From f45a286b8d62f710b519a98c7d4b75a0c34d5d10 Mon Sep 17 00:00:00 2001
From: Andrew Dolgov <fox@bah.org.ru>
Date: Mon, 22 Jun 2009 13:56:49 +0400
Subject: strip_tags_long: use htmlpurifier to properly reformat html content

---
 .../library/HTMLPurifier/Token/Comment.php         | 22 +++++++++
 .../library/HTMLPurifier/Token/Empty.php           | 11 +++++
 .../library/HTMLPurifier/Token/End.php             | 19 ++++++++
 .../library/HTMLPurifier/Token/Start.php           | 11 +++++
 .../library/HTMLPurifier/Token/Tag.php             | 56 ++++++++++++++++++++++
 .../library/HTMLPurifier/Token/Text.php            | 33 +++++++++++++
 6 files changed, 152 insertions(+)
 create mode 100755 lib/htmlpurifier/library/HTMLPurifier/Token/Comment.php
 create mode 100755 lib/htmlpurifier/library/HTMLPurifier/Token/Empty.php
 create mode 100755 lib/htmlpurifier/library/HTMLPurifier/Token/End.php
 create mode 100755 lib/htmlpurifier/library/HTMLPurifier/Token/Start.php
 create mode 100755 lib/htmlpurifier/library/HTMLPurifier/Token/Tag.php
 create mode 100755 lib/htmlpurifier/library/HTMLPurifier/Token/Text.php

(limited to 'lib/htmlpurifier/library/HTMLPurifier/Token')

diff --git a/lib/htmlpurifier/library/HTMLPurifier/Token/Comment.php b/lib/htmlpurifier/library/HTMLPurifier/Token/Comment.php
new file mode 100755
index 000000000..dc6bdcabb
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/Token/Comment.php
@@ -0,0 +1,22 @@
+<?php
+
+/**
+ * Concrete comment token class. Generally will be ignored.
+ */
+class HTMLPurifier_Token_Comment extends HTMLPurifier_Token
+{
+    public $data; /**< Character data within comment. */
+    public $is_whitespace = true;
+    /**
+     * Transparent constructor.
+     *
+     * @param $data String comment data.
+     */
+    public function __construct($data, $line = null, $col = null) {
+        $this->data = $data;
+        $this->line = $line;
+        $this->col  = $col;
+    }
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/Token/Empty.php b/lib/htmlpurifier/library/HTMLPurifier/Token/Empty.php
new file mode 100755
index 000000000..2a82b47ad
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/Token/Empty.php
@@ -0,0 +1,11 @@
+<?php
+
+/**
+ * Concrete empty token class.
+ */
+class HTMLPurifier_Token_Empty extends HTMLPurifier_Token_Tag
+{
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/Token/End.php b/lib/htmlpurifier/library/HTMLPurifier/Token/End.php
new file mode 100755
index 000000000..353e79daf
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/Token/End.php
@@ -0,0 +1,19 @@
+<?php
+
+/**
+ * Concrete end token class.
+ *
+ * @warning This class accepts attributes even though end tags cannot. This
+ * is for optimization reasons, as under normal circumstances, the Lexers
+ * do not pass attributes.
+ */
+class HTMLPurifier_Token_End extends HTMLPurifier_Token_Tag
+{
+    /**
+     * Token that started this node. Added by MakeWellFormed. Please
+     * do not edit this!
+     */
+    public $start;
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/Token/Start.php b/lib/htmlpurifier/library/HTMLPurifier/Token/Start.php
new file mode 100755
index 000000000..e0e14fc62
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/Token/Start.php
@@ -0,0 +1,11 @@
+<?php
+
+/**
+ * Concrete start token class.
+ */
+class HTMLPurifier_Token_Start extends HTMLPurifier_Token_Tag
+{
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/Token/Tag.php b/lib/htmlpurifier/library/HTMLPurifier/Token/Tag.php
new file mode 100755
index 000000000..798be028e
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/Token/Tag.php
@@ -0,0 +1,56 @@
+<?php
+
+/**
+ * Abstract class of a tag token (start, end or empty), and its behavior.
+ */
+class HTMLPurifier_Token_Tag extends HTMLPurifier_Token
+{
+    /**
+     * Static bool marker that indicates the class is a tag.
+     *
+     * This allows us to check objects with <tt>!empty($obj->is_tag)</tt>
+     * without having to use a function call <tt>is_a()</tt>.
+     */
+    public $is_tag = true;
+
+    /**
+     * The lower-case name of the tag, like 'a', 'b' or 'blockquote'.
+     *
+     * @note Strictly speaking, XML tags are case sensitive, so we shouldn't
+     * be lower-casing them, but these tokens cater to HTML tags, which are
+     * insensitive.
+     */
+    public $name;
+
+    /**
+     * Associative array of the tag's attributes.
+     */
+    public $attr = array();
+
+    /**
+     * Non-overloaded constructor, which lower-cases passed tag name.
+     *
+     * @param $name String name.
+     * @param $attr Associative array of attributes.
+     */
+    public function __construct($name, $attr = array(), $line = null, $col = null) {
+        $this->name = ctype_lower($name) ? $name : strtolower($name);
+        foreach ($attr as $key => $value) {
+            // normalization only necessary when key is not lowercase
+            if (!ctype_lower($key)) {
+                $new_key = strtolower($key);
+                if (!isset($attr[$new_key])) {
+                    $attr[$new_key] = $attr[$key];
+                }
+                if ($new_key !== $key) {
+                    unset($attr[$key]);
+                }
+            }
+        }
+        $this->attr = $attr;
+        $this->line = $line;
+        $this->col  = $col;
+    }
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/Token/Text.php b/lib/htmlpurifier/library/HTMLPurifier/Token/Text.php
new file mode 100755
index 000000000..82efd823d
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/Token/Text.php
@@ -0,0 +1,33 @@
+<?php
+
+/**
+ * Concrete text token class.
+ *
+ * Text tokens comprise of regular parsed character data (PCDATA) and raw
+ * character data (from the CDATA sections). Internally, their
+ * data is parsed with all entities expanded. Surprisingly, the text token
+ * does have a "tag name" called #PCDATA, which is how the DTD represents it
+ * in permissible child nodes.
+ */
+class HTMLPurifier_Token_Text extends HTMLPurifier_Token
+{
+
+    public $name = '#PCDATA'; /**< PCDATA tag name compatible with DTD. */
+    public $data; /**< Parsed character data of text. */
+    public $is_whitespace; /**< Bool indicating if node is whitespace. */
+
+    /**
+     * Constructor, accepts data and determines if it is whitespace.
+     *
+     * @param $data String parsed character data.
+     */
+    public function __construct($data, $line = null, $col = null) {
+        $this->data = $data;
+        $this->is_whitespace = ctype_space($data);
+        $this->line = $line;
+        $this->col  = $col;
+    }
+
+}
+
+// vim: et sw=4 sts=4
-- 
cgit v1.2.3