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

Avoid implicit looping and switching 

The while loop has been replaced with gotos where appropriate, and
switching has been replaced with a series of if-blocks in line with the
same logic in the tokenizer.
ns
J. King 4 years ago
parent
e3a271f06b
commit
f8b9cf2c2b
1 changed files with 1134 additions and 1142 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. 122
      lib/TreeBuilder.php

122
lib/TreeBuilder.php
View File

@ -199,22 +199,18 @@ class TreeBuilder {
}
}
protected function parseTokenInHTMLContent(Token $token, int $insertionMode = null) {
protected function parseTokenInHTMLContent(Token $token, int $insertionMode = null): bool {
ProcessToken:
$insertionMode = $insertionMode ?? $this->insertionMode;
// Loop used when processing the token under different rules; always breaks.
$iterations = 0;
while (true) {
assert((function() use ($insertionMode) {
$mode = self::INSERTION_MODE_NAMES[$insertionMode] ?? $insertionMode;
$this->debugLog .= " Mode: $mode\n";
return true;
})());
assert($iterations++ < 50, new LoopException("Probable infinite loop detected in HTML content handling"));
# 13.2.6.4. The rules for parsing tokens in HTML content
switch ($insertionMode) {
# 13.2.6.4.1. The "initial" insertion mode
case self::INITIAL_MODE:
if ($insertionMode === self::INITIAL_MODE) {
# A character token that is one of U+0009 CHARACTER TABULATION, U+000A LINE FEED
# (LF), U+000C FORM FEED (FF), U+000D CARRIAGE RETURN (CR), or U+0020 SPACE
// OPTIMIZATION: Will check for multiple space characters at once as character
@ -371,12 +367,11 @@ class TreeBuilder {
# token.
$this->insertionMode = self::BEFORE_HTML_MODE;
$insertionMode = self::BEFORE_HTML_MODE;
continue 2;
goto ProcessToken;
};
}
break;
# 13.2.6.4.2. The "before html" insertion mode
case self::BEFORE_HTML_MODE:
elseif ($insertionMode === self::BEFORE_HTML_MODE) {
# A DOCTYPE token
if ($token instanceof DOCTYPEToken) {
$this->error(ParseError::UNEXPECTED_DOCTYPE);
@ -420,17 +415,16 @@ class TreeBuilder {
# Switch the insertion mode to "before head", then reprocess the token.
$this->insertionMode = self::BEFORE_HEAD_MODE;
$insertionMode = self::BEFORE_HEAD_MODE;
continue 2;
goto ProcessToken;
}
# The document element can end up being removed from the Document object, e.g.,
# by scripts; nothing in particular happens in such cases, content continues
# by scripts; nothing in particular happens in such cases, content goto ProcessTokens
# being appended to the nodes as described in the next section.
// Good to know. There's no scripting in this implementation, though.
break;
}
# 13.2.6.4.3. The "before head" insertion mode
case self::BEFORE_HEAD_MODE:
elseif ($insertionMode === self::BEFORE_HEAD_MODE) {
# A character token that is one of U+0009 CHARACTER TABULATION, U+000A LINE FEED
# (LF), U+000C FORM FEED (FF), U+000D CARRIAGE RETURN (CR), or U+0020 SPACE
// OPTIMIZATION: Will check for multiple space characters at once as character
@ -452,7 +446,7 @@ class TreeBuilder {
elseif ($token instanceof StartTagToken && $token->name === 'html') {
# Process the token using the rules for the "in body" insertion mode.
$insertionMode = self::IN_BODY_MODE;
continue 2;
goto ProcessToken;
}
# A start tag whose tag name is "head"
elseif ($token instanceof StartTagToken && $token->name === 'head') {
@ -484,12 +478,11 @@ class TreeBuilder {
$insertionMode = self::IN_HEAD_MODE;
# Reprocess the current token.
continue 2;
goto ProcessToken;
}
}
break;
# 13.2.6.4.4. The "in head" insertion mode
case self::IN_HEAD_MODE:
elseif ($insertionMode === self::IN_HEAD_MODE) {
# A character token that is one of U+0009 CHARACTER TABULATION, U+000A LINE FEED
# (LF), U+000C FORM FEED (FF), U+000D CARRIAGE RETURN (CR), or U+0020 SPACE
// OPTIMIZATION: Will check for multiple space characters at once as character
@ -513,7 +506,7 @@ class TreeBuilder {
if ($token->name === 'html') {
# Process the token using the rules for the "in body" insertion mode.
$insertionMode = self::IN_BODY_MODE;
continue 2;
goto ProcessToken;
}
# A start tag whose tag name is one of: "base", "basefont", "bgsound", "link"
elseif ($token->name === 'base' || $token->name === 'basefont' || $token->name === 'bgsound' || $token->name === 'link') {
@ -632,7 +625,7 @@ class TreeBuilder {
$this->insertionMode = self::AFTER_HEAD_MODE;
$insertionMode = self::AFTER_HEAD_MODE;
# Reprocess the token.
continue 2;
goto ProcessToken;
}
}
elseif ($token instanceof EndTagToken) {
@ -655,7 +648,7 @@ class TreeBuilder {
$this->insertionMode = self::AFTER_HEAD_MODE;
$insertionMode = self::AFTER_HEAD_MODE;
# Reprocess the token.
continue 2;
goto ProcessToken;
}
# An end tag whose tag name is "template"
elseif ($token->name === 'template') {
@ -702,12 +695,11 @@ class TreeBuilder {
$this->insertionMode = self::AFTER_HEAD_MODE;
$insertionMode = self::AFTER_HEAD_MODE;
# Reprocess the token.
continue 2;
goto ProcessToken;
}
}
break;
# 13.2.6.4.5. The "in head noscript" insertion mode
case self::IN_HEAD_NOSCRIPT_MODE:
elseif ($insertionMode === self::IN_HEAD_NOSCRIPT_MODE) {
# DOCTYPE token
if ($token instanceof DOCTYPEToken) {
# Parse error.
@ -718,14 +710,14 @@ class TreeBuilder {
if ($token->name === 'html') {
# Process the token using the rules for the "in body" insertion mode.
$insertionMode = self::IN_BODY_MODE;
continue 2;
goto ProcessToken;
}
# A start tag whose tag name is one of: "basefont", "bgsound", "link", "meta",
# "noframes", "style"
elseif ($token->name === 'basefont' || $token->name === 'bgsound' || $token->name === 'link' || $token->name === 'meta' || $token->name === 'noframes' || $token->name === 'style'){
# Process the token using the rules for the "in head" insertion mode.
$insertionMode = self::IN_HEAD_MODE;
continue 2;
goto ProcessToken;
}
# A start tag whose tag name is one of: "head", "noscript"
elseif ($token->name === 'head' || $token->name === 'noscript') {
@ -745,7 +737,7 @@ class TreeBuilder {
$this->insertionMode = self::IN_HEAD_MODE;
$insertionMode = self::IN_HEAD_MODE;
# Reprocess the token.
continue 2;
goto ProcessToken;
}
}
elseif ($token instanceof EndTagToken) {
@ -770,7 +762,7 @@ class TreeBuilder {
$this->insertionMode = self::IN_HEAD_MODE;
$insertionMode = self::IN_HEAD_MODE;
# Reprocess the token.
continue 2;
goto ProcessToken;
}
# Any other end tag
else {
@ -787,7 +779,7 @@ class TreeBuilder {
$token instanceof CommentToken) {
# Process the token using the rules for the "in head" insertion mode.
$insertionMode = self::IN_HEAD_MODE;
continue 2;
goto ProcessToken;
}
# Anything else
else {
@ -800,12 +792,11 @@ class TreeBuilder {
$this->insertionMode = self::IN_HEAD_MODE;
$insertionMode = self::IN_HEAD_MODE;
# Reprocess the token.
continue 2;
goto ProcessToken;
}
}
break;
# 13.2.6.4.6. The "after head" insertion mode
case self::AFTER_HEAD_MODE:
elseif ($insertionMode === self::AFTER_HEAD_MODE) {
# A character token that is one of U+0009 CHARACTER TABULATION, U+000A LINE FEED
# (LF), U+000C FORM FEED (FF), U+000D CARRIAGE RETURN (CR), or U+0020 SPACE
// OPTIMIZATION: Will check for multiple space characters at once as character
@ -829,7 +820,7 @@ class TreeBuilder {
if ($token->name === 'html') {
# Process the token using the rules for the "in body" insertion mode.
$insertionMode = self::IN_BODY_MODE;
continue 2;
goto ProcessToken;
}
# A start tag whose tag name is "body"
elseif ($token->name === 'body') {
@ -879,7 +870,7 @@ class TreeBuilder {
$this->insertionMode = self::IN_BODY_MODE;
$insertionMode = self::IN_BODY_MODE;
# Reprocess the current token.
continue 2;
goto ProcessToken;
}
}
elseif ($token instanceof EndTagToken) {
@ -887,7 +878,7 @@ class TreeBuilder {
if ($token->name === 'template') {
# Process the token using the rules for the "in head" insertion mode.
$insertionMode = self::IN_HEAD_MODE;
continue 2;
goto ProcessToken;
}
# An end tag whose tag name is one of: "body", "html", "br"
elseif ($token->name === 'body' || $token->name === 'html' || $token->name === 'br') {
@ -899,7 +890,7 @@ class TreeBuilder {
$this->insertionMode = self::IN_BODY_MODE;
$insertionMode = self::IN_BODY_MODE;
# Reprocess the current token.
continue 2;
goto ProcessToken;
}
# Any other end tag
else {
@ -915,19 +906,25 @@ class TreeBuilder {
$this->insertionMode = self::IN_BODY_MODE;
$insertionMode = self::IN_BODY_MODE;
# Reprocess the current token.
continue 2;
goto ProcessToken;
}
}
break;
# 13.2.6.4.7. The "in body" insertion mode
case self::IN_BODY_MODE:
if ($token instanceof CharacterToken) {
elseif ($insertionMode === self::IN_BODY_MODE) {
# A character token that is U+0000 NULL
if ($token instanceof CharacterToken && $token->data === "\0") {
# Parse error. Ignore the token
// DEVIATION: the parse error is already reported by the tokenizer;
// this is probably an oversight in the specification, so we don't
// report it a second time
}
# A character token that is one of U+0009 CHARACTER TABULATION, U+000A LINE FEED
# (LF), U+000C FORM FEED (FF), U+000D CARRIAGE RETURN (CR), or U+0020 SPACE
#
# Any other character token
// Space characters and any other characters are exactly the same except any
// other characters sets the frameset-ok flag to "not ok".
elseif ($token instanceof CharacterToken) {
# Reconstruct the active formatting elements, if any.
$this->activeFormattingElementsList->reconstruct();
@ -975,7 +972,7 @@ class TreeBuilder {
elseif ($token->name === 'base' || $token->name === 'basefont' || $token->name === 'bgsound' || $token->name === 'link' || $token->name === 'meta' || $token->name === 'noframes' || $token->name === 'script' || $token->name === 'style' || $token->name === 'template' || $token->name === 'title') {
# Process the token using the rules for the "in head" insertion mode.
$insertionMode = self::IN_HEAD_MODE;
continue 2;
goto ProcessToken;
}
# A start tag whose tag name is "body"
elseif ($token->name === 'body') {
@ -1093,7 +1090,7 @@ class TreeBuilder {
// Process the next token
$token = $nextToken;
continue 2;
goto ProcessToken;
}
# A start tag whose tag name is "form"
elseif ($token->name === 'form') {
@ -1145,13 +1142,13 @@ class TreeBuilder {
$this->stack->popUntil('li');
# 4. Jump to the step labeled Done below.
break;
return true;
}
# 4. If node is in the special category, but is not an address, div, or p
# element, then jump to the step labeled Done below.
if ($nodeName !== 'address' && $nodeName !== 'div' && $nodeName !== 'p' && $this->isElementSpecial($node)) {
break;
return true;
}
# 5. Otherwise, set node to the previous entry in the stack of open elements and
@ -1196,13 +1193,13 @@ class TreeBuilder {
$this->stack->popUntil(['dd', 'dt']);
# 4. Jump to the step labeled Done below.
break;
return true;
}
# 5. If node is in the special category, but is not an address, div, or p
# element, then jump to the step labeled Done below.
if ($nodeName !== 'address' && $nodeName !== 'div' && $nodeName !== 'p' && $this->isElementSpecial($node)) {
break;
return true;
}
# 6. Otherwise, set node to the previous entry in the stack of open elements and
@ -1264,7 +1261,7 @@ class TreeBuilder {
if ($token->name === 'template') {
# Process the token using the rules for the "in head" insertion mode.
$insertionMode = self::IN_HEAD_MODE;
continue 2;
goto ProcessToken;
}
# An end tag whose tag name is "body"
# An end tag whose tag name is "html"
@ -1290,7 +1287,7 @@ class TreeBuilder {
return false;
}) !== -1) {
$this->error(ParseError::UNEXPECTED_END_TAG, 'body');
break;
return true;
}
# Switch the insertion mode to "after body".
@ -1300,7 +1297,7 @@ class TreeBuilder {
// an html end tag the token is reprocessed.
if ($token->name === 'html') {
# Reprocess the token.
continue 2;
goto ProcessToken;
}
}
}
@ -1382,7 +1379,7 @@ class TreeBuilder {
# 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) {
$insertionMode = self::IN_TEMPLATE_MODE;
continue 2;
goto ProcessToken;
}
# Otherwise, follow these steps:
@ -1400,26 +1397,21 @@ class TreeBuilder {
return false;
}) !== -1) {
$this->error(ParseError::UNEXPECTED_END_TAG, 'body');
break;
return true;
}
# 2. Stop parsing.
// Abort!
}
break;
}
break;
// IMPLEMENTATION PENDING
else {
throw new \Exception("NOT IMPLEMENTED");
}
return true;
}
protected function parseTokenInForeignContent(Token $token): bool {
if (self::$debug) {
echo "Foreign Content\n";
}
$currentNode = $this->stack->currentNode;
$currentNodeName = $this->stack->currentNodeName;
$currentNodeNamespace = $this->stack->currentNodeNamespace;

Write Preview
Loading…
Cancel
Save