aboutsummaryrefslogtreecommitdiffstats
path: root/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS
diff options
context:
space:
mode:
Diffstat (limited to 'lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS')
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/AlphaValue.php21
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Background.php87
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/BackgroundPosition.php133
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Border.php43
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Color.php78
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Composite.php38
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/DenyElementDecorator.php28
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Filter.php54
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Font.php149
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/FontFamily.php197
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Ident.php24
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/ImportantDecorator.php40
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Length.php47
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/ListStyle.php78
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Multiple.php58
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Number.php69
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Percentage.php40
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/TextDecoration.php38
-rw-r--r--lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/URI.php61
19 files changed, 1283 insertions, 0 deletions
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/AlphaValue.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/AlphaValue.php
new file mode 100644
index 000000000..292c040d4
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/AlphaValue.php
@@ -0,0 +1,21 @@
+<?php
+
+class HTMLPurifier_AttrDef_CSS_AlphaValue extends HTMLPurifier_AttrDef_CSS_Number
+{
+
+ public function __construct() {
+ parent::__construct(false); // opacity is non-negative, but we will clamp it
+ }
+
+ public function validate($number, $config, $context) {
+ $result = parent::validate($number, $config, $context);
+ if ($result === false) return $result;
+ $float = (float) $result;
+ if ($float < 0.0) $result = '0';
+ if ($float > 1.0) $result = '1';
+ return $result;
+ }
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Background.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Background.php
new file mode 100644
index 000000000..3a3d20cd6
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Background.php
@@ -0,0 +1,87 @@
+<?php
+
+/**
+ * Validates shorthand CSS property background.
+ * @warning Does not support url tokens that have internal spaces.
+ */
+class HTMLPurifier_AttrDef_CSS_Background extends HTMLPurifier_AttrDef
+{
+
+ /**
+ * Local copy of component validators.
+ * @note See HTMLPurifier_AttrDef_Font::$info for a similar impl.
+ */
+ protected $info;
+
+ public function __construct($config) {
+ $def = $config->getCSSDefinition();
+ $this->info['background-color'] = $def->info['background-color'];
+ $this->info['background-image'] = $def->info['background-image'];
+ $this->info['background-repeat'] = $def->info['background-repeat'];
+ $this->info['background-attachment'] = $def->info['background-attachment'];
+ $this->info['background-position'] = $def->info['background-position'];
+ }
+
+ public function validate($string, $config, $context) {
+
+ // regular pre-processing
+ $string = $this->parseCDATA($string);
+ if ($string === '') return false;
+
+ // munge rgb() decl if necessary
+ $string = $this->mungeRgb($string);
+
+ // assumes URI doesn't have spaces in it
+ $bits = explode(' ', strtolower($string)); // bits to process
+
+ $caught = array();
+ $caught['color'] = false;
+ $caught['image'] = false;
+ $caught['repeat'] = false;
+ $caught['attachment'] = false;
+ $caught['position'] = false;
+
+ $i = 0; // number of catches
+ $none = false;
+
+ foreach ($bits as $bit) {
+ if ($bit === '') continue;
+ foreach ($caught as $key => $status) {
+ if ($key != 'position') {
+ if ($status !== false) continue;
+ $r = $this->info['background-' . $key]->validate($bit, $config, $context);
+ } else {
+ $r = $bit;
+ }
+ if ($r === false) continue;
+ if ($key == 'position') {
+ if ($caught[$key] === false) $caught[$key] = '';
+ $caught[$key] .= $r . ' ';
+ } else {
+ $caught[$key] = $r;
+ }
+ $i++;
+ break;
+ }
+ }
+
+ if (!$i) return false;
+ if ($caught['position'] !== false) {
+ $caught['position'] = $this->info['background-position']->
+ validate($caught['position'], $config, $context);
+ }
+
+ $ret = array();
+ foreach ($caught as $value) {
+ if ($value === false) continue;
+ $ret[] = $value;
+ }
+
+ if (empty($ret)) return false;
+ return implode(' ', $ret);
+
+ }
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/BackgroundPosition.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/BackgroundPosition.php
new file mode 100644
index 000000000..fae82eaec
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/BackgroundPosition.php
@@ -0,0 +1,133 @@
+<?php
+
+/* W3C says:
+ [ // adjective and number must be in correct order, even if
+ // you could switch them without introducing ambiguity.
+ // some browsers support that syntax
+ [
+ <percentage> | <length> | left | center | right
+ ]
+ [
+ <percentage> | <length> | top | center | bottom
+ ]?
+ ] |
+ [ // this signifies that the vertical and horizontal adjectives
+ // can be arbitrarily ordered, however, there can only be two,
+ // one of each, or none at all
+ [
+ left | center | right
+ ] ||
+ [
+ top | center | bottom
+ ]
+ ]
+ top, left = 0%
+ center, (none) = 50%
+ bottom, right = 100%
+*/
+
+/* QuirksMode says:
+ keyword + length/percentage must be ordered correctly, as per W3C
+
+ Internet Explorer and Opera, however, support arbitrary ordering. We
+ should fix it up.
+
+ Minor issue though, not strictly necessary.
+*/
+
+// control freaks may appreciate the ability to convert these to
+// percentages or something, but it's not necessary
+
+/**
+ * Validates the value of background-position.
+ */
+class HTMLPurifier_AttrDef_CSS_BackgroundPosition extends HTMLPurifier_AttrDef
+{
+
+ protected $length;
+ protected $percentage;
+
+ public function __construct() {
+ $this->length = new HTMLPurifier_AttrDef_CSS_Length();
+ $this->percentage = new HTMLPurifier_AttrDef_CSS_Percentage();
+ }
+
+ public function validate($string, $config, $context) {
+ $string = $this->parseCDATA($string);
+ $bits = explode(' ', $string);
+
+ $keywords = array();
+ $keywords['h'] = false; // left, right
+ $keywords['v'] = false; // top, bottom
+ $keywords['ch'] = false; // center (first word)
+ $keywords['cv'] = false; // center (second word)
+ $measures = array();
+
+ $i = 0;
+
+ $lookup = array(
+ 'top' => 'v',
+ 'bottom' => 'v',
+ 'left' => 'h',
+ 'right' => 'h',
+ 'center' => 'c'
+ );
+
+ foreach ($bits as $bit) {
+ if ($bit === '') continue;
+
+ // test for keyword
+ $lbit = ctype_lower($bit) ? $bit : strtolower($bit);
+ if (isset($lookup[$lbit])) {
+ $status = $lookup[$lbit];
+ if ($status == 'c') {
+ if ($i == 0) {
+ $status = 'ch';
+ } else {
+ $status = 'cv';
+ }
+ }
+ $keywords[$status] = $lbit;
+ $i++;
+ }
+
+ // test for length
+ $r = $this->length->validate($bit, $config, $context);
+ if ($r !== false) {
+ $measures[] = $r;
+ $i++;
+ }
+
+ // test for percentage
+ $r = $this->percentage->validate($bit, $config, $context);
+ if ($r !== false) {
+ $measures[] = $r;
+ $i++;
+ }
+
+ }
+
+ if (!$i) return false; // no valid values were caught
+
+ $ret = array();
+
+ // first keyword
+ if ($keywords['h']) $ret[] = $keywords['h'];
+ elseif ($keywords['ch']) {
+ $ret[] = $keywords['ch'];
+ $keywords['cv'] = false; // prevent re-use: center = center center
+ }
+ elseif (count($measures)) $ret[] = array_shift($measures);
+
+ if ($keywords['v']) $ret[] = $keywords['v'];
+ elseif ($keywords['cv']) $ret[] = $keywords['cv'];
+ elseif (count($measures)) $ret[] = array_shift($measures);
+
+ if (empty($ret)) return false;
+ return implode(' ', $ret);
+
+ }
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Border.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Border.php
new file mode 100644
index 000000000..42a1d1b4a
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Border.php
@@ -0,0 +1,43 @@
+<?php
+
+/**
+ * Validates the border property as defined by CSS.
+ */
+class HTMLPurifier_AttrDef_CSS_Border extends HTMLPurifier_AttrDef
+{
+
+ /**
+ * Local copy of properties this property is shorthand for.
+ */
+ protected $info = array();
+
+ public function __construct($config) {
+ $def = $config->getCSSDefinition();
+ $this->info['border-width'] = $def->info['border-width'];
+ $this->info['border-style'] = $def->info['border-style'];
+ $this->info['border-top-color'] = $def->info['border-top-color'];
+ }
+
+ public function validate($string, $config, $context) {
+ $string = $this->parseCDATA($string);
+ $string = $this->mungeRgb($string);
+ $bits = explode(' ', $string);
+ $done = array(); // segments we've finished
+ $ret = ''; // return value
+ foreach ($bits as $bit) {
+ foreach ($this->info as $propname => $validator) {
+ if (isset($done[$propname])) continue;
+ $r = $validator->validate($bit, $config, $context);
+ if ($r !== false) {
+ $ret .= $r . ' ';
+ $done[$propname] = true;
+ break;
+ }
+ }
+ }
+ return rtrim($ret);
+ }
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Color.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Color.php
new file mode 100644
index 000000000..07f95a671
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Color.php
@@ -0,0 +1,78 @@
+<?php
+
+/**
+ * Validates Color as defined by CSS.
+ */
+class HTMLPurifier_AttrDef_CSS_Color extends HTMLPurifier_AttrDef
+{
+
+ public function validate($color, $config, $context) {
+
+ static $colors = null;
+ if ($colors === null) $colors = $config->get('Core.ColorKeywords');
+
+ $color = trim($color);
+ if ($color === '') return false;
+
+ $lower = strtolower($color);
+ if (isset($colors[$lower])) return $colors[$lower];
+
+ if (strpos($color, 'rgb(') !== false) {
+ // rgb literal handling
+ $length = strlen($color);
+ if (strpos($color, ')') !== $length - 1) return false;
+ $triad = substr($color, 4, $length - 4 - 1);
+ $parts = explode(',', $triad);
+ if (count($parts) !== 3) return false;
+ $type = false; // to ensure that they're all the same type
+ $new_parts = array();
+ foreach ($parts as $part) {
+ $part = trim($part);
+ if ($part === '') return false;
+ $length = strlen($part);
+ if ($part[$length - 1] === '%') {
+ // handle percents
+ if (!$type) {
+ $type = 'percentage';
+ } elseif ($type !== 'percentage') {
+ return false;
+ }
+ $num = (float) substr($part, 0, $length - 1);
+ if ($num < 0) $num = 0;
+ if ($num > 100) $num = 100;
+ $new_parts[] = "$num%";
+ } else {
+ // handle integers
+ if (!$type) {
+ $type = 'integer';
+ } elseif ($type !== 'integer') {
+ return false;
+ }
+ $num = (int) $part;
+ if ($num < 0) $num = 0;
+ if ($num > 255) $num = 255;
+ $new_parts[] = (string) $num;
+ }
+ }
+ $new_triad = implode(',', $new_parts);
+ $color = "rgb($new_triad)";
+ } else {
+ // hexadecimal handling
+ if ($color[0] === '#') {
+ $hex = substr($color, 1);
+ } else {
+ $hex = $color;
+ $color = '#' . $color;
+ }
+ $length = strlen($hex);
+ if ($length !== 3 && $length !== 6) return false;
+ if (!ctype_xdigit($hex)) return false;
+ }
+
+ return $color;
+
+ }
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Composite.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Composite.php
new file mode 100644
index 000000000..de1289cba
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Composite.php
@@ -0,0 +1,38 @@
+<?php
+
+/**
+ * Allows multiple validators to attempt to validate attribute.
+ *
+ * Composite is just what it sounds like: a composite of many validators.
+ * This means that multiple HTMLPurifier_AttrDef objects will have a whack
+ * at the string. If one of them passes, that's what is returned. This is
+ * especially useful for CSS values, which often are a choice between
+ * an enumerated set of predefined values or a flexible data type.
+ */
+class HTMLPurifier_AttrDef_CSS_Composite extends HTMLPurifier_AttrDef
+{
+
+ /**
+ * List of HTMLPurifier_AttrDef objects that may process strings
+ * @todo Make protected
+ */
+ public $defs;
+
+ /**
+ * @param $defs List of HTMLPurifier_AttrDef objects
+ */
+ public function __construct($defs) {
+ $this->defs = $defs;
+ }
+
+ public function validate($string, $config, $context) {
+ foreach ($this->defs as $i => $def) {
+ $result = $this->defs[$i]->validate($string, $config, $context);
+ if ($result !== false) return $result;
+ }
+ return false;
+ }
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/DenyElementDecorator.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/DenyElementDecorator.php
new file mode 100644
index 000000000..6599c5b2d
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/DenyElementDecorator.php
@@ -0,0 +1,28 @@
+<?php
+
+/**
+ * Decorator which enables CSS properties to be disabled for specific elements.
+ */
+class HTMLPurifier_AttrDef_CSS_DenyElementDecorator extends HTMLPurifier_AttrDef
+{
+ public $def, $element;
+
+ /**
+ * @param $def Definition to wrap
+ * @param $element Element to deny
+ */
+ public function __construct($def, $element) {
+ $this->def = $def;
+ $this->element = $element;
+ }
+ /**
+ * Checks if CurrentToken is set and equal to $this->element
+ */
+ public function validate($string, $config, $context) {
+ $token = $context->get('CurrentToken', true);
+ if ($token && $token->name == $this->element) return false;
+ return $this->def->validate($string, $config, $context);
+ }
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Filter.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Filter.php
new file mode 100644
index 000000000..147894b86
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Filter.php
@@ -0,0 +1,54 @@
+<?php
+
+/**
+ * Microsoft's proprietary filter: CSS property
+ * @note Currently supports the alpha filter. In the future, this will
+ * probably need an extensible framework
+ */
+class HTMLPurifier_AttrDef_CSS_Filter extends HTMLPurifier_AttrDef
+{
+
+ protected $intValidator;
+
+ public function __construct() {
+ $this->intValidator = new HTMLPurifier_AttrDef_Integer();
+ }
+
+ public function validate($value, $config, $context) {
+ $value = $this->parseCDATA($value);
+ if ($value === 'none') return $value;
+ // if we looped this we could support multiple filters
+ $function_length = strcspn($value, '(');
+ $function = trim(substr($value, 0, $function_length));
+ if ($function !== 'alpha' &&
+ $function !== 'Alpha' &&
+ $function !== 'progid:DXImageTransform.Microsoft.Alpha'
+ ) return false;
+ $cursor = $function_length + 1;
+ $parameters_length = strcspn($value, ')', $cursor);
+ $parameters = substr($value, $cursor, $parameters_length);
+ $params = explode(',', $parameters);
+ $ret_params = array();
+ $lookup = array();
+ foreach ($params as $param) {
+ list($key, $value) = explode('=', $param);
+ $key = trim($key);
+ $value = trim($value);
+ if (isset($lookup[$key])) continue;
+ if ($key !== 'opacity') continue;
+ $value = $this->intValidator->validate($value, $config, $context);
+ if ($value === false) continue;
+ $int = (int) $value;
+ if ($int > 100) $value = '100';
+ if ($int < 0) $value = '0';
+ $ret_params[] = "$key=$value";
+ $lookup[$key] = true;
+ }
+ $ret_parameters = implode(',', $ret_params);
+ $ret_function = "$function($ret_parameters)";
+ return $ret_function;
+ }
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Font.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Font.php
new file mode 100644
index 000000000..699ee0b70
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Font.php
@@ -0,0 +1,149 @@
+<?php
+
+/**
+ * Validates shorthand CSS property font.
+ */
+class HTMLPurifier_AttrDef_CSS_Font extends HTMLPurifier_AttrDef
+{
+
+ /**
+ * Local copy of component validators.
+ *
+ * @note If we moved specific CSS property definitions to their own
+ * classes instead of having them be assembled at run time by
+ * CSSDefinition, this wouldn't be necessary. We'd instantiate
+ * our own copies.
+ */
+ protected $info = array();
+
+ public function __construct($config) {
+ $def = $config->getCSSDefinition();
+ $this->info['font-style'] = $def->info['font-style'];
+ $this->info['font-variant'] = $def->info['font-variant'];
+ $this->info['font-weight'] = $def->info['font-weight'];
+ $this->info['font-size'] = $def->info['font-size'];
+ $this->info['line-height'] = $def->info['line-height'];
+ $this->info['font-family'] = $def->info['font-family'];
+ }
+
+ public function validate($string, $config, $context) {
+
+ static $system_fonts = array(
+ 'caption' => true,
+ 'icon' => true,
+ 'menu' => true,
+ 'message-box' => true,
+ 'small-caption' => true,
+ 'status-bar' => true
+ );
+
+ // regular pre-processing
+ $string = $this->parseCDATA($string);
+ if ($string === '') return false;
+
+ // check if it's one of the keywords
+ $lowercase_string = strtolower($string);
+ if (isset($system_fonts[$lowercase_string])) {
+ return $lowercase_string;
+ }
+
+ $bits = explode(' ', $string); // bits to process
+ $stage = 0; // this indicates what we're looking for
+ $caught = array(); // which stage 0 properties have we caught?
+ $stage_1 = array('font-style', 'font-variant', 'font-weight');
+ $final = ''; // output
+
+ for ($i = 0, $size = count($bits); $i < $size; $i++) {
+ if ($bits[$i] === '') continue;
+ switch ($stage) {
+
+ // attempting to catch font-style, font-variant or font-weight
+ case 0:
+ foreach ($stage_1 as $validator_name) {
+ if (isset($caught[$validator_name])) continue;
+ $r = $this->info[$validator_name]->validate(
+ $bits[$i], $config, $context);
+ if ($r !== false) {
+ $final .= $r . ' ';
+ $caught[$validator_name] = true;
+ break;
+ }
+ }
+ // all three caught, continue on
+ if (count($caught) >= 3) $stage = 1;
+ if ($r !== false) break;
+
+ // attempting to catch font-size and perhaps line-height
+ case 1:
+ $found_slash = false;
+ if (strpos($bits[$i], '/') !== false) {
+ list($font_size, $line_height) =
+ explode('/', $bits[$i]);
+ if ($line_height === '') {
+ // ooh, there's a space after the slash!
+ $line_height = false;
+ $found_slash = true;
+ }
+ } else {
+ $font_size = $bits[$i];
+ $line_height = false;
+ }
+ $r = $this->info['font-size']->validate(
+ $font_size, $config, $context);
+ if ($r !== false) {
+ $final .= $r;
+ // attempt to catch line-height
+ if ($line_height === false) {
+ // we need to scroll forward
+ for ($j = $i + 1; $j < $size; $j++) {
+ if ($bits[$j] === '') continue;
+ if ($bits[$j] === '/') {
+ if ($found_slash) {
+ return false;
+ } else {
+ $found_slash = true;
+ continue;
+ }
+ }
+ $line_height = $bits[$j];
+ break;
+ }
+ } else {
+ // slash already found
+ $found_slash = true;
+ $j = $i;
+ }
+ if ($found_slash) {
+ $i = $j;
+ $r = $this->info['line-height']->validate(
+ $line_height, $config, $context);
+ if ($r !== false) {
+ $final .= '/' . $r;
+ }
+ }
+ $final .= ' ';
+ $stage = 2;
+ break;
+ }
+ return false;
+
+ // attempting to catch font-family
+ case 2:
+ $font_family =
+ implode(' ', array_slice($bits, $i, $size - $i));
+ $r = $this->info['font-family']->validate(
+ $font_family, $config, $context);
+ if ($r !== false) {
+ $final .= $r . ' ';
+ // processing completed successfully
+ return rtrim($final);
+ }
+ return false;
+ }
+ }
+ return false;
+ }
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/FontFamily.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/FontFamily.php
new file mode 100644
index 000000000..0d9a4e12c
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/FontFamily.php
@@ -0,0 +1,197 @@
+<?php
+
+/**
+ * Validates a font family list according to CSS spec
+ */
+class HTMLPurifier_AttrDef_CSS_FontFamily extends HTMLPurifier_AttrDef
+{
+
+ protected $mask = null;
+
+ public function __construct() {
+ $this->mask = '- ';
+ for ($c = 'a'; $c <= 'z'; $c++) $this->mask .= $c;
+ for ($c = 'A'; $c <= 'Z'; $c++) $this->mask .= $c;
+ for ($c = '0'; $c <= '9'; $c++) $this->mask .= $c; // cast-y, but should be fine
+ // special bytes used by UTF-8
+ for ($i = 0x80; $i <= 0xFF; $i++) {
+ // We don't bother excluding invalid bytes in this range,
+ // because the our restriction of well-formed UTF-8 will
+ // prevent these from ever occurring.
+ $this->mask .= chr($i);
+ }
+
+ /*
+ PHP's internal strcspn implementation is
+ O(length of string * length of mask), making it inefficient
+ for large masks. However, it's still faster than
+ preg_match 8)
+ for (p = s1;;) {
+ spanp = s2;
+ do {
+ if (*spanp == c || p == s1_end) {
+ return p - s1;
+ }
+ } while (spanp++ < (s2_end - 1));
+ c = *++p;
+ }
+ */
+ // possible optimization: invert the mask.
+ }
+
+ public function validate($string, $config, $context) {
+ static $generic_names = array(
+ 'serif' => true,
+ 'sans-serif' => true,
+ 'monospace' => true,
+ 'fantasy' => true,
+ 'cursive' => true
+ );
+ $allowed_fonts = $config->get('CSS.AllowedFonts');
+
+ // assume that no font names contain commas in them
+ $fonts = explode(',', $string);
+ $final = '';
+ foreach($fonts as $font) {
+ $font = trim($font);
+ if ($font === '') continue;
+ // match a generic name
+ if (isset($generic_names[$font])) {
+ if ($allowed_fonts === null || isset($allowed_fonts[$font])) {
+ $final .= $font . ', ';
+ }
+ continue;
+ }
+ // match a quoted name
+ if ($font[0] === '"' || $font[0] === "'") {
+ $length = strlen($font);
+ if ($length <= 2) continue;
+ $quote = $font[0];
+ if ($font[$length - 1] !== $quote) continue;
+ $font = substr($font, 1, $length - 2);
+ }
+
+ $font = $this->expandCSSEscape($font);
+
+ // $font is a pure representation of the font name
+
+ if ($allowed_fonts !== null && !isset($allowed_fonts[$font])) {
+ continue;
+ }
+
+ if (ctype_alnum($font) && $font !== '') {
+ // very simple font, allow it in unharmed
+ $final .= $font . ', ';
+ continue;
+ }
+
+ // bugger out on whitespace. form feed (0C) really
+ // shouldn't show up regardless
+ $font = str_replace(array("\n", "\t", "\r", "\x0C"), ' ', $font);
+
+ // Here, there are various classes of characters which need
+ // to be treated differently:
+ // - Alphanumeric characters are essentially safe. We
+ // handled these above.
+ // - Spaces require quoting, though most parsers will do
+ // the right thing if there aren't any characters that
+ // can be misinterpreted
+ // - Dashes rarely occur, but they fairly unproblematic
+ // for parsing/rendering purposes.
+ // The above characters cover the majority of Western font
+ // names.
+ // - Arbitrary Unicode characters not in ASCII. Because
+ // most parsers give little thought to Unicode, treatment
+ // of these codepoints is basically uniform, even for
+ // punctuation-like codepoints. These characters can
+ // show up in non-Western pages and are supported by most
+ // major browsers, for example: "MS 明朝" is a
+ // legitimate font-name
+ // <http://ja.wikipedia.org/wiki/MS_明朝>. See
+ // the CSS3 spec for more examples:
+ // <http://www.w3.org/TR/2011/WD-css3-fonts-20110324/localizedfamilynames.png>
+ // You can see live samples of these on the Internet:
+ // <http://www.google.co.jp/search?q=font-family+MS+明朝|ゴシック>
+ // However, most of these fonts have ASCII equivalents:
+ // for example, 'MS Mincho', and it's considered
+ // professional to use ASCII font names instead of
+ // Unicode font names. Thanks Takeshi Terada for
+ // providing this information.
+ // The following characters, to my knowledge, have not been
+ // used to name font names.
+ // - Single quote. While theoretically you might find a
+ // font name that has a single quote in its name (serving
+ // as an apostrophe, e.g. Dave's Scribble), I haven't
+ // been able to find any actual examples of this.
+ // Internet Explorer's cssText translation (which I
+ // believe is invoked by innerHTML) normalizes any
+ // quoting to single quotes, and fails to escape single
+ // quotes. (Note that this is not IE's behavior for all
+ // CSS properties, just some sort of special casing for
+ // font-family). So a single quote *cannot* be used
+ // safely in the font-family context if there will be an
+ // innerHTML/cssText translation. Note that Firefox 3.x
+ // does this too.
+ // - Double quote. In IE, these get normalized to
+ // single-quotes, no matter what the encoding. (Fun
+ // fact, in IE8, the 'content' CSS property gained
+ // support, where they special cased to preserve encoded
+ // double quotes, but still translate unadorned double
+ // quotes into single quotes.) So, because their
+ // fixpoint behavior is identical to single quotes, they
+ // cannot be allowed either. Firefox 3.x displays
+ // single-quote style behavior.
+ // - Backslashes are reduced by one (so \\ -> \) every
+ // iteration, so they cannot be used safely. This shows
+ // up in IE7, IE8 and FF3
+ // - Semicolons, commas and backticks are handled properly.
+ // - The rest of the ASCII punctuation is handled properly.
+ // We haven't checked what browsers do to unadorned
+ // versions, but this is not important as long as the
+ // browser doesn't /remove/ surrounding quotes (as IE does
+ // for HTML).
+ //
+ // With these results in hand, we conclude that there are
+ // various levels of safety:
+ // - Paranoid: alphanumeric, spaces and dashes(?)
+ // - International: Paranoid + non-ASCII Unicode
+ // - Edgy: Everything except quotes, backslashes
+ // - NoJS: Standards compliance, e.g. sod IE. Note that
+ // with some judicious character escaping (since certain
+ // types of escaping doesn't work) this is theoretically
+ // OK as long as innerHTML/cssText is not called.
+ // We believe that international is a reasonable default
+ // (that we will implement now), and once we do more
+ // extensive research, we may feel comfortable with dropping
+ // it down to edgy.
+
+ // Edgy: alphanumeric, spaces, dashes and Unicode. Use of
+ // str(c)spn assumes that the string was already well formed
+ // Unicode (which of course it is).
+ if (strspn($font, $this->mask) !== strlen($font)) {
+ continue;
+ }
+
+ // Historical:
+ // In the absence of innerHTML/cssText, these ugly
+ // transforms don't pose a security risk (as \\ and \"
+ // might--these escapes are not supported by most browsers).
+ // We could try to be clever and use single-quote wrapping
+ // when there is a double quote present, but I have choosen
+ // not to implement that. (NOTE: you can reduce the amount
+ // of escapes by one depending on what quoting style you use)
+ // $font = str_replace('\\', '\\5C ', $font);
+ // $font = str_replace('"', '\\22 ', $font);
+ // $font = str_replace("'", '\\27 ', $font);
+
+ // font possibly with spaces, requires quoting
+ $final .= "'$font', ";
+ }
+ $final = rtrim($final, ', ');
+ if ($final === '') return false;
+ return $final;
+ }
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Ident.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Ident.php
new file mode 100644
index 000000000..779794a0b
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Ident.php
@@ -0,0 +1,24 @@
+<?php
+
+/**
+ * Validates based on {ident} CSS grammar production
+ */
+class HTMLPurifier_AttrDef_CSS_Ident extends HTMLPurifier_AttrDef
+{
+
+ public function validate($string, $config, $context) {
+
+ $string = trim($string);
+
+ // early abort: '' and '0' (strings that convert to false) are invalid
+ if (!$string) return false;
+
+ $pattern = '/^(-?[A-Za-z_][A-Za-z_\-0-9]*)$/';
+ if (!preg_match($pattern, $string)) return false;
+ return $string;
+
+ }
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/ImportantDecorator.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/ImportantDecorator.php
new file mode 100644
index 000000000..4e6b35e5a
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/ImportantDecorator.php
@@ -0,0 +1,40 @@
+<?php
+
+/**
+ * Decorator which enables !important to be used in CSS values.
+ */
+class HTMLPurifier_AttrDef_CSS_ImportantDecorator extends HTMLPurifier_AttrDef
+{
+ public $def, $allow;
+
+ /**
+ * @param $def Definition to wrap
+ * @param $allow Whether or not to allow !important
+ */
+ public function __construct($def, $allow = false) {
+ $this->def = $def;
+ $this->allow = $allow;
+ }
+ /**
+ * Intercepts and removes !important if necessary
+ */
+ public function validate($string, $config, $context) {
+ // test for ! and important tokens
+ $string = trim($string);
+ $is_important = false;
+ // :TODO: optimization: test directly for !important and ! important
+ if (strlen($string) >= 9 && substr($string, -9) === 'important') {
+ $temp = rtrim(substr($string, 0, -9));
+ // use a temp, because we might want to restore important
+ if (strlen($temp) >= 1 && substr($temp, -1) === '!') {
+ $string = rtrim(substr($temp, 0, -1));
+ $is_important = true;
+ }
+ }
+ $string = $this->def->validate($string, $config, $context);
+ if ($this->allow && $is_important) $string .= ' !important';
+ return $string;
+ }
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Length.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Length.php
new file mode 100644
index 000000000..a07ec5813
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Length.php
@@ -0,0 +1,47 @@
+<?php
+
+/**
+ * Represents a Length as defined by CSS.
+ */
+class HTMLPurifier_AttrDef_CSS_Length extends HTMLPurifier_AttrDef
+{
+
+ protected $min, $max;
+
+ /**
+ * @param HTMLPurifier_Length $max Minimum length, or null for no bound. String is also acceptable.
+ * @param HTMLPurifier_Length $max Maximum length, or null for no bound. String is also acceptable.
+ */
+ public function __construct($min = null, $max = null) {
+ $this->min = $min !== null ? HTMLPurifier_Length::make($min) : null;
+ $this->max = $max !== null ? HTMLPurifier_Length::make($max) : null;
+ }
+
+ public function validate($string, $config, $context) {
+ $string = $this->parseCDATA($string);
+
+ // Optimizations
+ if ($string === '') return false;
+ if ($string === '0') return '0';
+ if (strlen($string) === 1) return false;
+
+ $length = HTMLPurifier_Length::make($string);
+ if (!$length->isValid()) return false;
+
+ if ($this->min) {
+ $c = $length->compareTo($this->min);
+ if ($c === false) return false;
+ if ($c < 0) return false;
+ }
+ if ($this->max) {
+ $c = $length->compareTo($this->max);
+ if ($c === false) return false;
+ if ($c > 0) return false;
+ }
+
+ return $length->toString();
+ }
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/ListStyle.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/ListStyle.php
new file mode 100644
index 000000000..4406868c0
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/ListStyle.php
@@ -0,0 +1,78 @@
+<?php
+
+/**
+ * Validates shorthand CSS property list-style.
+ * @warning Does not support url tokens that have internal spaces.
+ */
+class HTMLPurifier_AttrDef_CSS_ListStyle extends HTMLPurifier_AttrDef
+{
+
+ /**
+ * Local copy of component validators.
+ * @note See HTMLPurifier_AttrDef_CSS_Font::$info for a similar impl.
+ */
+ protected $info;
+
+ public function __construct($config) {
+ $def = $config->getCSSDefinition();
+ $this->info['list-style-type'] = $def->info['list-style-type'];
+ $this->info['list-style-position'] = $def->info['list-style-position'];
+ $this->info['list-style-image'] = $def->info['list-style-image'];
+ }
+
+ public function validate($string, $config, $context) {
+
+ // regular pre-processing
+ $string = $this->parseCDATA($string);
+ if ($string === '') return false;
+
+ // assumes URI doesn't have spaces in it
+ $bits = explode(' ', strtolower($string)); // bits to process
+
+ $caught = array();
+ $caught['type'] = false;
+ $caught['position'] = false;
+ $caught['image'] = false;
+
+ $i = 0; // number of catches
+ $none = false;
+
+ foreach ($bits as $bit) {
+ if ($i >= 3) return; // optimization bit
+ if ($bit === '') continue;
+ foreach ($caught as $key => $status) {
+ if ($status !== false) continue;
+ $r = $this->info['list-style-' . $key]->validate($bit, $config, $context);
+ if ($r === false) continue;
+ if ($r === 'none') {
+ if ($none) continue;
+ else $none = true;
+ if ($key == 'image') continue;
+ }
+ $caught[$key] = $r;
+ $i++;
+ break;
+ }
+ }
+
+ if (!$i) return false;
+
+ $ret = array();
+
+ // construct type
+ if ($caught['type']) $ret[] = $caught['type'];
+
+ // construct image
+ if ($caught['image']) $ret[] = $caught['image'];
+
+ // construct position
+ if ($caught['position']) $ret[] = $caught['position'];
+
+ if (empty($ret)) return false;
+ return implode(' ', $ret);
+
+ }
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Multiple.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Multiple.php
new file mode 100644
index 000000000..4d62a40d7
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Multiple.php
@@ -0,0 +1,58 @@
+<?php
+
+/**
+ * Framework class for strings that involve multiple values.
+ *
+ * Certain CSS properties such as border-width and margin allow multiple
+ * lengths to be specified. This class can take a vanilla border-width
+ * definition and multiply it, usually into a max of four.
+ *
+ * @note Even though the CSS specification isn't clear about it, inherit
+ * can only be used alone: it will never manifest as part of a multi
+ * shorthand declaration. Thus, this class does not allow inherit.
+ */
+class HTMLPurifier_AttrDef_CSS_Multiple extends HTMLPurifier_AttrDef
+{
+
+ /**
+ * Instance of component definition to defer validation to.
+ * @todo Make protected
+ */
+ public $single;
+
+ /**
+ * Max number of values allowed.
+ * @todo Make protected
+ */
+ public $max;
+
+ /**
+ * @param $single HTMLPurifier_AttrDef to multiply
+ * @param $max Max number of values allowed (usually four)
+ */
+ public function __construct($single, $max = 4) {
+ $this->single = $single;
+ $this->max = $max;
+ }
+
+ public function validate($string, $config, $context) {
+ $string = $this->parseCDATA($string);
+ if ($string === '') return false;
+ $parts = explode(' ', $string); // parseCDATA replaced \r, \t and \n
+ $length = count($parts);
+ $final = '';
+ for ($i = 0, $num = 0; $i < $length && $num < $this->max; $i++) {
+ if (ctype_space($parts[$i])) continue;
+ $result = $this->single->validate($parts[$i], $config, $context);
+ if ($result !== false) {
+ $final .= $result . ' ';
+ $num++;
+ }
+ }
+ if ($final === '') return false;
+ return rtrim($final);
+ }
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Number.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Number.php
new file mode 100644
index 000000000..3f99e12ec
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Number.php
@@ -0,0 +1,69 @@
+<?php
+
+/**
+ * Validates a number as defined by the CSS spec.
+ */
+class HTMLPurifier_AttrDef_CSS_Number extends HTMLPurifier_AttrDef
+{
+
+ /**
+ * Bool indicating whether or not only positive values allowed.
+ */
+ protected $non_negative = false;
+
+ /**
+ * @param $non_negative Bool indicating whether negatives are forbidden
+ */
+ public function __construct($non_negative = false) {
+ $this->non_negative = $non_negative;
+ }
+
+ /**
+ * @warning Some contexts do not pass $config, $context. These
+ * variables should not be used without checking HTMLPurifier_Length
+ */
+ public function validate($number, $config, $context) {
+
+ $number = $this->parseCDATA($number);
+
+ if ($number === '') return false;
+ if ($number === '0') return '0';
+
+ $sign = '';
+ switch ($number[0]) {
+ case '-':
+ if ($this->non_negative) return false;
+ $sign = '-';
+ case '+':
+ $number = substr($number, 1);
+ }
+
+ if (ctype_digit($number)) {
+ $number = ltrim($number, '0');
+ return $number ? $sign . $number : '0';
+ }
+
+ // Period is the only non-numeric character allowed
+ if (strpos($number, '.') === false) return false;
+
+ list($left, $right) = explode('.', $number, 2);
+
+ if ($left === '' && $right === '') return false;
+ if ($left !== '' && !ctype_digit($left)) return false;
+
+ $left = ltrim($left, '0');
+ $right = rtrim($right, '0');
+
+ if ($right === '') {
+ return $left ? $sign . $left : '0';
+ } elseif (!ctype_digit($right)) {
+ return false;
+ }
+
+ return $sign . $left . '.' . $right;
+
+ }
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Percentage.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Percentage.php
new file mode 100644
index 000000000..c34b8fc3c
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Percentage.php
@@ -0,0 +1,40 @@
+<?php
+
+/**
+ * Validates a Percentage as defined by the CSS spec.
+ */
+class HTMLPurifier_AttrDef_CSS_Percentage extends HTMLPurifier_AttrDef
+{
+
+ /**
+ * Instance of HTMLPurifier_AttrDef_CSS_Number to defer number validation
+ */
+ protected $number_def;
+
+ /**
+ * @param Bool indicating whether to forbid negative values
+ */
+ public function __construct($non_negative = false) {
+ $this->number_def = new HTMLPurifier_AttrDef_CSS_Number($non_negative);
+ }
+
+ public function validate($string, $config, $context) {
+
+ $string = $this->parseCDATA($string);
+
+ if ($string === '') return false;
+ $length = strlen($string);
+ if ($length === 1) return false;
+ if ($string[$length - 1] !== '%') return false;
+
+ $number = substr($string, 0, $length - 1);
+ $number = $this->number_def->validate($number, $config, $context);
+
+ if ($number === false) return false;
+ return "$number%";
+
+ }
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/TextDecoration.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/TextDecoration.php
new file mode 100644
index 000000000..772c922d8
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/TextDecoration.php
@@ -0,0 +1,38 @@
+<?php
+
+/**
+ * Validates the value for the CSS property text-decoration
+ * @note This class could be generalized into a version that acts sort of
+ * like Enum except you can compound the allowed values.
+ */
+class HTMLPurifier_AttrDef_CSS_TextDecoration extends HTMLPurifier_AttrDef
+{
+
+ public function validate($string, $config, $context) {
+
+ static $allowed_values = array(
+ 'line-through' => true,
+ 'overline' => true,
+ 'underline' => true,
+ );
+
+ $string = strtolower($this->parseCDATA($string));
+
+ if ($string === 'none') return $string;
+
+ $parts = explode(' ', $string);
+ $final = '';
+ foreach ($parts as $part) {
+ if (isset($allowed_values[$part])) {
+ $final .= $part . ' ';
+ }
+ }
+ $final = rtrim($final);
+ if ($final === '') return false;
+ return $final;
+
+ }
+
+}
+
+// vim: et sw=4 sts=4
diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/URI.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/URI.php
new file mode 100644
index 000000000..c2f767e57
--- /dev/null
+++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/URI.php
@@ -0,0 +1,61 @@
+<?php
+
+/**
+ * Validates a URI in CSS syntax, which uses url('http://example.com')
+ * @note While theoretically speaking a URI in a CSS document could
+ * be non-embedded, as of CSS2 there is no such usage so we're
+ * generalizing it. This may need to be changed in the future.
+ * @warning Since HTMLPurifier_AttrDef_CSS blindly uses semicolons as
+ * the separator, you cannot put a literal semicolon in
+ * in the URI. Try percent encoding it, in that case.
+ */
+class HTMLPurifier_AttrDef_CSS_URI extends HTMLPurifier_AttrDef_URI
+{
+
+ public function __construct() {
+ parent::__construct(true); // always embedded
+ }
+
+ public function validate($uri_string, $config, $context) {
+ // parse the URI out of the string and then pass it onto
+ // the parent object
+
+ $uri_string = $this->parseCDATA($uri_string);
+ if (strpos($uri_string, 'url(') !== 0) return false;
+ $uri_string = substr($uri_string, 4);
+ $new_length = strlen($uri_string) - 1;
+ if ($uri_string[$new_length] != ')') return false;
+ $uri = trim(substr($uri_string, 0, $new_length));
+
+ if (!empty($uri) && ($uri[0] == "'" || $uri[0] == '"')) {
+ $quote = $uri[0];
+ $new_length = strlen($uri) - 1;
+ if ($uri[$new_length] !== $quote) return false;
+ $uri = substr($uri, 1, $new_length - 1);
+ }
+
+ $uri = $this->expandCSSEscape($uri);
+
+ $result = parent::validate($uri, $config, $context);
+
+ if ($result === false) return false;
+
+ // extra sanity check; should have been done by URI
+ $result = str_replace(array('"', "\\", "\n", "\x0c", "\r"), "", $result);
+
+ // suspicious characters are ()'; we're going to percent encode
+ // them for safety.
+ $result = str_replace(array('(', ')', "'"), array('%28', '%29', '%27'), $result);
+
+ // there's an extra bug where ampersands lose their escaping on
+ // an innerHTML cycle, so a very unlucky query parameter could
+ // then change the meaning of the URL. Unfortunately, there's
+ // not much we can do about that...
+
+ return "url(\"$result\")";
+
+ }
+
+}
+
+// vim: et sw=4 sts=4