MensBeam
/
HTML-Parser
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
Browse Source

Overhaul open elements stack

ns
J. King 3 years ago
parent
5376ed43a5
commit
979cec628e
4 changed files with 247 additions and 393 deletions
Whitespace
Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL
Split View
Diff Options
Show Stats Download Patch File Download Diff File
  1. 6
      lib/ActiveFormattingElementsList.php
  2. 496
      lib/OpenElementsStack.php
  3. 16
      lib/TemplateInsertionModesStack.php
  4. 122
      lib/TreeBuilder.php

6
lib/ActiveFormattingElementsList.php
View File

@ -17,8 +17,10 @@ namespace dW\HTML5;
class ActiveFormattingElementsList extends Stack {
protected $_storage = [];
protected $stack;
protected $tree;
public function __construct(OpenElementsStack $stack) {
public function __construct(TreeBuilder $tree, OpenElementsStack $stack) {
$this->tree = $tree;
$this->stack = $stack;
}
@ -143,7 +145,7 @@ class ActiveFormattingElementsList extends Stack {
# 8. Create: Insert an HTML element for the token for which the element entry
# was created, to obtain new element.
create:
$element = TreeBuilder::insertStartTagToken($entry['token']);
$element = $this->tree->insertStartTagToken($entry['token']);
# 9. Replace the entry for entry in the list with an entry for new element.
$this->_storage[key($this->_storage)]['element'] = $element;

496
lib/OpenElementsStack.php
View File

@ -2,7 +2,91 @@
declare(strict_types=1);
namespace dW\HTML5;
class OpenElementsStack extends Stack {
class OpenElementsStack extends \splStack {
protected const IMPLIED_END_TAGS = [
'dd' => true,
'dt' => true,
'li' => true,
'optgroup' => true,
'option' => true,
'p' => true,
'rb' => true,
'rp' => true,
'rt' => true,
'rtc' => true,
];
protected const IMPLIED_END_TAGS_THOROUGH = [
'caption' => true,
'colgroup' => true,
'dd' => true,
'dt' => true,
'li' => true,
'optgroup' => true,
'option' => true,
'p' => true,
'rb' => true,
'rp' => true,
'rt' => true,
'rtc' => true,
'tbody' => true,
'td' => true,
'tfoot' => true,
'th' => true,
'thead' => true,
'tr' => true,
];
protected const GENERAL_SCOPE = [
Parser::HTML_NAMESPACE => [
'applet',
'caption',
'html',
'table',
'td',
'th',
'marquee',
'object',
'template'
],
Parser::MATHML_NAMESPACE => [
'mi',
'mo',
'mn',
'ms',
'mtext',
'annotation-xml'
],
Parser::SVG_NAMESPACE => [
'foreignObject',
'desc',
'title'
],
];
protected const LIST_ITEM_SCOPE = [
// everything in general scope, and these in the HTML namespace
'ol',
'ul',
];
protected const BUTTON_SCOPE = [
// everything in general scope, and these in the HTML namespace
'button',
];
protected const TABLE_SCOPE = [
Parser::HTML_NAMESPACE => [
'html',
'table',
'template',
],
];
protected const SELECT_SCOPE = [
// all elements EXCEPT these
Parser::HTML_NAMESPACE => [
'optgroup',
'option',
],
];
protected $fragmentCase;
protected $fragmentContext;
@ -18,88 +102,68 @@ class OpenElementsStack extends Stack {
$this->fragmentContext = $fragmentContext;
}
public function popUntil($target) {
if ($target instanceof Element) {
do {
$node = $this->pop;
} while (!$node->isSameNode($target));
} elseif (is_string($target)) {
do {
$poppedNodeName = $this->pop()->nodeName;
} while ($poppedNodeName !== $target);
} elseif (is_array($target)) {
do {
$poppedNodeName = $this->pop()->nodeName;
} while (!in_array($poppedNodeName, $target));
} else {
throw new Exception(Exception::STACK_ELEMENT_STRING_ARRAY_EXPECTED);
}
public function popUntil(string ...$target): void {
do {
$node = $this->pop();
} while (!in_array($node->nodeName, $target));
}
public function search($needle): int {
if (!$needle) {
return -1;
}
public function popUntilSame(Element $target): void {
do {
$node = $this->pop();
} while (!$node->isSameNode($target));
}
if ($needle instanceof \DOMElement) {
foreach (array_reverse($this->_storage) as $key => $value) {
if ($value->isSameNode($needle)) {
return $key;
}
}
} elseif (is_string($needle)) {
foreach (array_reverse($this->_storage) as $key => $value) {
if ($value->nodeName === $needle) {
return $key;
}
}
} elseif ($needle instanceof \Closure) {
foreach (array_reverse($this->_storage) as $key => $value) {
if ($needle($value) === true) {
return $key;
}
public function find(string ...$name): int {
foreach ($this as $k => $node) {
if (in_array($node->nodeName, $name)) {
return $k;
}
}
return -1;
}
// Remove an arbitrary element from the array.
public function remove($target) {
$key = $this->search($target);
if ($key === -1) {
return;
} elseif ($key === count($this->_storage) - 1) {
$this->pop();
return;
public function findNot(string ...$name): int {
foreach ($this as $k => $node) {
if (!in_array($node->nodeName, $name)) {
return $k;
}
}
unset($this->_storage[$key]);
$this->_storage = array_values($this->_storage);
return -1;
}
public function generateImpliedEndTags(array $exclude = []) {
$tags = ['caption', 'colgroup', 'dd', 'dt', 'li', 'optgroup', 'option', 'p', 'rb', 'rp', 'rt', 'rtc', 'tbody', 'td', 'tfoot', 'th', 'thead', 'tr'];
if (count($exclude) > 0) {
$modified = false;
foreach ($exclude as $e) {
$key = array_search($e, $tags);
if ($key !== false) {
unset($tags[$key]);
$modified = true;
}
public function findSame(\DOMElement $node): int {
foreach ($this as $k => $node) {
if ($node->isSameNode($node)) {
return $k;
}
}
return -1;
}
if ($modified) {
$tags = array_values($tags);
}
public function generateImpliedEndTags(string ...$exclude): void {
# When the steps below require the UA to generate implied end tags,
# then, while the current node is {elided list of element names},
# the UA must pop the current node off the stack of open elements.
#
# If a step requires the UA to generate implied end tags but lists
# an element to exclude from the process, then the UA must perform
# the above steps as if that element was not in the above list.
$map = self::IMPLIED_END_TAGS;
foreach($exclude as $name) {
$map[$name] = false;
}
while (!$this->isEmpty() && ($map[$this->top()->nodeName] ?? false)) {
$this->pop();
}
}
$currentNodeName = end($this->_storage)->nodeName;
while (in_array($currentNodeName, $tags)) {
public function generateImpliedEndTagsThoroughly(): void {
# When the steps below require the UA to generate all implied end tags
# thoroughly, then, while the current node is {elided list of element names},
# the UA must pop the current node off the stack of open elements.
while (!$this->isEmpty() && (self::IMPLIED_END_TAGS_THOROUGH[$this->top()->nodeName] ?? false)) {
$this->pop();
$currentNodeName = end($this->_storage)->nodeName;
}
}
@ -108,296 +172,102 @@ class OpenElementsStack extends Stack {
# it has that element in the specific scope consisting of the following element
# types:
#
# applet
# caption
# html
# table
# td
# th
# marquee
# object
# template
# MathML mi
# MathML mo
# MathML mn
# MathML ms
# MathML mtext
# MathML annotation-xml
# SVG foreignObject
# SVG desc
# SVG title
$list = [
Parser::HTML_NAMESPACE => [
'applet',
'caption',
'html',
'table',
'td',
'th',
'marquee',
'object',
'template'
],
Parser::MATHML_NAMESPACE => [
'mi',
'mo',
'mn',
'ms',
'mtext',
'annotation-xml'
],
Parser::SVG_NAMESPACE => [
'foreignObject',
'desc',
'title'
]
];
return $this->hasElementInScopeHandler($target, $list);
# {elided}
return $this->hasElementInScopeHandler($target, self::GENERAL_SCOPE);
}
public function hasElementInListItemScope($target): bool {
# The stack of open elements is said to have a particular element in list item scope when it has that element in the specific scope consisting of the following element types:
#
# All the element types listed above for the has an element in scope algorithm.
# ol in the HTML namespace
# ul in the HTML namespace
$list = [
Parser::HTML_NAMESPACE => [
'applet',
'caption',
'html',
'table',
'td',
'th',
'marquee',
'object',
'template',
'ol',
'ul'
],
Parser::MATHML_NAMESPACE => [
'mi',
'mo',
'mn',
'ms',
'mtext',
'annotation-xml'
],
Parser::SVG_NAMESPACE => [
'foreignObject',
'desc',
'title'
]
];
return $this->hasElementInScopeHandler($target, $list);
$scope = self::GENERAL_SCOPE;
$scope[Parser::HTML_NAMESPACE] = array_merge($scope[Parser::HTML_NAMESPACE], self::LIST_ITEM_SCOPE);
return $this->hasElementInScopeHandler($target, $scope);
}
public function hasElementInButtonScope($target): bool {
# The stack of open elements is said to have a particular element in button
# scope when it has that element in the specific scope consisting of the
# following element types:
#
# All the element types listed above for the has an element in scope algorithm.
# button in the HTML namespace
$list = [
Parser::HTML_NAMESPACE => [
'applet',
'caption',
'html',
'table',
'td',
'th',
'marquee',
'object',
'template',
'button'
],
Parser::MATHML_NAMESPACE => [
'mi',
'mo',
'mn',
'ms',
'mtext',
'annotation-xml'
],
Parser::SVG_NAMESPACE => [
'foreignObject',
'desc',
'title'
]
];
return $this->hasElementInScopeHandler($target, $list);
$scope = self::GENERAL_SCOPE;
$scope[Parser::HTML_NAMESPACE] = array_merge($scope[Parser::HTML_NAMESPACE], self::BUTTON_SCOPE);
return $this->hasElementInScopeHandler($target, $scope);
}
public function hasElementInTableScope($target): bool {
# The stack of open elements is said to have a particular element in table scope
# when it has that element in the specific scope consisting of the following
# element types:
#
# All the element types listed above for the has an element in scope algorithm.
# html in the HTML namespace
# table in the HTML namespace
# template in the HTML namespace
// Not sure what to do here. I am going to assume the elements without a
// namespace in the element types listed above are meant for the HTML namespace.
// If so then these listed here are redundant. My interpretation therefore has
// this being an alias for hasElementInScope.
return $this->hasElementInScope($target);
return $this->hasElementInScopeHandler($target, self::TABLE_SCOPE);
}
public function hasElementInSelectScope(string $target): bool {
# The stack of open elements is said to have a particular element in select
# scope when it has that element in the specific scope consisting of all element
# types except the following:
# The stack of open elements is said to have a particular element
# in select scope when it has that element in the specific scope
# consisting of all element types EXCEPT the following:
#
# All the element types listed above for the has an element in scope algorithm.
# optgroup in the HTML namespace
# option in the HTML namespace
$list = [
Parser::HTML_NAMESPACE => [
'applet',
'caption',
'html',
'table',
'td',
'th',
'marquee',
'object',
'template',
'button',
'optgroup',
'option'
],
Parser::MATHML_NAMESPACE => [
'mi',
'mo',
'mn',
'ms',
'mtext',
'annotation-xml'
],
Parser::SVG_NAMESPACE => [
'foreignObject',
'desc',
'title'
]
];
return $this->hasElementInScopeHandler($target, $list);
return $this->hasElementInScopeHandler($target, self::SELECT_SCOPE, false);
}
protected function hasElementInScopeHandler($target, array $list): bool {
# 1. Initialize node to be the current node (the bottommost node of the stack).
// Handled by loop.
foreach (array_reverse($this->_storage) as $node) {
# 2. If node is the target node, terminate in a match state.
if ($target instanceof \DOMElement) {
protected function hasElementInScopeHandler($target, array $list, $matchType = true): bool {
assert(is_string($target) || $target instanceof \DOMElement, new \Exception("Invalid input type"));
# The stack of open elements is said to have an element target node
# in a specific scope consisting of a list of element types list
# when the following algorithm terminates in a match state:
if ($target instanceof \DOMElement) {
# Initialize node to be the current node (the bottommost node of the stack).
foreach ($this as $node) {
# If node is the target node, terminate in a match state.
if ($node->isSameNode($target)) {
return true;
}
} elseif (is_string($target)) {
# Otherwise, if node is one of the element types in list, terminate in a failure state.
$ns = $node->namespaceURI ?? Parser::HTML_NAMESPACE;
if (in_array($node->nodeName, $list[$ns] ?? []) === $matchType) {
return false;
}
# Otherwise, set node to the previous entry in the stack of
# open elements and return to step 2. (This will never fail,
# since the loop will always terminate in the previous step
# if the top of the stack — an html element — is reached.)
}
} else {
# Initialize node to be the current node (the bottommost node of the stack).
foreach ($this as $node) {
# If node is the target node, terminate in a match state.
if ($node->nodeName === $target) {
return true;
}
}
# 3. Otherwise, if node is one of the element types in list, terminate in a
# failure state.
else {
foreach ($list as $namespace => $subList) {
if ($namespace === Parser::HTML_NAMESPACE) {
$namespace = '';
}
if ($node->namespaceURI !== $namespace) {
continue;
}
foreach ($subList as $name) {
if ($node->nodeName === $name) {
return false;
}
}
# Otherwise, if node is one of the element types in list, terminate in a failure state.
$ns = $node->namespaceURI ?? Parser::HTML_NAMESPACE;
if (in_array($node->nodeName, $list[$ns] ?? []) === $matchType) {
return false;
}
# Otherwise, set node to the previous entry in the stack of
# open elements and return to step 2. (This will never fail,
# since the loop will always terminate in the previous step
# if the top of the stack — an html element — is reached.)
}
# Otherwise, set node to the previous entry in the stack of open elements and
# return to step 2. (This will never fail, since the loop will always terminate
# in the previous step if the top of the stack — an html element — is reached.)
// Handled by loop.
}
return false;
}
public function __get($property) {
$value = parent::__get($property);
if (!is_null($value)) {
return $value;
}
switch ($property) {
case 'adjustedCurrentNode':
# The adjusted current node is the context element if the parser was created by
# the HTML fragment parsing algorithm and the stack of open elements has only one
# element in it (fragment case); otherwise, the adjusted current node is the
# current node.
return ($this->fragmentCase && $this->length === 1) ? $this->fragmentContext : $this->currentNode;
break;
return ($this->fragmentCase && count($this) === 1) ? $this->fragmentContext : $this->__get('currentNode');
case 'adjustedCurrentNodeName':
$adjustedCurrentNode = $this->adjustedCurrentNode;
$adjustedCurrentNode = $this->__get('adjustedCurrentNode');
return (!is_null($adjustedCurrentNode)) ? $adjustedCurrentNode->nodeName : null;
break;
case 'adjustedCurrentNodeNamespace':
$adjustedCurrentNode = $this->adjustedCurrentNode;
$adjustedCurrentNode = $this->__get('adjustedCurrentNode');
return (!is_null($adjustedCurrentNode)) ? $adjustedCurrentNode->namespaceURI: null;
break;
case 'currentNode':
$currentNode = end($this->_storage);
return ($currentNode) ? $currentNode : null;
break;
return $this->isEmpty() ? null : $this->top();
case 'currentNodeName':
$currentNode = $this->currentNode;
$currentNode = $this->__get('currentNode');
return ($currentNode && $currentNode->nodeType) ? $currentNode->nodeName : null;
break;
case 'currentNodeNamespace':
$currentNode = $this->currentNode;
$currentNode = $this->__get('currentNode');
return (!is_null($currentNode)) ? $currentNode->namespaceURI: null;
break;
default: return null;
}
}
// Used when listing expected elements when returning parse errors
public function __toString(): string {
if (count($this->_storage) > 1) {
// Don't output the name of the root element.
for ($i = 1, $temp = []; $i < count($this->_storage) - 1; $i++) {
$temp[] = $this->_storage[$i]->nodeName;
}
return implode(', ', array_unique($temp));
} else {
return '';
default:
return null;
}
}
}

16
lib/TemplateInsertionModesStack.php
View File

@ -2,19 +2,13 @@
declare(strict_types=1);
namespace dW\HTML5;
class TemplateInsertionModesStack extends Stack {
class TemplateInsertionModesStack extends \SplStack {
public function __get($property) {
$value = parent::__get($property);
if (!is_null($value)) {
return $value;
}
switch ($property) {
case 'currentMode': return
$currentMode = end($this->_storage);
return ($currentMode) ? $currentMode : null;
break;
default: return null;
case 'currentMode':
return $this->isEmpty() ? null : $this->top();
default:
return null;
}
}
}

122
lib/TreeBuilder.php
View File

@ -193,8 +193,23 @@ class TreeBuilder {
'ychannelselector' => 'yChannelSelector',
'zoomandpan' => 'zoomAndPan',
];
# The following elements have varying levels of special parsing rules: HTML’s
# address, applet, area, article, aside, base, basefont, bgsound, blockquote,
# body, br, button, caption, center, col, colgroup, dd, details, dir, div, dl,
# dt, embed, fieldset, figcaption, figure, footer, form, frame, frameset, h1,
# h2, h3, h4, h5, h6, head, header, hr, html, iframe, img, input, li, link,
# listing, main, marquee, meta, nav, noembed, noframes, noscript, object, ol, p,
# param, plaintext, pre, script, section, select, source, style, summary, table,
# tbody, td, template, textarea, tfoot, th, thead, title, tr, track, ul, wbr,
# xmp; MathML mi, MathML mo, MathML mn, MathML ms, MathML mtext, and MathML
# annotation-xml; and SVG foreignObject, SVG desc, and SVG title.
protected const SPECIAL_ELEMENTS = [
Parser::HTML_NAMESPACE => ['address', 'applet', 'area', 'article', 'aside', 'base', 'basefont', 'bgsound', 'blockquote', 'body', 'br', 'button', 'caption', 'center', 'col', 'colgroup', 'dd', 'details', 'dir', 'div', 'dl', 'dt', 'embed', 'fieldset', 'figcaption', 'figure', 'footer', 'form', 'frame', 'frameset', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'head', 'header', 'hr', 'html', 'iframe', 'img', 'input', 'li', 'link', 'listing', 'main', 'marquee', 'meta', 'nav', 'noembed', 'noframes', 'noscript', 'object', 'ol', 'p', 'param', 'plaintext', 'pre', 'script', 'section', 'select', 'source', 'style', 'summary', 'table', 'tbody', 'td', 'template', 'textarea', 'tfoot', 'th', 'thead', 'title', 'tr', 'track', 'ul', 'wbr', 'xmp'],
Parser::MATHML_NAMESPACE => ['mi', 'mo', 'mn', 'ms', 'mtext', 'annotation-xml'],
Parser::SVG_NAMESPACE => ['foreignObject', 'desc', 'title'],
];
public function __construct(Document $dom, $formElement, bool $fragmentCase = false, $fragmentContext = null, OpenElementsStack $stack, Stack $templateInsertionModes, Tokenizer $tokenizer, ParseError $errorHandler, Data $data) {
public function __construct(Document $dom, $formElement, bool $fragmentCase = false, $fragmentContext = null, OpenElementsStack $stack, TemplateInsertionModesStack $templateInsertionModes, Tokenizer $tokenizer, ParseError $errorHandler, Data $data) {
// If the form element isn't an instance of DOMElement that has a node name of
// "form" or null then there's a problem.
if (!is_null($formElement) && !($formElement instanceof \DOMElement && $formElement->nodeName === 'form')) {
@ -221,7 +236,7 @@ class TreeBuilder {
$this->errorHandler = $errorHandler;
// Initialize the list of active formatting elements.
$this->activeFormattingElementsList = new ActiveFormattingElementsList($stack);
$this->activeFormattingElementsList = new ActiveFormattingElementsList($this, $stack);
$this->insertionMode = self::INITIAL_MODE;
$this->quirksMode = self::QUIRKS_MODE_OFF;
@ -238,6 +253,7 @@ class TreeBuilder {
assert($iterations++ < 50, new LoopException("Probable infinite loop detected in HTML content handling"));
$adjustedCurrentNode = $this->stack->adjustedCurrentNode;
$adjustedCurrentNodeName = $this->stack->adjustedCurrentNodeName;
assert(!$adjustedCurrentNode || $adjustedCurrentNodeName, new \Exception("The adjusted current node must have a name if not null"));
$adjustedCurrentNodeNamespace = $this->stack->adjustedCurrentNodeNamespace;
# 13.2.6 Tree construction
@ -246,7 +262,7 @@ class TreeBuilder {
# appropriate steps from the following list, known as the tree construction dispatcher:
#
# If the stack of open elements is empty
if ($this->stack->length === 0 ||
if (count($this->stack) === 0 ||
# If the adjusted current node is an element in the HTML namespace
// PHP's DOM returns null when the namespace isn't specified... eg. HTML.
is_null($adjustedCurrentNodeNamespace) || (
@ -747,7 +763,7 @@ class TreeBuilder {
elseif ($token->name === 'template') {
# If there is no template element on the stack of open elements, then this is a
# parse error; ignore the token.
if ($this->stack->search('template') === -1) {
if ($this->stack->find('template') === -1) {
$this->error(ParseError::UNEXPECTED_END_TAG);
}
# Otherwise, run these steps:
@ -916,7 +932,7 @@ class TreeBuilder {
$this->parseTokenInHTMLContent($token, self::IN_HEAD_MODE);
# Remove the node pointed to by the head element pointer from the stack of open
# elements. (It might not be the current node at this point.)
$key = $this->stack->search($this->headElement);
$key = $this->stack->findSame($this->headElement);
if ($key !== -1) {
unset($this->stack[$key]);
}
@ -1015,7 +1031,7 @@ class TreeBuilder {
$this->error(ParseError::UNEXPECTED_START_TAG, 'html');
# If there is a template element on the stack of open elements, then ignore the
# token.
if ($this->stack->search('template') === -1) {
if ($this->stack->find('template') === -1) {
# Otherwise, for each attribute on the token, check to see if the attribute is
# already present on the top element of the stack of open elements. If it is
# not, add the attribute and its corresponding value to that element.
@ -1040,7 +1056,7 @@ class TreeBuilder {
# If the second element on the stack of open elements is not a body element, if
# the stack of open elements has only one node on it, or if there is a template
# element on the stack of open elements, then ignore the token. (fragment case)
if (!($this->stack[1]->tagName !== 'body' || $this->stack->length === 1 || $this->stack->search('template') !== -1)) {
if (!($this->stack[1]->tagName !== 'body' || count($this->stack) === 1 || $this->stack->find('template') !== -1)) {
# Otherwise, set the frameset-ok flag to "not ok"; then, for each attribute on
# the token, check to see if the attribute is already present on the body
# element (the second element) on the stack of open elements, and if it is not,
@ -1064,7 +1080,7 @@ class TreeBuilder {
# element on the stack of open elements is not a body element, then ignore the
# token. (fragment case)
# If the frameset-ok flag is set to "not ok", ignore the token.
if (!($this->stack->length === 1 || $this->stack[1]->tagName !== 'body' || $this->framesetOk === false)) {
if (!(count($this->stack) === 1 || $this->stack[1]->tagName !== 'body' || $this->framesetOk === false)) {
# Otherwise, run the following steps:
#
# 1. Remove the second element on the stack of open elements from its parent
@ -1075,7 +1091,7 @@ class TreeBuilder {
}
# 2. Pop all the nodes from the bottom of the stack of open elements, from the
# current node up to, but not including, the root html element.
for ($i = $this->stack->length - 1; $i > 0; $i--) {
for ($i = count($this->stack) - 1; $i > 0; $i--) {
$this->stack->pop();
}
# 3. Insert an HTML element for the token.
@ -1094,7 +1110,6 @@ class TreeBuilder {
if ($this->stack->hasElementInButtonScope('p')) {
$this->closePElement();
}
# Insert an HTML element for the token.
$this->insertStartTagToken($token);
}
@ -1111,7 +1126,7 @@ class TreeBuilder {
# off the stack of open elements.
$currentNodeName = $this->stack->currentNodeName;
$currentNodeNamespace = $this->stack->currentNodeNamespace;
if ($currentNodeNamespace === '' && ($currentNodeName === 'h1' || $currentNodeName === 'h2' || $currentNodeName === 'h3' || $currentNodeName === 'h4' || $currentNodeName === 'h5' || $currentNodeName === 'h6')) {
if ($currentNodeNamespace === null && ($currentNodeName === 'h1' || $currentNodeName === 'h2' || $currentNodeName === 'h3' || $currentNodeName === 'h4' || $currentNodeName === 'h5' || $currentNodeName === 'h6')) {
$this->error(ParseError::UNEXPECTED_START_TAG, $token->name);
$this->stack->pop();
}
@ -1146,7 +1161,6 @@ class TreeBuilder {
$nextToken->data = substr($nextToken->data, 1);
}
}
// Process the next token
$token = $nextToken;
goto ProcessToken;
@ -1155,7 +1169,7 @@ class TreeBuilder {
elseif ($token->name === 'form') {
# If the form element pointer is not null, and there is no template element on
# the stack of open elements, then this is a parse error; ignore the token.
$templateInStack = ($this->stack->search('template') !== -1);
$templateInStack = ($this->stack->find('template') !== -1);
if (!is_null($this->formElement) && !$templateInStack) {
$this->error(ParseError::UNEXPECTED_START_TAG, $token->name);
}
@ -1183,13 +1197,13 @@ class TreeBuilder {
# 2. Initialize node to be the current node (the bottommost node of the stack).
# 3. Loop: If node is an li element, then run these substeps:
for ($i = $this->stack->length - 1; $i >= 0; $i--) {
for ($i = count($this->stack) - 1; $i >= 0; $i--) {
$node = $this->stack[$i];
$nodeName = $node->nodeName;
if ($nodeName === 'li') {
# 1. Generate implied end tags, except for li elements.
$this->stack->generateImpliedEndTags(["li"]);
$this->stack->generateImpliedEndTags("li");
# 2. If the current node is not an li element, then this is a parse error.
if ($this->stack->currentNodeName !== 'li') {
@ -1201,13 +1215,13 @@ class TreeBuilder {
$this->stack->popUntil('li');
# 4. Jump to the step labeled Done below.
return true;
break;
}
# 4. If node is in the special category, but is not an address, div, or p
# element, then jump to the step labeled Done below.
if ($nodeName !== 'address' && $nodeName !== 'div' && $nodeName !== 'p' && $this->isElementSpecial($node)) {
return true;
break;
}
# 5. Otherwise, set node to the previous entry in the stack of open elements and
@ -1230,7 +1244,7 @@ class TreeBuilder {
$this->framesetOk = false;
# 2. Initialize node to be the current node (the bottommost node of the stack).
for ($i = $this->stack->length - 1; $i >= 0; $i--) {
for ($i = count($this->stack) - 1; $i >= 0; $i--) {
$node = $this->stack[$i];
$nodeName = $node->nodeName;
@ -1240,7 +1254,7 @@ class TreeBuilder {
# 4. If node is a dt element, then run these substeps:
if ($nodeName === 'dd' || $nodeName === 'dt') {
# 1. Generate implied end tags, except for dd or dt elements.
$this->stack->generateImpliedEndTags(['dd', 'dt']);
$this->stack->generateImpliedEndTags('dd', 'dt');
# 2. If the current node is not a dd or dt element, then this is a parse error.
if ($this->stack->currentNodeName !== $nodeName) {
@ -1249,16 +1263,16 @@ class TreeBuilder {
# 3. Pop elements from the stack of open elements until a dd or dt element has been
# popped from the stack.
$this->stack->popUntil(['dd', 'dt']);
$this->stack->popUntil('dd', 'dt');
# 4. Jump to the step labeled Done below.
return true;
break;
}
# 5. If node is in the special category, but is not an address, div, or p
# element, then jump to the step labeled Done below.
if ($nodeName !== 'address' && $nodeName !== 'div' && $nodeName !== 'p' && $this->isElementSpecial($node)) {
return true;
break;
}
# 6. Otherwise, set node to the previous entry in the stack of open elements and
@ -1342,7 +1356,7 @@ class TreeBuilder {
# If the stack of open elements does not have a body element in scope, this is a
# parse error; ignore the token.
if (!$this->stack->hasElementInScope('body')) {
$this->error(ParseError::UNEXPECTED_END_TAG, 'body');
$this->error(ParseError::UNEXPECTED_END_TAG);
}
# Otherwise, if there is a node in the stack of open elements that is not either
# a dd element, a dt element, an li element, an optgroup element, an option
@ -1351,21 +1365,11 @@ class TreeBuilder {
# element, a tr element, the body element, or the html element, then this is a
# parse error.
else {
if ($this->stack->search(function($node) {
$n = $node->nodeName;
if ($n !== 'dd' && $n !== 'dt' && $n !== 'li' && $n !== 'optgroup' && $n !== 'option' && $n !== 'p' && $n !== 'rb' && $n !== 'rp' && $n !== 'rt' && $n !== 'rtc' && $n !== 'tbody' && $n !== 'td' && $n !== 'tfoot' && $n !== 'th' && $n !== 'thead' && $n !== 'tr' && $n !== 'body' && $n !== 'html') {
return true;
}
return false;
}) !== -1) {
$this->error(ParseError::UNEXPECTED_END_TAG, 'body');
return true;
if ($this->stack->findNot('dd', 'dt', 'li', 'optgroup', 'option', 'p', 'rb', 'rp', 'rt', 'rtc', 'tbody', 'td', 'tfoot', 'th', 'thead', 'tr', 'body', 'html') > -1) {
$this->error(ParseError::UNEXPECTED_END_TAG);
}
# Switch the insertion mode to "after body".
$this->insertionMode = self::AFTER_BODY_MODE;
// The only thing different between body and html here is that when processing
// an html end tag the token is reprocessed.
if ($token->name === 'html') {
@ -1405,7 +1409,7 @@ class TreeBuilder {
elseif ($token->name === 'form') {
# If there is no template element on the stack of open elements, then run these
# substeps:
if ($this->stack->search('template') === -1) {
if ($this->stack->find('template') === -1) {
# 1. Let node be the element that the form element pointer is set to, or null if it
# is not set to an element.
$node = $this->formElement;
@ -1424,7 +1428,10 @@ class TreeBuilder {
$this->error(ParseError::UNEXPECTED_END_TAG, $token->name);
}
# 6. Remove node from the stack of open elements
$this->stack->remove($node);
$key = $this->stack->findSame($node);
if ($key > -1) {
unset($this->stack[$key]);
}
}
# If there is a template element on the stack of open elements, then run these
# substeps instead:
@ -1452,7 +1459,7 @@ class TreeBuilder {
# An end-of-file token
elseif ($token instanceof EOFToken) {
# If the stack of template insertion modes is not empty, then process the token using the rules for the "in template" insertion mode.
if ($this->templateInsertionModes->length !== 0) {
if (count($this->templateInsertionModes) !== 0) {
$insertionMode = self::IN_TEMPLATE_MODE;
goto ProcessToken;
}
@ -1463,16 +1470,8 @@ class TreeBuilder {
# a p element, an rb element, an rp element, an rt element, an rtc element, a
# tbody element, a td element, a tfoot element, a th element, a thead element, a
# tr element, the body element, or the html element, then this is a parse error.
if ($this->stack->search(function($node) {
$n = $node->nodeName;
if ($n !== 'dd' && $n !== 'dt' && $n !== 'li' && $n !== 'optgroup' && $n !== 'option' && $n !== 'p' && $n !== 'rb' && $n !== 'rp' && $n !== 'rt' && $n !== 'rtc' && $n !== 'tbody' && $n !== 'td' && $n !== 'tfoot' && $n !== 'th' && $n !== 'thead' && $n !== 'tr' && $n !== 'body' && $n !== 'html') {
return true;
}
return false;
}) !== -1) {
if ($this->stack->findNot('dd', 'dt', 'li', 'optgroup', 'option', 'p', 'rb', 'rp', 'rt', 'rtc', 'tbody', 'td', 'tfoot', 'th', 'thead', 'tr', 'body', 'html') > -1) {
$this->error(ParseError::UNEXPECTED_END_TAG, 'body');
return true;
}
# 2. Stop parsing.
@ -1675,10 +1674,10 @@ class TreeBuilder {
# 3. Loop: If node's tag name, converted to ASCII lowercase, is the same as the
# tag name of the token, pop elements from the stack of open elements until node
# has been popped from the stack, and then abort these steps.
$count = $this->stack->length - 1;
$count = count($this->stack) - 1;
while (true) {
if (strtolower($nodeName) === $token->name) {
$this->stack->popUntil($node);
$this->stack->popUntilSame($node);
break;
}
@ -1728,12 +1727,12 @@ class TreeBuilder {
#
# 1. Let last template be the last template element in the stack of open
# elements, if any.
$lastTemplateKey = $this->stack->search('template');
$lastTemplateKey = $this->stack->find('template');
$lastTemplate = ($lastTemplateKey !== -1 ) ? $this->stack[$lastTemplateKey] : null;
# 2. Let last table be the last table element in the stack of open elements, if
# any.
$lastTableKey = $this->stack->search('table');
$lastTableKey = $this->stack->find('table');
$lastTable = ($lastTableKey !== -1 ) ? $this->stack[$lastTableKey] : null;
# 3. If there is a last template and either there is no last table, or there is
@ -2035,7 +2034,7 @@ class TreeBuilder {
$node = $this->stack->currentNode;
$nodeName = $this->stack->currentNodeName;
// Keeping up with the position, too.
$position = $this->stack->length - 1;
$position = count($this->stack) - 1;
# 3. Loop: If node is the first node in the stack of open elements, then set
# last to true, and, if the parser was originally created as part of the HTML
@ -2176,11 +2175,11 @@ class TreeBuilder {
# must run the following steps:
# 1. Generate implied end tags, except for p elements.
$this->stack->generateImpliedEndTags(["p"]);
$this->stack->generateImpliedEndTags("p");
# 2. If the current node is not a p element, then this is a parse error.
$currentNodeName = $this->stack->currentNodeName;
if ($currentNodeName !== 'p') {
$this->error(ParseError::UNEXPECTED_END_TAG, $currentNodeName);
$this->error(ParseError::UNEXPECTED_END_TAG);
}
# 3. Pop elements from the stack of open elements until a p element has been
# popped from the stack.
@ -2189,18 +2188,7 @@ class TreeBuilder {
protected function isElementSpecial(Element $element): bool {
$name = $element->nodeName;
$ns = $element->namespaceURI;
# The following elements have varying levels of special parsing rules: HTML’s
# address, applet, area, article, aside, base, basefont, bgsound, blockquote,
# body, br, button, caption, center, col, colgroup, dd, details, dir, div, dl,
# dt, embed, fieldset, figcaption, figure, footer, form, frame, frameset, h1,
# h2, h3, h4, h5, h6, head, header, hr, html, iframe, img, input, li, link,
# listing, main, marquee, meta, nav, noembed, noframes, noscript, object, ol, p,
# param, plaintext, pre, script, section, select, source, style, summary, table,
# tbody, td, template, textarea, tfoot, th, thead, title, tr, track, ul, wbr,
# xmp; MathML mi, MathML mo, MathML mn, MathML ms, MathML mtext, and MathML
# annotation-xml; and SVG foreignObject, SVG desc, and SVG title.
return (($ns === '' && ($name === 'address' || $name === 'applet' || $name === 'area' || $name === 'article' || $name === 'aside' || $name === 'base' || $name === 'basefont' || $name === 'bgsound' || $name === 'blockquote' || $name === 'body' || $name === 'br' || $name === 'button' || $name === 'caption' || $name === 'center' || $name === 'col' || $name === 'colgroup' || $name === 'dd' || $name === 'details' || $name === 'dir' || $name === 'div' || $name === 'dl' || $name === 'dt' || $name === 'embed' || $name === 'fieldset' || $name === 'figcaption' || $name === 'figure' || $name === 'footer' || $name === 'form' || $name === 'frame' || $name === 'frameset' || $name === 'h1' || $name === 'h2' || $name === 'h3' || $name === 'h4' || $name === 'h5' || $name === 'h6' || $name === 'head' || $name === 'header' || $name === 'hr' || $name === 'html' || $name === 'iframe' || $name === 'img' || $name === 'input' || $name === 'li' || $name === 'link' || $name === 'listing' || $name === 'main' || $name === 'marquee' || $name === 'meta' || $name === 'nav' || $name === 'noembed' || $name === 'noframes' || $name === 'noscript' || $name === 'object' || $name === 'ol' || $name === 'p' || $name === 'param' || $name === 'plaintext' || $name === 'pre' || $name === 'script' || $name === 'section' || $name === 'select' || $name === 'source' || $name === 'style' || $name === 'summary' || $name === 'table' || $name === 'tbody' || $name === 'td' || $name === 'template' || $name === 'textarea' || $name === 'tfoot' || $name === 'th' || $name === 'thead' || $name === 'title' || $name === 'tr' || $name === 'track' || $name === 'ul' || $name === 'wbr' || $name === 'xmp')) || ($ns === Parser::MATHML_NAMESPACE && ($name === 'mi' || $name === 'mo' || $name === 'mn' || $name === 'ms' || $name === 'mtext' || $name === 'annotation-xml')) || ($ns === Parser::SVG_NAMESPACE && ($name === 'foreignObject' || $name === 'desc' || $name === 'title')));
$ns = $element->namespaceURI ?? Parser::HTML_NAMESPACE;
return in_array($name, self::SPECIAL_ELEMENTS[$ns] ?? []);
}
}

Write Preview
Loading…
Cancel
Save