schema);
+
+foreach ($schema->info as $ns => $ns_array) {
+    saveHash($adapter->get($ns));
+    foreach ($ns_array as $dir => $x) {
+        saveHash($adapter->get($ns, $dir));
+    }
+}
+
+// vim: et sw=4 sts=4

htmlpurifier-4.10.0/maintenance/old-remove-require-once.php

@@ -0,0 +1,32 @@
+#!/usr/bin/php
+<?php
+
+chdir(dirname(__FILE__));
+require_once 'common.php';
+assertCli();
+
+echo "Please do not run this script. It is here for historical purposes only.";
+exit;
+
+/**
+ * @file
+ * Removes leading includes from files.
+ *
+ * @note
+ *      This does not remove inline includes; those must be handled manually.
+ */
+
+chdir(dirname(__FILE__) . '/../tests/HTMLPurifier');
+$FS = new FSTools();
+
+$files = $FS->globr('.', '*.php');
+foreach ($files as $file) {
+    if (substr_count(basename($file), '.') > 1) continue;
+    $old_code = file_get_contents($file);
+    $new_code = preg_replace("#^require_once .+[\n\r]*#m", '', $old_code);
+    if ($old_code !== $new_code) {
+        file_put_contents($file, $new_code);
+    }
+}
+
+// vim: et sw=4 sts=4

htmlpurifier-4.10.0/maintenance/old-remove-schema-def.php

@@ -0,0 +1,32 @@
+#!/usr/bin/php
+<?php
+
+chdir(dirname(__FILE__));
+require_once 'common.php';
+assertCli();
+
+echo "Please do not run this script. It is here for historical purposes only.";
+exit;
+
+/**
+ * @file
+ * Removes ConfigSchema function calls from source files.
+ */
+
+chdir(dirname(__FILE__) . '/../library/');
+$FS = new FSTools();
+
+$files = $FS->globr('.', '*.php');
+foreach ($files as $file) {
+    if (substr_count(basename($file), '.') > 1) continue;
+    $old_code = file_get_contents($file);
+    $new_code = preg_replace("#^HTMLPurifier_ConfigSchema::.+?\);[\n\r]*#ms", '', $old_code);
+    if ($old_code !== $new_code) {
+        file_put_contents($file, $new_code);
+    }
+    if (preg_match('#^\s+HTMLPurifier_ConfigSchema::#m', $new_code)) {
+        echo "Indented ConfigSchema call in $file\n";
+    }
+}
+
+// vim: et sw=4 sts=4

htmlpurifier-4.10.0/maintenance/regenerate-docs.sh

@@ -0,0 +1,5 @@
+#!/bin/bash -e
+./compile-doxygen.sh
+cd ../docs
+scp doxygen.tgz htmlpurifier.org:/home/ezyang/htmlpurifier.org
+ssh htmlpurifier.org "cd /home/ezyang/htmlpurifier.org && ./reload-docs.sh"

htmlpurifier-4.10.0/maintenance/remove-trailing-whitespace.php

@@ -0,0 +1,37 @@
+#!/usr/bin/php
+<?php
+
+chdir(dirname(__FILE__));
+require_once 'common.php';
+assertCli();
+
+/**
+ * @file
+ * Removes trailing whitespace from files.
+ */
+
+chdir(dirname(__FILE__) . '/..');
+$FS = new FSTools();
+
+$files = $FS->globr('.', '{,.}*', GLOB_BRACE);
+foreach ($files as $file) {
+    if (
+        !is_file($file) ||
+        prefix_is('./.git', $file) ||
+        prefix_is('./docs/doxygen', $file) ||
+        postfix_is('.ser', $file) ||
+        postfix_is('.tgz', $file) ||
+        postfix_is('.patch', $file) ||
+        postfix_is('.dtd', $file) ||
+        postfix_is('.ent', $file) ||
+        $file == './library/HTMLPurifier/Lexer/PH5P.php' ||
+        $file == './maintenance/PH5P.php'
+    ) continue;
+    $contents = file_get_contents($file);
+    $result = preg_replace('/^(.*?)[ \t]+(\r?)$/m', '\1\2', $contents, -1, $count);
+    if (!$count) continue;
+    echo "$file\n";
+    file_put_contents($file, $result);
+}
+
+// vim: et sw=4 sts=4

htmlpurifier-4.10.0/maintenance/rename-config.php

@@ -0,0 +1,84 @@
+#!/usr/bin/php
+<?php
+
+chdir(dirname(__FILE__));
+require_once 'common.php';
+require_once '../library/HTMLPurifier.auto.php';
+assertCli();
+
+/**
+ * @file
+ * Renames a configuration directive.  This involves renaming the file,
+ * adding an alias, and then regenerating the cache.  You still have to
+ * manually go through and fix any calls to the directive.
+ * @warning This script doesn't handle multi-stringhash files.
+ */
+
+$argv = $_SERVER['argv'];
+if (count($argv) < 3) {
+    echo "Usage: {$argv[0]} OldName NewName\n";
+    exit(1);
+}
+
+chdir('../library/HTMLPurifier/ConfigSchema/schema');
+
+$old = $argv[1];
+$new = $argv[2];
+
+if (!file_exists("$old.txt")) {
+    echo "Cannot move undefined configuration directive $old\n";
+    exit(1);
+}
+
+if ($old === $new) {
+    echo "Attempting to move to self, aborting\n";
+    exit(1);
+}
+
+if (file_exists("$new.txt")) {
+    echo "Cannot move to already defined directive $new\n";
+    exit(1);
+}
+
+$file = "$old.txt";
+$builder = new HTMLPurifier_ConfigSchema_InterchangeBuilder();
+$interchange = new HTMLPurifier_ConfigSchema_Interchange();
+$builder->buildFile($interchange, $file);
+$contents = file_get_contents($file);
+
+if (strpos($contents, "\r\n") !== false) {
+    $nl = "\r\n";
+} elseif (strpos($contents, "\r") !== false) {
+    $nl = "\r";
+} else {
+    $nl = "\n";
+}
+
+// replace name with new name
+$contents = str_replace($old, $new, $contents);
+
+if ($interchange->directives[$old]->aliases) {
+    $pos_alias = strpos($contents, 'ALIASES:');
+    $pos_ins = strpos($contents, $nl, $pos_alias);
+    if ($pos_ins === false) $pos_ins = strlen($contents);
+    $contents =
+        substr($contents, 0, $pos_ins) . ", $old" . substr($contents, $pos_ins);
+    file_put_contents($file, $contents);
+} else {
+    $lines = explode($nl, $contents);
+    $insert = false;
+    foreach ($lines as $n => $line) {
+        if (strncmp($line, '--', 2) === 0) {
+            $insert = $n;
+            break;
+        }
+    }
+    if (!$insert) {
+        $lines[] = "ALIASES: $old";
+    } else {
+        array_splice($lines, $insert, 0, "ALIASES: $old");
+    }
+    file_put_contents($file, implode($nl, $lines));
+}
+
+rename("$old.txt", "$new.txt") || exit(1);

htmlpurifier-4.10.0/maintenance/update-config.php

@@ -0,0 +1,34 @@
+#!/usr/bin/php
+<?php
+
+chdir(dirname(__FILE__));
+require_once 'common.php';
+assertCli();
+
+/**
+ * @file
+ * Converts all instances of $config->set and $config->get to the new
+ * format, as described by docs/dev-config-bcbreaks.txt
+ */
+
+$FS = new FSTools();
+chdir(dirname(__FILE__) . '/..');
+$raw_files = $FS->globr('.', '*.php');
+foreach ($raw_files as $file) {
+    $file = substr($file, 2); // rm leading './'
+    if (strpos($file, 'library/standalone/') === 0) continue;
+    if (strpos($file, 'maintenance/update-config.php') === 0) continue;
+    if (strpos($file, 'test-settings.php') === 0) continue;
+    if (substr_count($file, '.') > 1) continue; // rm meta files
+    // process the file
+    $contents = file_get_contents($file);
+    $contents = preg_replace(
+        "#config->(set|get)\('(.+?)', '(.+?)'#",
+        "config->\\1('\\2.\\3'",
+        $contents
+    );
+    if ($contents === '') continue;
+    file_put_contents($file, $contents);
+}
+
+// vim: et sw=4 sts=4