MensBeam
/
HTML-DOM
3
0
Fork 0
Code Issues Pull Requests Releases Activity
Browse Source

Fixed an issue with tokenization of attributes

split-manual
Dustin Wilson 6 years ago
parent
d95f3e37e4
commit
85894ed1ea
2 changed files with 85 additions and 86 deletions
Whitespace
Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL
Unified View
Diff Options
Show Stats Download Patch File Download Diff File
  1. 17
      lib/Token.php
  2. 154
      lib/Tokenizer.php

17
lib/Token.php
View File

@ -55,7 +55,7 @@ class StartTagToken extends TagToken {
} }
public function getAttribute(string $name) { public function getAttribute(string $name) {
$key = $this->getAttributeKey($name); $key = $this->_getAttributeKey($name);
return (isset($this->attributes[$key])) ? $this->attributes[$key] : null; return (isset($this->attributes[$key])) ? $this->attributes[$key] : null;
} }
@ -65,29 +65,30 @@ class StartTagToken extends TagToken {
} }
public function removeAttribute(string $name) { public function removeAttribute(string $name) {
unset($this->attributes[$this->getAttributeKey($name)]); unset($this->attributes[$this->_getAttributeKey($name)]);
} }
public function setAttribute(string $name, string $value, string $namespace = Parser::HTML_NAMESPACE) { public function setAttribute(string $name, string $value, string $namespace = Parser::HTML_NAMESPACE) {
$key = $this->_getAttributeKey($name); $key = $this->_getAttributeKey($name);
$attribute = new TokenAttr($name, $value, $namespace);
if (is_null($key)) { if (is_null($key)) {
$this->attributes[] = $attribute; $this->attributes[] = new TokenAttr($name, $value, $namespace);
} else { } else {
$this->attributes[$key] = $attribute; $attribute = &$this->attributes[$key];
$attribute->name = $name;
$attribute->value = $value;
$attribute->namespace = $namespace;
} }
} }
private function _getAttributeKey(string $name) { private function _getAttributeKey(string $name) {
$key = null;
foreach ($this->attributes as $key => $a) { foreach ($this->attributes as $key => $a) {
if ($a->name === $name) { if ($a->name === $name) {
break; return $key;
} }
} }
return $key; return null;
} }
} }

154
lib/Tokenizer.php
View File

@ -1474,9 +1474,13 @@ class Tokenizer {
# code point), and its value to the empty string. Switch to the attribute name # code point), and its value to the empty string. Switch to the attribute name
# state. # state.
// DEVIATION: Will use a buffer for the attribute name instead. // Need to add the current attribute to the token, if necessary.
$attributeName = strtolower($char); if ($attribute) {
$attributeValue = ''; $token->attributes[] = $attribute;
$attribute = null;
}
$attribute = new TokenAttr(strtolower($char), '');
$this->state = self::ATTRIBUTE_NAME_STATE; $this->state = self::ATTRIBUTE_NAME_STATE;
} }
# EOF # EOF
@ -1503,9 +1507,13 @@ class Tokenizer {
ParseError::trigger(ParseError::UNEXPECTED_CHARACTER, $char, 'attribute name'); ParseError::trigger(ParseError::UNEXPECTED_CHARACTER, $char, 'attribute name');
} }
// DEVIATION: Will use a buffer for the attribute name instead. // Need to add the current attribute to the token, if necessary.
$attributeName = $char; if ($attribute) {
$attributeValue = ''; $token->attributes[] = $attribute;
$attribute = null;
}
$attribute = new TokenAttr($char, '');
$this->state = self::ATTRIBUTE_NAME_STATE; $this->state = self::ATTRIBUTE_NAME_STATE;
} }
@ -1521,48 +1529,27 @@ class Tokenizer {
# "LF" (U+000A) # "LF" (U+000A)
# "FF" (U+000C) # "FF" (U+000C)
# U+0020 SPACE # U+0020 SPACE
if ($char === "\t" || $char === "\n" || $char === "\x0c" || $char === ' ') {
if ($token->hasAttribute($attributeName)) {
ParseError::trigger(ParseError::ATTRIBUTE_EXISTS, $attributeName);
}
# Switch to the after attribute name state.
$this->state = self::AFTER_ATTRIBUTE_NAME_STATE;
}
# "/" (U+002F) # "/" (U+002F)
elseif ($char === '/') { # U+003E GREATER-THAN SIGN (>)
if ($token->hasAttribute($attributeName)) { # EOF
ParseError::trigger(ParseError::ATTRIBUTE_EXISTS, $attributeName); if ($char === "\t" || $char === "\n" || $char === "\x0c" || $char === ' ' || $char === '/' || $char === '>' || $char === '') {
if ($token->hasAttribute($attribute->name)) {
ParseError::trigger(ParseError::ATTRIBUTE_EXISTS, $attribute->name);
} }
# Switch to the self-closing start tag state. # Reconsume in the after attribute name state.
$this->state = self::SELF_CLOSING_START_TAG_STATE; $this->data->unconsume();
$this->state = self::AFTER_ATTRIBUTE_NAME_STATE;
} }
# "=" (U+003D) # "=" (U+003D)
elseif ($char === '=') { elseif ($char === '=') {
if ($token->hasAttribute($attributeName)) { if ($token->hasAttribute($attribute->name)) {
ParseError::trigger(ParseError::ATTRIBUTE_EXISTS, $attributeName); ParseError::trigger(ParseError::ATTRIBUTE_EXISTS, $attribute->name);
} }
# Switch to the before attribute value state. # Switch to the before attribute value state.
$this->state = self::BEFORE_ATTRIBUTE_VALUE_STATE; $this->state = self::BEFORE_ATTRIBUTE_VALUE_STATE;
} }
# ">" (U+003E)
elseif ($char === '>') {
if ($token->hasAttribute($attributeName)) {
ParseError::trigger(ParseError::ATTRIBUTE_EXISTS, $attributeName);
}
# Switch to the data state. Emit the current tag token.
$this->state = self::DATA_STATE;
// Need to add the current attribute name and value to the token if necessary.
if ($attributeName) {
$token->setAttribute($attributeName, $attributeValue);
}
return $token;
}
# Uppercase ASCII letter # Uppercase ASCII letter
elseif (ctype_upper($char)) { elseif (ctype_upper($char)) {
# Append the lowercase version of the current input character (add 0x0020 to the # Append the lowercase version of the current input character (add 0x0020 to the
@ -1570,22 +1557,14 @@ class Tokenizer {
// OPTIMIZATION: Consume all characters that are uppercase ASCII letters to prevent // OPTIMIZATION: Consume all characters that are uppercase ASCII letters to prevent
// having to loop back through here every single time. // having to loop back through here every single time.
$attributeName .= strtolower($char.$this->data->consumeWhile(self::CTYPE_UPPER)); $attribute->name .= strtolower($char.$this->data->consumeWhile(self::CTYPE_UPPER));
}
# EOF
elseif ($char === '') {
# Parse error. Switch to the data state. Reconsume the EOF character.
ParseError::trigger(ParseError::UNEXPECTED_EOF, 'attribute name');
$this->state = self::DATA_STATE;
$this->data->unconsume();
} }
# U+0022 QUOTATION MARK (") # U+0022 QUOTATION MARK (")
# "'" (U+0027) # "'" (U+0027)
# "<" (U+003C) # "<" (U+003C)
# "=" (U+003D)
# Anything else # Anything else
else { else {
# Quotes, less than sign, equals: # Quotes, less than sign:
# Parse error. Treat it as per the "anything else" entry below. # Parse error. Treat it as per the "anything else" entry below.
# Anything else: # Anything else:
# Append the current input character to the current attribute's name. # Append the current input character to the current attribute's name.
@ -1598,7 +1577,7 @@ class Tokenizer {
// characters. // characters.
// OPTIMIZATION: Consume all characters that aren't listed above to prevent having // OPTIMIZATION: Consume all characters that aren't listed above to prevent having
// to loop back through here every single time. // to loop back through here every single time.
$attributeName .= $char.$this->data->consumeUntil("\t\n\x0c /=>\"'<".self::CTYPE_UPPER); $attribute->name .= $char.$this->data->consumeUntil("\t\n\x0c /=>\"'<".self::CTYPE_UPPER);
} }
# When the user agent leaves the attribute name state (and before emitting the tag # When the user agent leaves the attribute name state (and before emitting the tag
@ -1625,26 +1604,26 @@ class Tokenizer {
if ($char === "\t" || $char === "\n" || $char === "\x0c" || $char === ' ') { if ($char === "\t" || $char === "\n" || $char === "\x0c" || $char === ' ') {
# Ignore the character. # Ignore the character.
} }
# "/" (U+002F) # U+002F SOLIDUS (/)
elseif ($char === '/') { elseif ($char === '/') {
# Switch to the self-closing start tag state. # Switch to the self-closing start tag state.
$this->state = self::SELF_CLOSING_START_TAG_STATE; $this->state = self::SELF_CLOSING_START_TAG_STATE;
} }
# "=" (U+003D) # U+003D EQUALS SIGN (=)
elseif ($char === '=') { elseif ($char === '=') {
# Switch to the before attribute value state. # Switch to the before attribute value state.
$this->state = self::BEFORE_ATTRIBUTE_VALUE_STATE; $this->state = self::BEFORE_ATTRIBUTE_VALUE_STATE;
} }
# ">" (U+003E) # U+003E GREATER-THAN SIGN (>)
elseif ($char === '>') { elseif ($char === '>') {
# Switch to the data state. Emit the current tag token. // Need to add the current attribute to the token, if necessary.
$this->state = self::DATA_STATE; if ($attribute) {
$token->attributes[] = $attribute;
// Need to add the current attribute name and value to the token if necessary. $attribute = null;
if ($attributeName) {
$token->setAttribute($attributeName, $attributeValue);
} }
# Switch to the data state. Emit the current tag token.
$this->state = self::DATA_STATE;
return $token; return $token;
} }
# Uppercase ASCII letter # Uppercase ASCII letter
@ -1654,9 +1633,13 @@ class Tokenizer {
# code point), and its value to the empty string. Switch to the attribute name # code point), and its value to the empty string. Switch to the attribute name
# state. # state.
// DEVIATION: Will use a buffer for the attribute name instead. // Need to add the current attribute to the token, if necessary.
$attributeName = strtolower($char); if ($attribute) {
$attributeValue = ''; $token->attributes[] = $attribute;
$attribute = null;
}
$attribute = new TokenAttr(strtolower($char), '');
$this->state = self::ATTRIBUTE_NAME_STATE; $this->state = self::ATTRIBUTE_NAME_STATE;
} }
# EOF # EOF
@ -1683,8 +1666,13 @@ class Tokenizer {
ParseError::trigger(ParseError::UNEXPECTED_CHARACTER, $char, 'attribute name, attribute value, or tag end'); ParseError::trigger(ParseError::UNEXPECTED_CHARACTER, $char, 'attribute name, attribute value, or tag end');
} }
$attributeName = $char; // Need to add the current attribute to the token, if necessary.
$attributeValue = ''; if ($attribute) {
$token->attributes[] = $attribute;
$attribute = null;
}
$attribute = new TokenAttr($char, '');
$this->state = self::ATTRIBUTE_NAME_STATE; $this->state = self::ATTRIBUTE_NAME_STATE;
} }
@ -1726,9 +1714,10 @@ class Tokenizer {
ParseError::trigger(ParseError::UNEXPECTED_END_OF_TAG, 'attribute value'); ParseError::trigger(ParseError::UNEXPECTED_END_OF_TAG, 'attribute value');
$this->state = self::DATA_STATE; $this->state = self::DATA_STATE;
// Need to add the current attribute name and value to the token if necessary. // Need to add the current attribute to the token, if necessary.
if ($attributeName) { if ($attribute) {
$token->setAttribute($attributeName, $attributeValue); $token->attributes[] = $attribute;
$attribute = null;
} }
return $token; return $token;
@ -1755,7 +1744,7 @@ class Tokenizer {
ParseError::trigger(ParseError::UNEXPECTED_CHARACTER, $char, 'attribute value'); ParseError::trigger(ParseError::UNEXPECTED_CHARACTER, $char, 'attribute value');
} }
$attributeValue .= $char; $attribute->value .= $char;
$this->state = self::ATTRIBUTE_VALUE_UNQUOTED_STATE; $this->state = self::ATTRIBUTE_VALUE_UNQUOTED_STATE;
} }
@ -1786,7 +1775,7 @@ class Tokenizer {
// DEVIATION: This implementation does the character reference consuming in a // DEVIATION: This implementation does the character reference consuming in a
// function for which it is more suited for. // function for which it is more suited for.
$attributeValue .= $this->data->consumeCharacterReference('"', true); $attribute->value .= $this->data->consumeCharacterReference('"', true);
} }
# EOF # EOF
elseif ($char === '') { elseif ($char === '') {
@ -1800,7 +1789,7 @@ class Tokenizer {
# Append the current input character to the current attribute's value. # Append the current input character to the current attribute's value.
// OPTIMIZATION: Consume all characters that aren't listed above to prevent having // OPTIMIZATION: Consume all characters that aren't listed above to prevent having
// to loop back through here every single time. // to loop back through here every single time.
$attributeValue .= $char.$this->data->consumeUntil('"&'); $attribute->value .= $char.$this->data->consumeUntil('"&');
} }
continue; continue;
@ -1831,7 +1820,7 @@ class Tokenizer {
# DEVIATION: This implementation does the character reference consuming in a # DEVIATION: This implementation does the character reference consuming in a
# function for which it is more suited for. # function for which it is more suited for.
$attributeValue .= $this->data->consumeCharacterReference("'", true); $attribute->value .= $this->data->consumeCharacterReference("'", true);
} }
# EOF # EOF
elseif ($char === '') { elseif ($char === '') {
@ -1846,7 +1835,7 @@ class Tokenizer {
// OPTIMIZATION: Consume all characters that aren't listed above to prevent having // OPTIMIZATION: Consume all characters that aren't listed above to prevent having
// to loop back through here every single time. // to loop back through here every single time.
$attributeValue .= $char.$this->data->consumeUntil("'&"); $attribute->value .= $char.$this->data->consumeUntil("'&");
} }
continue; continue;
@ -1883,16 +1872,17 @@ class Tokenizer {
// DEVIATION: This implementation does the character reference consuming in a // DEVIATION: This implementation does the character reference consuming in a
// function for which it is more suited for. // function for which it is more suited for.
$attributeValue .= $this->data->consumeCharacterReference('>', true); $attribute->value .= $this->data->consumeCharacterReference('>', true);
} }
# ">" (U+003E) # ">" (U+003E)
elseif ($char === '>') { elseif ($char === '>') {
# Switch to the data state. Emit the current tag token. # Switch to the data state. Emit the current tag token.
$this->state = self::DATA_STATE; $this->state = self::DATA_STATE;
// Need to add the current attribute name and value to the token if necessary. // Need to add the current attribute to the token, if necessary.
if ($attributeName) { if ($attribute) {
$token->setAttribute($attributeName, $attributeValue); $token->attributes[] = $attribute;
$attribute = null;
} }
return $token; return $token;
@ -1921,7 +1911,7 @@ class Tokenizer {
// OPTIMIZATION: Consume all characters that aren't listed above to prevent having // OPTIMIZATION: Consume all characters that aren't listed above to prevent having
// to loop back through here every single time. // to loop back through here every single time.
$attributeValue .= $char.$this->data->consumeUntil("\t\n\x0c &>\"'<=`"); $attribute->value .= $char.$this->data->consumeUntil("\t\n\x0c &>\"'<=`");
} }
continue; continue;
@ -1950,9 +1940,10 @@ class Tokenizer {
# Switch to the data state. Emit the current tag token. # Switch to the data state. Emit the current tag token.
$this->state = self::DATA_STATE; $this->state = self::DATA_STATE;
// Need to add the current attribute name and value to the token if necessary. // Need to add the current attribute to the token, if necessary.
if ($attributeName) { if ($attribute) {
$token->setAttribute($attributeName, $attributeValue); $token->attributes[] = $attribute;
$attribute = null;
} }
return $token; return $token;
@ -1986,6 +1977,13 @@ class Tokenizer {
# Emit the current tag token. # Emit the current tag token.
$token->selfClosing = true; $token->selfClosing = true;
$this->state = self::DATA_STATE; $this->state = self::DATA_STATE;
// Need to add the current attribute to the token, if necessary.
if ($attribute) {
$token->attributes[] = $attribute;
$attribute = null;
}
return $token; return $token;
} }
# EOF # EOF

Write Preview
Loading…
Cancel
Save