diff --git a/lib/Parser/AttributeSetter.php b/lib/Parser/AttributeSetter.php
new file mode 100644
index 0000000..9ad63f1
--- /dev/null
+++ b/lib/Parser/AttributeSetter.php
@@ -0,0 +1,50 @@
+appendChild($d->createElement("html"));
+ try {
+ $a = $d->createAttributeNS($namespaceURI, $qualifiedName);
+ // @codeCoverageIgnoreStart
+ } catch (\DOMException $e) {
+ // The attribute name is invalid for XML 1.0 Second Edition
+ // Replace any offending characters with "UHHHHHH" where H are the
+ // uppercase hexadecimal digits of the character's code point
+ // NOTE: This case is never encountered by the parser
+ $qualifiedName = self::coerceName($qualifiedName, true);
+ $a = $d->createAttributeNS($namespaceURI, $qualifiedName);
+ }
+ // @codeCoverageIgnoreEnd
+ $a->value = self::escapeString($value, true);
+ $element->setAttributeNodeNS($element->ownerDocument->importNode($a));
+ } else {
+ try {
+ $element->setAttributeNS($namespaceURI, $qualifiedName, $value);
+ } catch (\DOMException $e) {
+ // The attribute name is invalid for XML 1.0 Second Edition
+ // Replace any offending characters with "UHHHHHH" where H are the
+ // uppercase hexadecimal digits of the character's code point
+ $qualifiedName = self::coerceName($qualifiedName, ($namespaceURI !== null));
+ $element->setAttributeNS($namespaceURI, $qualifiedName, $value);
+ $this->mangledAttributes = true;
+ }
+ if ($qualifiedName === "id" && $namespaceURI === null) {
+ $element->setIdAttribute($qualifiedName, true);
+ }
+ }
+ }
+}
diff --git a/lib/Parser/Data.php b/lib/Parser/Data.php
index 181c3e9..d0fa6e7 100644
--- a/lib/Parser/Data.php
+++ b/lib/Parser/Data.php
@@ -112,15 +112,17 @@ class Data {
else {
$char = "\n";
}
+ } elseif ($char === '') {
+ $this->eof = true;
}
// unless we're peeking, track line and column position, and whether we've hit EOF
if ($this->track) {
- if ($char === "\n") {
+ if ($char === '') {
+ // do nothing
+ } elseif ($char === "\n") {
$this->newlines[$this->data->posChar()] = $this->_column;
$this->_column = 0;
$this->_line++;
- } elseif ($char === '') {
- $this->eof = true;
} else {
$this->_column++;
$len = strlen($char);
@@ -155,7 +157,7 @@ class Data {
$this->error(ParseError::NONCHARACTER_IN_INPUT_STREAM);
$this->lastError = $here;
} elseif ($tail === 0xBFBD && $this->data->posErr === $here) {
- $this->error(ParseError::NONCHARACTER_IN_INPUT_STREAM, $this->data->posByte);
+ $this->error(ParseError::NONCHARACTER_IN_INPUT_STREAM);
$this->lastError = $here;
}
}
@@ -184,20 +186,21 @@ class Data {
$here = $this->data->posChar();
// if the previous character was a normalized CR+LF pair, we need to go back two
if (isset($this->normalized[$here])) {
- $this->data->seek(-1);
+ // NOTE: This case is never encountered by the parser
+ $this->data->seek(-1); // @codeCoverageIgnore
}
// recalculate line and column positions, if requested
if ($retreatPointer && $this->track) {
- $col = $this->newlines[$here] ?? 0;
- if ($col) {
- $this->_column = $col;
+ // NOTE: These cases are never encountered by the parser
+ // @codeCoverageIgnoreStart
+ if ($col = $this->newlines[$here] ?? 0) {
+ $this->_column = $col + 1;
$this->_line--;
- } else {
+ } elseif ($this->astrals[$here] ?? false) {
$this->_column--;
- if ($this->astrals[$here] ?? false) {
- $this->_column--;
- }
}
+ // @codeCoverageIgnoreEnd
+ $this->_column--;
}
$this->data->seek(-1);
}
@@ -252,21 +255,22 @@ class Data {
do {
// If the current position is the start of a line,
// get the column position of the end of the previous line
+ // NOTE: These cases are never encountered by the parser
+ // @codeCoverageIgnoreStart
if (isset($this->newlines[$pos])) {
$line--;
- $col = $this->newlines[$pos];
+ $col = $this->newlines[$pos] + 1;
// If the newline was a normalized CR+LF pair,
// go back one extra character
if (isset($this->normalized[$pos])) {
$pos--;
}
- } else {
- $col--;
+ } elseif ($this->astrals[$pos] ?? false) {
// supplementary plane characters count as two
- if ($this->astrals[$pos] ?? false) {
- $this->_column--;
- }
+ $col--;
}
+ // @codeCoverageIgnoreEnd
+ $col--;
$pos--;
} while (++$relativePos < 0);
return [$line, $col];
@@ -274,19 +278,23 @@ class Data {
return [$this->_line, $this->_column + $relativePos];
}
} else {
- return [0, 0];
+ return [0, 0]; // @codeCoverageIgnore
}
}
public function __get($property) {
switch ($property) {
- case 'column': return $this->_column;
- break;
- case 'line': return $this->_line;
- break;
- case 'pointer': return $this->data->posChar();
- break;
- default: return null;
+ case 'column':
+ return $this->_column; // @codeCoverageIgnore
+ break;
+ case 'line':
+ return $this->_line; // @codeCoverageIgnore
+ break;
+ case 'pointer':
+ return $this->data->posChar();
+ break;
+ default:
+ return null; // @codeCoverageIgnore
}
}
diff --git a/lib/Parser/Exception.php b/lib/Parser/Exception.php
index b9a1164..10b48ee 100644
--- a/lib/Parser/Exception.php
+++ b/lib/Parser/Exception.php
@@ -11,10 +11,13 @@ class Exception extends \Exception {
public const FAILED_CREATING_DOCUMENT = 102;
public const INVALID_DOCUMENT_CLASS = 103;
+ public const UNSUPPORTED_NODE_TYPE = 201;
+
protected static $messages = [
101 => 'Fragment\'s quirks mode must be one of Parser::NO_QUIRKS_MODE, Parser::LIMITED_QUIRKS_MODE, or Parser::QUIRKS_MODE',
102 => 'Unable to create instance of configured document class "%s"',
103 => 'Configured document class "%s" must be a subclass of \DOMDocument',
+ 201 => 'Unable to serialize unsupported node type %s',
];
public function __construct(int $code, array $args = [], \Throwable $previous = null) {
diff --git a/lib/Parser/NameCoercion.php b/lib/Parser/NameCoercion.php
index eba16bf..9a0dc7b 100644
--- a/lib/Parser/NameCoercion.php
+++ b/lib/Parser/NameCoercion.php
@@ -10,7 +10,7 @@ use MensBeam\Intl\Encoding\UTF8;
trait NameCoercion {
/** @codeCoverageIgnore */
- protected function coerceNameFifthEdition(string $name): string {
+ protected static function coerceNameFifthEdition(string $name): string {
// This matches the inverse of the production of NameChar in XML 1.0 Fifth Edition,
// with the added exclusion of ":" from allowed characters
// See https://www.w3.org/TR/REC-xml/#NT-NameStartChar
@@ -30,27 +30,34 @@ trait NameCoercion {
return $name;
}
- protected function coerceName(string $name): string {
- // This matches the inverse of the production of Name in XML 1.0 Fourth Edition,
- // with the added exclusion of ":" from allowed characters
- // See https://www.w3.org/TR/2006/REC-xml-20060816/#NT-NameChar
- preg_match_all('/[^_\.\-\x{41}-\x{5A}\x{61}-\x{7A}\x{C0}-\x{D6}\x{D8}-\x{F6}\x{F8}-\x{FF}\x{100}-\x{131}\x{134}-\x{13E}\x{141}-\x{148}\x{14A}-\x{17E}\x{180}-\x{1C3}\x{1CD}-\x{1F0}\x{1F4}-\x{1F5}\x{1FA}-\x{217}\x{250}-\x{2A8}\x{2BB}-\x{2C1}\x{386}\x{388}-\x{38A}\x{38C}\x{38E}-\x{3A1}\x{3A3}-\x{3CE}\x{3D0}-\x{3D6}\x{3DA}\x{3DC}\x{3DE}\x{3E0}\x{3E2}-\x{3F3}\x{401}-\x{40C}\x{40E}-\x{44F}\x{451}-\x{45C}\x{45E}-\x{481}\x{490}-\x{4C4}\x{4C7}-\x{4C8}\x{4CB}-\x{4CC}\x{4D0}-\x{4EB}\x{4EE}-\x{4F5}\x{4F8}-\x{4F9}\x{531}-\x{556}\x{559}\x{561}-\x{586}\x{5D0}-\x{5EA}\x{5F0}-\x{5F2}\x{621}-\x{63A}\x{641}-\x{64A}\x{671}-\x{6B7}\x{6BA}-\x{6BE}\x{6C0}-\x{6CE}\x{6D0}-\x{6D3}\x{6D5}\x{6E5}-\x{6E6}\x{905}-\x{939}\x{93D}\x{958}-\x{961}\x{985}-\x{98C}\x{98F}-\x{990}\x{993}-\x{9A8}\x{9AA}-\x{9B0}\x{9B2}\x{9B6}-\x{9B9}\x{9DC}-\x{9DD}\x{9DF}-\x{9E1}\x{9F0}-\x{9F1}\x{A05}-\x{A0A}\x{A0F}-\x{A10}\x{A13}-\x{A28}\x{A2A}-\x{A30}\x{A32}-\x{A33}\x{A35}-\x{A36}\x{A38}-\x{A39}\x{A59}-\x{A5C}\x{A5E}\x{A72}-\x{A74}\x{A85}-\x{A8B}\x{A8D}\x{A8F}-\x{A91}\x{A93}-\x{AA8}\x{AAA}-\x{AB0}\x{AB2}-\x{AB3}\x{AB5}-\x{AB9}\x{ABD}\x{AE0}\x{B05}-\x{B0C}\x{B0F}-\x{B10}\x{B13}-\x{B28}\x{B2A}-\x{B30}\x{B32}-\x{B33}\x{B36}-\x{B39}\x{B3D}\x{B5C}-\x{B5D}\x{B5F}-\x{B61}\x{B85}-\x{B8A}\x{B8E}-\x{B90}\x{B92}-\x{B95}\x{B99}-\x{B9A}\x{B9C}\x{B9E}-\x{B9F}\x{BA3}-\x{BA4}\x{BA8}-\x{BAA}\x{BAE}-\x{BB5}\x{BB7}-\x{BB9}\x{C05}-\x{C0C}\x{C0E}-\x{C10}\x{C12}-\x{C28}\x{C2A}-\x{C33}\x{C35}-\x{C39}\x{C60}-\x{C61}\x{C85}-\x{C8C}\x{C8E}-\x{C90}\x{C92}-\x{CA8}\x{CAA}-\x{CB3}\x{CB5}-\x{CB9}\x{CDE}\x{CE0}-\x{CE1}\x{D05}-\x{D0C}\x{D0E}-\x{D10}\x{D12}-\x{D28}\x{D2A}-\x{D39}\x{D60}-\x{D61}\x{E01}-\x{E2E}\x{E30}\x{E32}-\x{E33}\x{E40}-\x{E45}\x{E81}-\x{E82}\x{E84}\x{E87}-\x{E88}\x{E8A}\x{E8D}\x{E94}-\x{E97}\x{E99}-\x{E9F}\x{EA1}-\x{EA3}\x{EA5}\x{EA7}\x{EAA}-\x{EAB}\x{EAD}-\x{EAE}\x{EB0}\x{EB2}-\x{EB3}\x{EBD}\x{EC0}-\x{EC4}\x{F40}-\x{F47}\x{F49}-\x{F69}\x{10A0}-\x{10C5}\x{10D0}-\x{10F6}\x{1100}\x{1102}-\x{1103}\x{1105}-\x{1107}\x{1109}\x{110B}-\x{110C}\x{110E}-\x{1112}\x{113C}\x{113E}\x{1140}\x{114C}\x{114E}\x{1150}\x{1154}-\x{1155}\x{1159}\x{115F}-\x{1161}\x{1163}\x{1165}\x{1167}\x{1169}\x{116D}-\x{116E}\x{1172}-\x{1173}\x{1175}\x{119E}\x{11A8}\x{11AB}\x{11AE}-\x{11AF}\x{11B7}-\x{11B8}\x{11BA}\x{11BC}-\x{11C2}\x{11EB}\x{11F0}\x{11F9}\x{1E00}-\x{1E9B}\x{1EA0}-\x{1EF9}\x{1F00}-\x{1F15}\x{1F18}-\x{1F1D}\x{1F20}-\x{1F45}\x{1F48}-\x{1F4D}\x{1F50}-\x{1F57}\x{1F59}\x{1F5B}\x{1F5D}\x{1F5F}-\x{1F7D}\x{1F80}-\x{1FB4}\x{1FB6}-\x{1FBC}\x{1FBE}\x{1FC2}-\x{1FC4}\x{1FC6}-\x{1FCC}\x{1FD0}-\x{1FD3}\x{1FD6}-\x{1FDB}\x{1FE0}-\x{1FEC}\x{1FF2}-\x{1FF4}\x{1FF6}-\x{1FFC}\x{2126}\x{212A}-\x{212B}\x{212E}\x{2180}-\x{2182}\x{3041}-\x{3094}\x{30A1}-\x{30FA}\x{3105}-\x{312C}\x{AC00}-\x{D7A3}\x{4E00}-\x{9FA5}\x{3007}\x{3021}-\x{3029}\x{30}-\x{39}\x{660}-\x{669}\x{6F0}-\x{6F9}\x{966}-\x{96F}\x{9E6}-\x{9EF}\x{A66}-\x{A6F}\x{AE6}-\x{AEF}\x{B66}-\x{B6F}\x{BE7}-\x{BEF}\x{C66}-\x{C6F}\x{CE6}-\x{CEF}\x{D66}-\x{D6F}\x{E50}-\x{E59}\x{ED0}-\x{ED9}\x{F20}-\x{F29}\x{300}-\x{345}\x{360}-\x{361}\x{483}-\x{486}\x{591}-\x{5A1}\x{5A3}-\x{5B9}\x{5BB}-\x{5BD}\x{5BF}\x{5C1}-\x{5C2}\x{5C4}\x{64B}-\x{652}\x{670}\x{6D6}-\x{6DC}\x{6DD}-\x{6DF}\x{6E0}-\x{6E4}\x{6E7}-\x{6E8}\x{6EA}-\x{6ED}\x{901}-\x{903}\x{93C}\x{93E}-\x{94C}\x{94D}\x{951}-\x{954}\x{962}-\x{963}\x{981}-\x{983}\x{9BC}\x{9BE}\x{9BF}\x{9C0}-\x{9C4}\x{9C7}-\x{9C8}\x{9CB}-\x{9CD}\x{9D7}\x{9E2}-\x{9E3}\x{A02}\x{A3C}\x{A3E}\x{A3F}\x{A40}-\x{A42}\x{A47}-\x{A48}\x{A4B}-\x{A4D}\x{A70}-\x{A71}\x{A81}-\x{A83}\x{ABC}\x{ABE}-\x{AC5}\x{AC7}-\x{AC9}\x{ACB}-\x{ACD}\x{B01}-\x{B03}\x{B3C}\x{B3E}-\x{B43}\x{B47}-\x{B48}\x{B4B}-\x{B4D}\x{B56}-\x{B57}\x{B82}-\x{B83}\x{BBE}-\x{BC2}\x{BC6}-\x{BC8}\x{BCA}-\x{BCD}\x{BD7}\x{C01}-\x{C03}\x{C3E}-\x{C44}\x{C46}-\x{C48}\x{C4A}-\x{C4D}\x{C55}-\x{C56}\x{C82}-\x{C83}\x{CBE}-\x{CC4}\x{CC6}-\x{CC8}\x{CCA}-\x{CCD}\x{CD5}-\x{CD6}\x{D02}-\x{D03}\x{D3E}-\x{D43}\x{D46}-\x{D48}\x{D4A}-\x{D4D}\x{D57}\x{E31}\x{E34}-\x{E3A}\x{E47}-\x{E4E}\x{EB1}\x{EB4}-\x{EB9}\x{EBB}-\x{EBC}\x{EC8}-\x{ECD}\x{F18}-\x{F19}\x{F35}\x{F37}\x{F39}\x{F3E}\x{F3F}\x{F71}-\x{F84}\x{F86}-\x{F8B}\x{F90}-\x{F95}\x{F97}\x{F99}-\x{FAD}\x{FB1}-\x{FB7}\x{FB9}\x{20D0}-\x{20DC}\x{20E1}\x{302A}-\x{302F}\x{3099}\x{309A}\x{B7}\x{2D0}\x{2D1}\x{387}\x{640}\x{E46}\x{EC6}\x{3005}\x{3031}-\x{3035}\x{309D}-\x{309E}\x{30FC}-\x{30FE}]/u', $name, $m);
- foreach (array_unique($m[0], \SORT_STRING) as $c) {
- $o = (new UTF8($c))->nextCode();
- $esc = "U".str_pad(strtoupper(dechex($o)), 6, "0", \STR_PAD_LEFT);
- $name = str_replace($c, $esc, $name);
- }
- // Apply stricter rules to the first character
- if (preg_match('/^[^_\x{41}-\x{5A}\x{61}-\x{7A}\x{C0}-\x{D6}\x{D8}-\x{F6}\x{F8}-\x{FF}\x{100}-\x{131}\x{134}-\x{13E}\x{141}-\x{148}\x{14A}-\x{17E}\x{180}-\x{1C3}\x{1CD}-\x{1F0}\x{1F4}-\x{1F5}\x{1FA}-\x{217}\x{250}-\x{2A8}\x{2BB}-\x{2C1}\x{386}\x{388}-\x{38A}\x{38C}\x{38E}-\x{3A1}\x{3A3}-\x{3CE}\x{3D0}-\x{3D6}\x{3DA}\x{3DC}\x{3DE}\x{3E0}\x{3E2}-\x{3F3}\x{401}-\x{40C}\x{40E}-\x{44F}\x{451}-\x{45C}\x{45E}-\x{481}\x{490}-\x{4C4}\x{4C7}-\x{4C8}\x{4CB}-\x{4CC}\x{4D0}-\x{4EB}\x{4EE}-\x{4F5}\x{4F8}-\x{4F9}\x{531}-\x{556}\x{559}\x{561}-\x{586}\x{5D0}-\x{5EA}\x{5F0}-\x{5F2}\x{621}-\x{63A}\x{641}-\x{64A}\x{671}-\x{6B7}\x{6BA}-\x{6BE}\x{6C0}-\x{6CE}\x{6D0}-\x{6D3}\x{6D5}\x{6E5}-\x{6E6}\x{905}-\x{939}\x{93D}\x{958}-\x{961}\x{985}-\x{98C}\x{98F}-\x{990}\x{993}-\x{9A8}\x{9AA}-\x{9B0}\x{9B2}\x{9B6}-\x{9B9}\x{9DC}-\x{9DD}\x{9DF}-\x{9E1}\x{9F0}-\x{9F1}\x{A05}-\x{A0A}\x{A0F}-\x{A10}\x{A13}-\x{A28}\x{A2A}-\x{A30}\x{A32}-\x{A33}\x{A35}-\x{A36}\x{A38}-\x{A39}\x{A59}-\x{A5C}\x{A5E}\x{A72}-\x{A74}\x{A85}-\x{A8B}\x{A8D}\x{A8F}-\x{A91}\x{A93}-\x{AA8}\x{AAA}-\x{AB0}\x{AB2}-\x{AB3}\x{AB5}-\x{AB9}\x{ABD}\x{AE0}\x{B05}-\x{B0C}\x{B0F}-\x{B10}\x{B13}-\x{B28}\x{B2A}-\x{B30}\x{B32}-\x{B33}\x{B36}-\x{B39}\x{B3D}\x{B5C}-\x{B5D}\x{B5F}-\x{B61}\x{B85}-\x{B8A}\x{B8E}-\x{B90}\x{B92}-\x{B95}\x{B99}-\x{B9A}\x{B9C}\x{B9E}-\x{B9F}\x{BA3}-\x{BA4}\x{BA8}-\x{BAA}\x{BAE}-\x{BB5}\x{BB7}-\x{BB9}\x{C05}-\x{C0C}\x{C0E}-\x{C10}\x{C12}-\x{C28}\x{C2A}-\x{C33}\x{C35}-\x{C39}\x{C60}-\x{C61}\x{C85}-\x{C8C}\x{C8E}-\x{C90}\x{C92}-\x{CA8}\x{CAA}-\x{CB3}\x{CB5}-\x{CB9}\x{CDE}\x{CE0}-\x{CE1}\x{D05}-\x{D0C}\x{D0E}-\x{D10}\x{D12}-\x{D28}\x{D2A}-\x{D39}\x{D60}-\x{D61}\x{E01}-\x{E2E}\x{E30}\x{E32}-\x{E33}\x{E40}-\x{E45}\x{E81}-\x{E82}\x{E84}\x{E87}-\x{E88}\x{E8A}\x{E8D}\x{E94}-\x{E97}\x{E99}-\x{E9F}\x{EA1}-\x{EA3}\x{EA5}\x{EA7}\x{EAA}-\x{EAB}\x{EAD}-\x{EAE}\x{EB0}\x{EB2}-\x{EB3}\x{EBD}\x{EC0}-\x{EC4}\x{F40}-\x{F47}\x{F49}-\x{F69}\x{10A0}-\x{10C5}\x{10D0}-\x{10F6}\x{1100}\x{1102}-\x{1103}\x{1105}-\x{1107}\x{1109}\x{110B}-\x{110C}\x{110E}-\x{1112}\x{113C}\x{113E}\x{1140}\x{114C}\x{114E}\x{1150}\x{1154}-\x{1155}\x{1159}\x{115F}-\x{1161}\x{1163}\x{1165}\x{1167}\x{1169}\x{116D}-\x{116E}\x{1172}-\x{1173}\x{1175}\x{119E}\x{11A8}\x{11AB}\x{11AE}-\x{11AF}\x{11B7}-\x{11B8}\x{11BA}\x{11BC}-\x{11C2}\x{11EB}\x{11F0}\x{11F9}\x{1E00}-\x{1E9B}\x{1EA0}-\x{1EF9}\x{1F00}-\x{1F15}\x{1F18}-\x{1F1D}\x{1F20}-\x{1F45}\x{1F48}-\x{1F4D}\x{1F50}-\x{1F57}\x{1F59}\x{1F5B}\x{1F5D}\x{1F5F}-\x{1F7D}\x{1F80}-\x{1FB4}\x{1FB6}-\x{1FBC}\x{1FBE}\x{1FC2}-\x{1FC4}\x{1FC6}-\x{1FCC}\x{1FD0}-\x{1FD3}\x{1FD6}-\x{1FDB}\x{1FE0}-\x{1FEC}\x{1FF2}-\x{1FF4}\x{1FF6}-\x{1FFC}\x{2126}\x{212A}-\x{212B}\x{212E}\x{2180}-\x{2182}\x{3041}-\x{3094}\x{30A1}-\x{30FA}\x{3105}-\x{312C}\x{AC00}-\x{D7A3}\x{4E00}-\x{9FA5}\x{3007}\x{3021}-\x{3029}]/u', $name, $m)) {
- $c = (string) $m[0];
- $o = (new UTF8($c))->nextCode();
- $esc = "U".str_pad(strtoupper(dechex($o)), 6, "0", \STR_PAD_LEFT);
- $name = $esc.substr($name, strlen($c));
+ protected static function coerceName(string $name, bool $prefixed = false): string {
+ if ($prefixed) {
+ $name = explode(":", $name, 2);
+ } else {
+ $name = [$name];
}
- return $name;
+ return implode(":", array_map(function($name) {
+ // This matches the inverse of the production of Name in XML 1.0 Fourth Edition,
+ // with the added exclusion of ":" from allowed characters
+ // See https://www.w3.org/TR/2006/REC-xml-20060816/#NT-NameChar
+ preg_match_all('/[^_\.\-\x{41}-\x{5A}\x{61}-\x{7A}\x{C0}-\x{D6}\x{D8}-\x{F6}\x{F8}-\x{FF}\x{100}-\x{131}\x{134}-\x{13E}\x{141}-\x{148}\x{14A}-\x{17E}\x{180}-\x{1C3}\x{1CD}-\x{1F0}\x{1F4}-\x{1F5}\x{1FA}-\x{217}\x{250}-\x{2A8}\x{2BB}-\x{2C1}\x{386}\x{388}-\x{38A}\x{38C}\x{38E}-\x{3A1}\x{3A3}-\x{3CE}\x{3D0}-\x{3D6}\x{3DA}\x{3DC}\x{3DE}\x{3E0}\x{3E2}-\x{3F3}\x{401}-\x{40C}\x{40E}-\x{44F}\x{451}-\x{45C}\x{45E}-\x{481}\x{490}-\x{4C4}\x{4C7}-\x{4C8}\x{4CB}-\x{4CC}\x{4D0}-\x{4EB}\x{4EE}-\x{4F5}\x{4F8}-\x{4F9}\x{531}-\x{556}\x{559}\x{561}-\x{586}\x{5D0}-\x{5EA}\x{5F0}-\x{5F2}\x{621}-\x{63A}\x{641}-\x{64A}\x{671}-\x{6B7}\x{6BA}-\x{6BE}\x{6C0}-\x{6CE}\x{6D0}-\x{6D3}\x{6D5}\x{6E5}-\x{6E6}\x{905}-\x{939}\x{93D}\x{958}-\x{961}\x{985}-\x{98C}\x{98F}-\x{990}\x{993}-\x{9A8}\x{9AA}-\x{9B0}\x{9B2}\x{9B6}-\x{9B9}\x{9DC}-\x{9DD}\x{9DF}-\x{9E1}\x{9F0}-\x{9F1}\x{A05}-\x{A0A}\x{A0F}-\x{A10}\x{A13}-\x{A28}\x{A2A}-\x{A30}\x{A32}-\x{A33}\x{A35}-\x{A36}\x{A38}-\x{A39}\x{A59}-\x{A5C}\x{A5E}\x{A72}-\x{A74}\x{A85}-\x{A8B}\x{A8D}\x{A8F}-\x{A91}\x{A93}-\x{AA8}\x{AAA}-\x{AB0}\x{AB2}-\x{AB3}\x{AB5}-\x{AB9}\x{ABD}\x{AE0}\x{B05}-\x{B0C}\x{B0F}-\x{B10}\x{B13}-\x{B28}\x{B2A}-\x{B30}\x{B32}-\x{B33}\x{B36}-\x{B39}\x{B3D}\x{B5C}-\x{B5D}\x{B5F}-\x{B61}\x{B85}-\x{B8A}\x{B8E}-\x{B90}\x{B92}-\x{B95}\x{B99}-\x{B9A}\x{B9C}\x{B9E}-\x{B9F}\x{BA3}-\x{BA4}\x{BA8}-\x{BAA}\x{BAE}-\x{BB5}\x{BB7}-\x{BB9}\x{C05}-\x{C0C}\x{C0E}-\x{C10}\x{C12}-\x{C28}\x{C2A}-\x{C33}\x{C35}-\x{C39}\x{C60}-\x{C61}\x{C85}-\x{C8C}\x{C8E}-\x{C90}\x{C92}-\x{CA8}\x{CAA}-\x{CB3}\x{CB5}-\x{CB9}\x{CDE}\x{CE0}-\x{CE1}\x{D05}-\x{D0C}\x{D0E}-\x{D10}\x{D12}-\x{D28}\x{D2A}-\x{D39}\x{D60}-\x{D61}\x{E01}-\x{E2E}\x{E30}\x{E32}-\x{E33}\x{E40}-\x{E45}\x{E81}-\x{E82}\x{E84}\x{E87}-\x{E88}\x{E8A}\x{E8D}\x{E94}-\x{E97}\x{E99}-\x{E9F}\x{EA1}-\x{EA3}\x{EA5}\x{EA7}\x{EAA}-\x{EAB}\x{EAD}-\x{EAE}\x{EB0}\x{EB2}-\x{EB3}\x{EBD}\x{EC0}-\x{EC4}\x{F40}-\x{F47}\x{F49}-\x{F69}\x{10A0}-\x{10C5}\x{10D0}-\x{10F6}\x{1100}\x{1102}-\x{1103}\x{1105}-\x{1107}\x{1109}\x{110B}-\x{110C}\x{110E}-\x{1112}\x{113C}\x{113E}\x{1140}\x{114C}\x{114E}\x{1150}\x{1154}-\x{1155}\x{1159}\x{115F}-\x{1161}\x{1163}\x{1165}\x{1167}\x{1169}\x{116D}-\x{116E}\x{1172}-\x{1173}\x{1175}\x{119E}\x{11A8}\x{11AB}\x{11AE}-\x{11AF}\x{11B7}-\x{11B8}\x{11BA}\x{11BC}-\x{11C2}\x{11EB}\x{11F0}\x{11F9}\x{1E00}-\x{1E9B}\x{1EA0}-\x{1EF9}\x{1F00}-\x{1F15}\x{1F18}-\x{1F1D}\x{1F20}-\x{1F45}\x{1F48}-\x{1F4D}\x{1F50}-\x{1F57}\x{1F59}\x{1F5B}\x{1F5D}\x{1F5F}-\x{1F7D}\x{1F80}-\x{1FB4}\x{1FB6}-\x{1FBC}\x{1FBE}\x{1FC2}-\x{1FC4}\x{1FC6}-\x{1FCC}\x{1FD0}-\x{1FD3}\x{1FD6}-\x{1FDB}\x{1FE0}-\x{1FEC}\x{1FF2}-\x{1FF4}\x{1FF6}-\x{1FFC}\x{2126}\x{212A}-\x{212B}\x{212E}\x{2180}-\x{2182}\x{3041}-\x{3094}\x{30A1}-\x{30FA}\x{3105}-\x{312C}\x{AC00}-\x{D7A3}\x{4E00}-\x{9FA5}\x{3007}\x{3021}-\x{3029}\x{30}-\x{39}\x{660}-\x{669}\x{6F0}-\x{6F9}\x{966}-\x{96F}\x{9E6}-\x{9EF}\x{A66}-\x{A6F}\x{AE6}-\x{AEF}\x{B66}-\x{B6F}\x{BE7}-\x{BEF}\x{C66}-\x{C6F}\x{CE6}-\x{CEF}\x{D66}-\x{D6F}\x{E50}-\x{E59}\x{ED0}-\x{ED9}\x{F20}-\x{F29}\x{300}-\x{345}\x{360}-\x{361}\x{483}-\x{486}\x{591}-\x{5A1}\x{5A3}-\x{5B9}\x{5BB}-\x{5BD}\x{5BF}\x{5C1}-\x{5C2}\x{5C4}\x{64B}-\x{652}\x{670}\x{6D6}-\x{6DC}\x{6DD}-\x{6DF}\x{6E0}-\x{6E4}\x{6E7}-\x{6E8}\x{6EA}-\x{6ED}\x{901}-\x{903}\x{93C}\x{93E}-\x{94C}\x{94D}\x{951}-\x{954}\x{962}-\x{963}\x{981}-\x{983}\x{9BC}\x{9BE}\x{9BF}\x{9C0}-\x{9C4}\x{9C7}-\x{9C8}\x{9CB}-\x{9CD}\x{9D7}\x{9E2}-\x{9E3}\x{A02}\x{A3C}\x{A3E}\x{A3F}\x{A40}-\x{A42}\x{A47}-\x{A48}\x{A4B}-\x{A4D}\x{A70}-\x{A71}\x{A81}-\x{A83}\x{ABC}\x{ABE}-\x{AC5}\x{AC7}-\x{AC9}\x{ACB}-\x{ACD}\x{B01}-\x{B03}\x{B3C}\x{B3E}-\x{B43}\x{B47}-\x{B48}\x{B4B}-\x{B4D}\x{B56}-\x{B57}\x{B82}-\x{B83}\x{BBE}-\x{BC2}\x{BC6}-\x{BC8}\x{BCA}-\x{BCD}\x{BD7}\x{C01}-\x{C03}\x{C3E}-\x{C44}\x{C46}-\x{C48}\x{C4A}-\x{C4D}\x{C55}-\x{C56}\x{C82}-\x{C83}\x{CBE}-\x{CC4}\x{CC6}-\x{CC8}\x{CCA}-\x{CCD}\x{CD5}-\x{CD6}\x{D02}-\x{D03}\x{D3E}-\x{D43}\x{D46}-\x{D48}\x{D4A}-\x{D4D}\x{D57}\x{E31}\x{E34}-\x{E3A}\x{E47}-\x{E4E}\x{EB1}\x{EB4}-\x{EB9}\x{EBB}-\x{EBC}\x{EC8}-\x{ECD}\x{F18}-\x{F19}\x{F35}\x{F37}\x{F39}\x{F3E}\x{F3F}\x{F71}-\x{F84}\x{F86}-\x{F8B}\x{F90}-\x{F95}\x{F97}\x{F99}-\x{FAD}\x{FB1}-\x{FB7}\x{FB9}\x{20D0}-\x{20DC}\x{20E1}\x{302A}-\x{302F}\x{3099}\x{309A}\x{B7}\x{2D0}\x{2D1}\x{387}\x{640}\x{E46}\x{EC6}\x{3005}\x{3031}-\x{3035}\x{309D}-\x{309E}\x{30FC}-\x{30FE}]/u', $name, $m);
+ foreach (array_unique($m[0], \SORT_STRING) as $c) {
+ $o = (new UTF8($c))->nextCode();
+ $esc = "U".str_pad(strtoupper(dechex($o)), 6, "0", \STR_PAD_LEFT);
+ $name = str_replace($c, $esc, $name);
+ }
+ // Apply stricter rules to the first character
+ if (preg_match('/^[^_\x{41}-\x{5A}\x{61}-\x{7A}\x{C0}-\x{D6}\x{D8}-\x{F6}\x{F8}-\x{FF}\x{100}-\x{131}\x{134}-\x{13E}\x{141}-\x{148}\x{14A}-\x{17E}\x{180}-\x{1C3}\x{1CD}-\x{1F0}\x{1F4}-\x{1F5}\x{1FA}-\x{217}\x{250}-\x{2A8}\x{2BB}-\x{2C1}\x{386}\x{388}-\x{38A}\x{38C}\x{38E}-\x{3A1}\x{3A3}-\x{3CE}\x{3D0}-\x{3D6}\x{3DA}\x{3DC}\x{3DE}\x{3E0}\x{3E2}-\x{3F3}\x{401}-\x{40C}\x{40E}-\x{44F}\x{451}-\x{45C}\x{45E}-\x{481}\x{490}-\x{4C4}\x{4C7}-\x{4C8}\x{4CB}-\x{4CC}\x{4D0}-\x{4EB}\x{4EE}-\x{4F5}\x{4F8}-\x{4F9}\x{531}-\x{556}\x{559}\x{561}-\x{586}\x{5D0}-\x{5EA}\x{5F0}-\x{5F2}\x{621}-\x{63A}\x{641}-\x{64A}\x{671}-\x{6B7}\x{6BA}-\x{6BE}\x{6C0}-\x{6CE}\x{6D0}-\x{6D3}\x{6D5}\x{6E5}-\x{6E6}\x{905}-\x{939}\x{93D}\x{958}-\x{961}\x{985}-\x{98C}\x{98F}-\x{990}\x{993}-\x{9A8}\x{9AA}-\x{9B0}\x{9B2}\x{9B6}-\x{9B9}\x{9DC}-\x{9DD}\x{9DF}-\x{9E1}\x{9F0}-\x{9F1}\x{A05}-\x{A0A}\x{A0F}-\x{A10}\x{A13}-\x{A28}\x{A2A}-\x{A30}\x{A32}-\x{A33}\x{A35}-\x{A36}\x{A38}-\x{A39}\x{A59}-\x{A5C}\x{A5E}\x{A72}-\x{A74}\x{A85}-\x{A8B}\x{A8D}\x{A8F}-\x{A91}\x{A93}-\x{AA8}\x{AAA}-\x{AB0}\x{AB2}-\x{AB3}\x{AB5}-\x{AB9}\x{ABD}\x{AE0}\x{B05}-\x{B0C}\x{B0F}-\x{B10}\x{B13}-\x{B28}\x{B2A}-\x{B30}\x{B32}-\x{B33}\x{B36}-\x{B39}\x{B3D}\x{B5C}-\x{B5D}\x{B5F}-\x{B61}\x{B85}-\x{B8A}\x{B8E}-\x{B90}\x{B92}-\x{B95}\x{B99}-\x{B9A}\x{B9C}\x{B9E}-\x{B9F}\x{BA3}-\x{BA4}\x{BA8}-\x{BAA}\x{BAE}-\x{BB5}\x{BB7}-\x{BB9}\x{C05}-\x{C0C}\x{C0E}-\x{C10}\x{C12}-\x{C28}\x{C2A}-\x{C33}\x{C35}-\x{C39}\x{C60}-\x{C61}\x{C85}-\x{C8C}\x{C8E}-\x{C90}\x{C92}-\x{CA8}\x{CAA}-\x{CB3}\x{CB5}-\x{CB9}\x{CDE}\x{CE0}-\x{CE1}\x{D05}-\x{D0C}\x{D0E}-\x{D10}\x{D12}-\x{D28}\x{D2A}-\x{D39}\x{D60}-\x{D61}\x{E01}-\x{E2E}\x{E30}\x{E32}-\x{E33}\x{E40}-\x{E45}\x{E81}-\x{E82}\x{E84}\x{E87}-\x{E88}\x{E8A}\x{E8D}\x{E94}-\x{E97}\x{E99}-\x{E9F}\x{EA1}-\x{EA3}\x{EA5}\x{EA7}\x{EAA}-\x{EAB}\x{EAD}-\x{EAE}\x{EB0}\x{EB2}-\x{EB3}\x{EBD}\x{EC0}-\x{EC4}\x{F40}-\x{F47}\x{F49}-\x{F69}\x{10A0}-\x{10C5}\x{10D0}-\x{10F6}\x{1100}\x{1102}-\x{1103}\x{1105}-\x{1107}\x{1109}\x{110B}-\x{110C}\x{110E}-\x{1112}\x{113C}\x{113E}\x{1140}\x{114C}\x{114E}\x{1150}\x{1154}-\x{1155}\x{1159}\x{115F}-\x{1161}\x{1163}\x{1165}\x{1167}\x{1169}\x{116D}-\x{116E}\x{1172}-\x{1173}\x{1175}\x{119E}\x{11A8}\x{11AB}\x{11AE}-\x{11AF}\x{11B7}-\x{11B8}\x{11BA}\x{11BC}-\x{11C2}\x{11EB}\x{11F0}\x{11F9}\x{1E00}-\x{1E9B}\x{1EA0}-\x{1EF9}\x{1F00}-\x{1F15}\x{1F18}-\x{1F1D}\x{1F20}-\x{1F45}\x{1F48}-\x{1F4D}\x{1F50}-\x{1F57}\x{1F59}\x{1F5B}\x{1F5D}\x{1F5F}-\x{1F7D}\x{1F80}-\x{1FB4}\x{1FB6}-\x{1FBC}\x{1FBE}\x{1FC2}-\x{1FC4}\x{1FC6}-\x{1FCC}\x{1FD0}-\x{1FD3}\x{1FD6}-\x{1FDB}\x{1FE0}-\x{1FEC}\x{1FF2}-\x{1FF4}\x{1FF6}-\x{1FFC}\x{2126}\x{212A}-\x{212B}\x{212E}\x{2180}-\x{2182}\x{3041}-\x{3094}\x{30A1}-\x{30FA}\x{3105}-\x{312C}\x{AC00}-\x{D7A3}\x{4E00}-\x{9FA5}\x{3007}\x{3021}-\x{3029}]/u', $name, $m)) {
+ $c = (string) $m[0];
+ $o = (new UTF8($c))->nextCode();
+ $esc = "U".str_pad(strtoupper(dechex($o)), 6, "0", \STR_PAD_LEFT);
+ $name = $esc.substr($name, strlen($c));
+ }
+ return $name;
+ }, $name));
}
- protected function uncoerceName(string $name): string {
+ protected static function uncoerceName(string $name): string {
preg_match_all('/U[0-9A-F]{6}/', $name, $m);
foreach (array_unique($m[0], \SORT_STRING) as $o) {
$c = UTF8::encode(hexdec(substr($o, 1)));
@@ -59,7 +66,7 @@ trait NameCoercion {
return $name;
}
- protected function escapeString(string $string, bool $attribute = false): string {
+ protected static function escapeString(string $string, bool $attribute = false): string {
# Escaping a string (for the purposes of the algorithm above) consists of
# running the following steps:
diff --git a/lib/Parser/ParseErrorEmitter.php b/lib/Parser/ParseErrorEmitter.php
index 68e5beb..316d344 100644
--- a/lib/Parser/ParseErrorEmitter.php
+++ b/lib/Parser/ParseErrorEmitter.php
@@ -78,7 +78,7 @@ trait ParseErrorEmitter {
// Count the number of replacements needed in the message.
$count = substr_count($message, '%s');
// If the number of replacements don't match the arguments then oops.
- assert(count($arg) === $count, new \Exception("Parse error message expects $count parameters"));
+ assert(count($arg) === $count, new \Exception("Message of parse error $code expects $count parameters"));
if ($count > 0) {
// Convert newlines and tabs in the arguments to words to better
@@ -89,7 +89,7 @@ trait ParseErrorEmitter {
} elseif ($value === "\t") {
return 'Tab';
} elseif ($value === null) {
- return 'nothing';
+ return 'nothing'; // @codeCoverageIgnore
} else {
return $value;
}
diff --git a/lib/Parser/Serializer.php b/lib/Parser/Serializer.php
new file mode 100644
index 0000000..a3c430e
--- /dev/null
+++ b/lib/Parser/Serializer.php
@@ -0,0 +1,232 @@
+namespaceURI ?? Parser::HTML_NAMESPACE, [Parser::HTML_NAMESPACE, Parser::SVG_NAMESPACE, Parser::MATHML_NAMESPACE])) {
+ $tagName = self::uncoerceName($n->localName);
+ } else {
+ $tagName = self::uncoerceName($n->tagName);
+ }
+ # Append a U+003C LESS-THAN SIGN character (<), followed by tagname.
+ $s .= "<$tagName";
+ # If current node's is value is not null, and the element does
+ # not have an is attribute in its attribute list, then
+ # append the string " is="", followed by current node's is
+ # value escaped as described below in attribute mode,
+ # followed by a U+0022 QUOTATION MARK character (").
+ // DEVIATION: We don't support custom elements
+ # For each attribute that the element has, append a
+ # U+0020 SPACE character, the attribute's serialized name as
+ # described below, a U+003D EQUALS SIGN character (=), a
+ # U+0022 QUOTATION MARK character ("), the attribute's
+ # value, escaped as described below in attribute mode, and
+ # a second U+0022 QUOTATION MARK character (").
+ foreach ($n->attributes as $a) {
+ # An attribute's serialized name for the purposes of the previous
+ # paragraph must be determined as follows:
+
+ # If the attribute has no namespace
+ if ($a->namespaceURI === null) {
+ # The attribute's serialized name is the attribute's local name.
+ $name = self::uncoerceName($a->localName);
+ }
+ # If the attribute is in the XML namespace
+ elseif ($a->namespaceURI === Parser::XML_NAMESPACE) {
+ # The attribute's serialized name is the string "xml:" followed
+ # by the attribute's local name.
+ $name = "xml:".self::uncoerceName($a->localName);
+ }
+ # If the attribute is in the XMLNS namespace...
+ elseif ($a->namespaceURI === Parser::XMLNS_NAMESPACE) {
+ # ... and the attribute's local name is xmlns
+ if ($a->localName === "xmlns") {
+ # The attribute's serialized name is the string "xmlns".
+ $name = "xmlns";
+ }
+ # ... and the attribute's local name is not xmlns
+ else {
+ # The attribute's serialized name is the string "xmlns:"
+ # followed by the attribute's local name.
+ $name = "xmlns:".self::uncoerceName($a->localName);
+ }
+ }
+ # If the attribute is in the XLink namespace
+ elseif ($a->namespaceURI === Parser::XLINK_NAMESPACE) {
+ # The attribute's serialized name is the string "xlink:"
+ # followed by the attribute's local name.
+ $name = "xlink:".self::uncoerceName($a->localName);
+ }
+ # If the attribute is in some other namespace
+ else {
+ # The attribute's serialized name is the attribute's qualified name.
+ $name = ($a->prefix !== "") ? $a->prefix.":".$a->name : $a->name;
+ }
+ $value = self::escapeString((string) $a->value, true);
+ $s .= " $name=\"$value\"";
+ }
+ # Append a U+003E GREATER-THAN SIGN character (>).
+ $s .= ">";
+ # If current node serializes as void, then continue on to the
+ # next child node at this point.
+ # Append the value of running the HTML fragment serialization
+ # algorithm on the current node element (thus recursing into
+ # this algorithm for that element), followed by a
+ # U+003C LESS-THAN SIGN character (<), a U+002F SOLIDUS
+ # character (/), tagname again, and finally a
+ # U+003E GREATER-THAN SIGN character (>).
+ if (($n->namespaceURI ?? Parser::HTML_NAMESPACE) !== Parser::HTML_NAMESPACE || !in_array($tagName, self::VOID_ELEMENTS)) {
+ # If the node is a template element, then let the node instead
+ # be the template element's template contents
+ # (a DocumentFragment node).
+ if (
+ ($n->namespaceURI ?? Parser::HTML_NAMESPACE) === Parser::HTML_NAMESPACE
+ && $n->tagName === "template"
+ && property_exists($n, "content")
+ && $n->content instanceof \DOMDocumentFragment
+ ) {
+ // NOTE: Treat template content as any other document
+ // fragment and just invoke the inner serializer
+ $s .= self::serializeInner($n->content)."";
+ } elseif ($n->hasChildNodes()) {
+ // If the element has children, store its tag name and
+ // continue the loop with its first child; its end
+ // tag will be written out further down
+ $stack[] = $tagName;
+ $n = $n->firstChild;
+ continue;
+ } else {
+ // Otherwise just append the end tag now
+ $s .= "";
+ }
+ }
+ }
+ # If current node is a Text node
+ elseif ($n instanceof \DOMText) {
+ # If the parent of current node is a style, script, xmp,
+ # iframe, noembed, noframes, or plaintext element, or
+ # if the parent of current node is a noscript element
+ # and scripting is enabled for the node, then append
+ # the value of current node's data IDL attribute literally.
+ $p = $n->parentNode;
+ if ($p instanceof \DOMElement && ($p->namespaceURI ?? Parser::HTML_NAMESPACE) === Parser::HTML_NAMESPACE && in_array($p->tagName, self::RAWTEXT_ELEMENTS)) {
+ // NOTE: scripting is assumed not to be enabled
+ $s .= $n->data;
+ }
+ # Otherwise, append the value of current node's data IDL attribute, escaped as described below.
+ else {
+ $s .= self::escapeString($n->data);
+ }
+ }
+ # If current node is a Comment
+ elseif ($n instanceof \DOMComment) {
+ # Append the literal string "" (U+002D HYPHEN-MINUS, U+002D HYPHEN-MINUS,
+ # U+003E GREATER-THAN SIGN).
+ $s .= "";
+ }
+ # If current node is a ProcessingInstruction
+ elseif ($n instanceof \DOMProcessingInstruction) {
+ # Append the literal string ").
+ $s .= "target)." ".$n->data.">";
+ }
+ # If current node is a DocumentType
+ elseif ($n instanceof \DOMDocumentType) {
+ # Append the literal string "" (U+003E GREATER-THAN SIGN).
+ $s .= "name).">";
+ }
+ // NOTE: Documents and document fragments have no outer content,
+ // so we can just serialize the inner content
+ elseif ($n instanceof \DOMDocument || $n instanceof \DOMDocumentFragment) {
+ return self::serializeInner($n);
+ } else {
+ throw new Exception(Exception::UNSUPPORTED_NODE_TYPE, [get_class($n)]);
+ }
+ // If the current node has no more siblings, go up the tree till a
+ // sibling is found or we've reached the original node
+ while (!$n->nextSibling && $stack) {
+ // Write out the stored end tag each time we go up the tree
+ $tagName = array_pop($stack);
+ $s .= "";
+ $n = $n->parentNode;
+ }
+ $n = $n->nextSibling;
+ } while ($stack); // Loop until we have traversed the subtree of the target node in full
+ return $s;
+ }
+
+ /** Serializes the children of an HTML DOM node to a string. This is equivalent to the innerHTML getter
+ *
+ * @param \DOMDocument|\DOMElement|\DOMDocumentFragment $node The node to serialize
+ */
+ public static function serializeInner(\DOMNode $node): string {
+ # Let s be a string, and initialize it to the empty string.
+ $s = "";
+
+ if ($node instanceof \DOMElement && ($node->namespaceURI ?? Parser::HTML_NAMESPACE) === Parser::HTML_NAMESPACE) {
+ # If the node serializes as void, then return the empty string.
+ if (in_array($node->tagName, self::VOID_ELEMENTS)) {
+ return "";
+ }
+ # If the node is a template element, then let the node instead
+ # be the template element's template contents
+ # (a DocumentFragment node).
+ elseif ($node->tagName === "template" && property_exists($node, "content") && $node->content instanceof \DOMDocumentFragment) {
+ // NOTE: template elements won't necessarily have a content
+ // property because PHP's DOM does not support this natively
+ $node = $node->content;
+ }
+ }
+ if ($node instanceof \DOMElement || $node instanceof \DOMDocument || $node instanceof \DOMDocumentFragment) {
+ # For each child node of the node, in tree order, run the following steps:
+ // NOTE: the steps in question are implemented in the "serializeOuter" routine
+ foreach ($node->childNodes as $n) {
+ $s .= self::serializeOuter($n);
+ }
+ } else {
+ throw new Exception(Exception::UNSUPPORTED_NODE_TYPE, [get_class($node)]);
+ }
+ return $s;
+ }
+}
diff --git a/lib/Parser/TreeConstructor.php b/lib/Parser/TreeConstructor.php
index f0b2a22..b8fa91f 100644
--- a/lib/Parser/TreeConstructor.php
+++ b/lib/Parser/TreeConstructor.php
@@ -9,7 +9,7 @@ namespace MensBeam\HTML\Parser;
use MensBeam\HTML\Parser;
class TreeConstructor {
- use ParseErrorEmitter, NameCoercion;
+ use ParseErrorEmitter, NameCoercion, AttributeSetter;
public $debugLog = "";
@@ -335,12 +335,6 @@ class TreeConstructor {
$iterations = 0;
$insertionMode = $this->insertionMode;
- // If element name coercison has occurred at some earlier point,
- // we must coerce all end tag names to match mangled start tags
- if ($token instanceof EndTagToken && $this->mangledElements) {
- $token->name = $this->coerceName($token->name);
- }
-
# 13.2.6 Tree construction
#
# As each token is emitted from the tokenizer, the user agent must follow the
@@ -379,6 +373,12 @@ class TreeConstructor {
return true;
})());
+ // If element name coercison has occurred at some earlier point,
+ // we must coerce all end tag names to match mangled start tags
+ if ($this->mangledElements && $token instanceof EndTagToken) {
+ $token->name = self::coerceName($token->name);
+ }
+
# 13.2.6.4. The rules for parsing tokens in HTML content
// OPTIMIZATION: Evaluation the "in body" mode first is
// faster for typical documents
@@ -401,7 +401,7 @@ class TreeConstructor {
// If attribute name coercison has occurred at some earlier point,
// we must coerce all attributes on html and body start tags in
// case they are relocated to existing elements
- $attrName = $this->mangledAttributes ? $this->coerceName($a->name) : $a->name;
+ $attrName = $this->mangledAttributes ? self::coerceName($a->name) : $a->name;
if (!$top->hasAttributeNS(null, $attrName)) {
$this->elementSetAttribute($top, null, $attrName, $a->value);
}
@@ -433,7 +433,7 @@ class TreeConstructor {
// If attribute name coercison has occurred at some earlier point,
// we must coerce all attributes on html and body start tags in
// case they are relocated to existing elements
- $attrName = $this->mangledAttributes ? $this->coerceName($a->name) : $a->name;
+ $attrName = $this->mangledAttributes ? self::coerceName($a->name) : $a->name;
if (!$body->hasAttributeNS(null, $attrName)) {
$this->elementSetAttribute($body, null, $attrName, $a->value);
}
@@ -521,7 +521,8 @@ class TreeConstructor {
if (strlen($nextToken->data) === 1 && $nextToken->data === "\n") {
continue;
} elseif (strpos($nextToken->data, "\n") === 0) {
- $nextToken->data = substr($nextToken->data, 1);
+ // NOTE: This case is not currently encountered by the parser due to special handling of newlines
+ $nextToken->data = substr($nextToken->data, 1); // @codeCoverageIgnore
}
}
// Process the next token
@@ -818,7 +819,8 @@ class TreeConstructor {
if (strlen($nextToken->data) === 1 && $nextToken->data === "\n") {
continue;
} elseif (strpos($nextToken->data, "\n") === 0) {
- $nextToken->data = substr($nextToken->data, 1);
+ // NOTE: This case is not currently encountered by the parser due to special handling of newlines
+ $nextToken->data = substr($nextToken->data, 1); // @codeCoverageIgnore
}
}
# Let the original insertion mode be the current insertion mode.
@@ -1065,7 +1067,7 @@ class TreeConstructor {
else {
# 1. If the stack of open elements does not have a form element in scope, then
# this is a parse error; return and ignore the token.
- if ($this->stack->hasElementInScope('form')) {
+ if (!$this->stack->hasElementInScope('form')) {
$this->error(ParseError::UNEXPECTED_END_TAG, $token->name);
continue;
}
@@ -1918,10 +1920,6 @@ class TreeConstructor {
elseif ($token->name === 'noframes' || $token->name === 'style') {
$this->parseGenericRawText($token);
}
- elseif ($token->name === 'noscript') {
- $this->insertStartTagToken($token);
- $this->insertionMode = self::IN_HEAD_NOSCRIPT_MODE;
- }
elseif ($token->name === 'script') {
$this->insertStartTagToken($token);
$this->tokenizer->state = Tokenizer::SCRIPT_DATA_STATE;
@@ -2714,7 +2712,9 @@ class TreeConstructor {
# element in table scope, then this is a parse error;
# ignore the token. (fragment case)
if (!$this->stack->hasElementInTableScope("td", "th")) {
- $this->error(ParseError::UNEXPECTED_START_TAG, $token->name);
+ // NOTE: This case appears to be unreachable
+ // See https://github.com/whatwg/html/issues/7242
+ $this->error(ParseError::UNEXPECTED_START_TAG, $token->name); //@codeCoverageIgnore
}
# Otherwise, close the cell (see below) and reprocess the token.
else {
@@ -2861,7 +2861,7 @@ class TreeConstructor {
# An end tag...
elseif ($token instanceof EndTagToken) {
# An end tag whose tag name is "template"
- if ($token->name === "tenplate") {
+ if ($token->name === "template") {
# Process the token using the rules for the "in head" insertion mode.
$insertionMode = self::IN_HEAD_MODE;
goto ProcessToken;
@@ -3262,10 +3262,8 @@ class TreeConstructor {
# Anything else
else {
# Parse error. Ignore the token.
- assert($token instanceof CharacterToken || $token instanceof TagToken, new \Exception("Invalid token class: ".get_class($token)));
- if ($token instanceof StartTagToken) {
- $this->error(ParseError::UNEXPECTED_START_TAG, $token->name);
- } elseif ($token instanceof EndTagToken) {
+ assert($token instanceof CharacterToken || $token instanceof EndTagToken, new \Exception("Invalid token class: ".get_class($token)));
+ if ($token instanceof EndTagToken) {
$this->error(ParseError::UNEXPECTED_END_TAG, $token->name);
} elseif ($token instanceof CharacterToken) {
$this->error(ParseError::UNEXPECTED_CHAR, $token->data, "exclude whitespace");
@@ -3434,6 +3432,12 @@ class TreeConstructor {
return true;
})());
+ // If element name coercison has occurred at some earlier point,
+ // we must coerce all end tag names to match mangled start tags
+ if ($this->mangledElements && $token instanceof EndTagToken) {
+ $token->name = self::coerceName($token->name, true);
+ }
+
# 13.2.6.5 The rules for parsing tokens in foreign content
#
# When the user agent is to apply the rules for parsing tokens in foreign
@@ -3587,7 +3591,8 @@ class TreeConstructor {
$node = $this->stack[$pos];
# If node's tag name, converted to ASCII lowercase, is not the
# same as the tag name of the token, then this is a parse error.
- if (strtolower($node->nodeName) !== $token->name) {
+ $nodeName = self::coerceName(strtolower(self::uncoerceName($node->nodeName)), true);
+ if ($nodeName !== $token->name) {
$this->error(ParseError::UNEXPECTED_END_TAG, $token->name);
}
do {
@@ -3598,7 +3603,8 @@ class TreeConstructor {
# If node's tag name, converted to ASCII lowercase, is the same as the
# tag name of the token, pop elements from the stack of open elements until node
# has been popped from the stack, and then abort these steps.
- if (strtolower($node->nodeName) === $token->name) {
+ $nodeName = self::coerceName(strtolower(self::uncoerceName($node->nodeName)), true);
+ if ($nodeName === $token->name) {
$this->stack->popUntilSame($node);
continue 2;
}
@@ -3673,14 +3679,8 @@ class TreeConstructor {
// NOTE: The "entry above" refers to the "in body" insertion mode
// Changes here should be mirrored there
foreach ($this->stack as $node) {
- if ($node->nodeName === $token->name && $node->namespaceURI === $this->htmlNamespace) {
- $this->stack->generateImpliedEndTags($token->name);
- if (!$node->isSameNode($this->stack->currentNode)) {
- $this->error($errorCode, $token->name);
- }
- $this->stack->popUntilSame($node);
- return;
- } elseif ($this->isElementSpecial($node)) {
+ // NOTE: Only the "is special" case is possible here
+ if ($this->isElementSpecial($node)) {
$this->error($errorCode, $token->name);
return;
}
@@ -3890,12 +3890,15 @@ class TreeConstructor {
// Abort!
}
else {
+ // NOTE: This is an edge case only possible via scripting
+ // @codeCoverageIgnoreStart
# 6. Let previous element be the element immediately above last table in the
# stack of open elements.
$previousElement = $this->stack[$lastTableIndex - 1];
# 7. Let adjusted insertion location be inside previous element, after its last
# child (if any).
$insertionLocation = $previousElement;
+ // @codeCoverageIgnoreEnd
}
}
# Otherwise let adjusted insertion location be inside target, after its last
@@ -4150,7 +4153,7 @@ class TreeConstructor {
# 17. Let node now be the node before node in the stack of open elements.
# 18. Return to the step labeled Loop.
}
- }
+ } // @codeCoverageIgnore
protected function closePElement(TagToken $token) {
# When the steps above say the UA is to close a p element, it means that the UA
@@ -4216,11 +4219,7 @@ class TreeConstructor {
// The element name is invalid for XML
// Replace any offending characters with "UHHHHHH" where H are the
// uppercase hexadecimal digits of the character's code point
- if ($namespace !== $this->htmlNamespace) {
- $qualifiedName = implode(":", array_map([$this, "coerceName"], explode(":", $token->name, 2)));
- } else {
- $qualifiedName = $this->coerceName($token->name);
- }
+ $qualifiedName = self::coerceName($token->name, ($namespace !== $this->htmlNamespace));
$element = $this->DOM->createElementNS($namespace, $qualifiedName);
$this->mangledElements = true;
}
@@ -4246,47 +4245,6 @@ class TreeConstructor {
return $element;
}
- public function elementSetAttribute(\DOMElement $element, ?string $namespaceURI, string $qualifiedName, string $value): void {
- if ($namespaceURI === Parser::XMLNS_NAMESPACE) {
- // NOTE: We create attribute nodes so that xmlns attributes
- // don't get lost; otherwise they cannot be serialized
- try {
- $a = @$element->ownerDocument->createAttributeNS($namespaceURI, $qualifiedName);
- } catch (\DOMException $e) {
- // FIXME: PHP has a fit here if the document element has a namespace and no prefix
- // A workaround does not seem to exist
- return;
- }
- if ($a === false) {
- // The document element does not exist yet, so we need
- // to insert this element into the document
- $element->ownerDocument->appendChild($element);
- $a = $element->ownerDocument->createAttributeNS($namespaceURI, $qualifiedName);
- $element->ownerDocument->removeChild($element);
- }
- $a->value = $this->escapeString($value, true);
- $element->setAttributeNodeNS($a);
- } else {
- try {
- $element->setAttributeNS($namespaceURI, $qualifiedName, $value);
- } catch (\DOMException $e) {
- // The attribute name is invalid for XML
- // Replace any offending characters with "UHHHHHH" where H are the
- // uppercase hexadecimal digits of the character's code point
- if ($namespaceURI !== null) {
- $qualifiedName = implode(":", array_map([$this, "coerceName"], explode(":", $qualifiedName, 2)));
- } else {
- $qualifiedName = $this->coerceName($qualifiedName);
- }
- $element->setAttributeNS($namespaceURI, $qualifiedName, $value);
- $this->mangledAttributes = true;
- }
- if ($qualifiedName === "id" && $namespaceURI === null) {
- $element->setIdAttribute($qualifiedName, true);
- }
- }
- }
-
public function isMathMLTextIntegrationPoint(\DOMElement $e): bool {
return ($e->namespaceURI === Parser::MATHML_NAMESPACE && (in_array($e->nodeName, ['mi', 'mo', 'mn', 'ms', 'mtext'])));
}
diff --git a/lib/Parser/ctype.php b/lib/Parser/ctype.php
index cd9d177..bae6e7b 100644
--- a/lib/Parser/ctype.php
+++ b/lib/Parser/ctype.php
@@ -11,6 +11,7 @@ namespace MensBeam\HTML\Parser;
// replacement, as they are designed only to evaluate
// single characters
+// @codeCoverageIgnoreStart
if (!extension_loaded("ctype")) {
function ctype_alnum(string $str): bool {
return ["a"=>true,"b"=>true,"c"=>true,"d"=>true,"e"=>true,"f"=>true,"g"=>true,"h"=>true,"i"=>true,"j"=>true,"k"=>true,"l"=>true,"m"=>true,"n"=>true,"o"=>true,"p"=>true,"q"=>true,"r"=>true,"s"=>true,"t"=>true,"u"=>true,"v"=>true,"w"=>true,"x"=>true,"y"=>true,"z"=>true,"A"=>true,"B"=>true,"C"=>true,"D"=>true,"E"=>true,"F"=>true,"G"=>true,"H"=>true,"I"=>true,"J"=>true,"K"=>true,"L"=>true,"M"=>true,"N"=>true,"O"=>true,"P"=>true,"Q"=>true,"R"=>true,"S"=>true,"T"=>true,"U"=>true,"V"=>true,"W"=>true,"X"=>true,"Y"=>true,"Z"=>true,"0"=>true,"1"=>true,"2"=>true,"3"=>true,"4"=>true,"5"=>true,"6"=>true,"7"=>true,"8"=>true,"9"=>true][$str] ?? false;
@@ -32,3 +33,4 @@ if (!extension_loaded("ctype")) {
return ["a"=>true,"b"=>true,"c"=>true,"d"=>true,"e"=>true,"f"=>true,"A"=>true,"B"=>true,"C"=>true,"D"=>true,"E"=>true,"F"=>true,"0"=>true,"1"=>true,"2"=>true,"3"=>true,"4"=>true,"5"=>true,"6"=>true,"7"=>true,"8"=>true,"9"=>true][$str] ?? false;
}
}
+// @codeCoverageIgnoreEnd
diff --git a/tests/cases/TestCharset.php b/tests/cases/TestCharset.php
index 144fcfb..639062e 100644
--- a/tests/cases/TestCharset.php
+++ b/tests/cases/TestCharset.php
@@ -117,6 +117,28 @@ class TestCharset extends \PHPUnit\Framework\TestCase {
$this->assertSame($exp, $act->encoding);
}
+ /**
+ * @dataProvider provideNonstandardDeclarationTests
+ * @covers \MensBeam\HTML\Parser\Data::__construct */
+ public function testNonstandardDeclarationTests(string $data, ?string $charset, ?string $fallback, int $bytesToScan, string $exp): void {
+ $config = new Config;
+ $config->encodingPrescanBytes = $bytesToScan;
+ $config->encodingFallback = $fallback;
+ $act = Parser::parse($data, $charset, $config);
+ $this->assertSame($exp, $act->encoding);
+ }
+
+ public function provideNonstandardDeclarationTests(): iterable {
+ return [
+ ["", null, null, 1024, "windows-1252"],
+ ["", null, null, 1024, "UTF-8"],
+ ["", null, "UTF-8", 1024, "UTF-8"],
+ ["", null, "UTF-7", 1024, "windows-1252"],
+ ];
+ }
+
public function provideStandardDeclarationTests() {
$tests = [];
$blacklist = ["xmldecl-3.html"];
diff --git a/tests/cases/TestParser.php b/tests/cases/TestParser.php
index e49d99c..9ccf580 100644
--- a/tests/cases/TestParser.php
+++ b/tests/cases/TestParser.php
@@ -13,6 +13,7 @@ use MensBeam\HTML\Parser\Exception;
/**
* @covers \MensBeam\HTML\Parser
+ * @covers \MensBeam\HTML\Parser\Exception
*/
class TestParser extends \PHPUnit\Framework\TestCase {
public function testParseADocument(): void {
@@ -33,6 +34,15 @@ class TestParser extends \PHPUnit\Framework\TestCase {
$this->assertInstanceOf(\DOMDocumentFragment::class, $out);
}
+ /** @covers \MensBeam\HTML\Parser\TreeConstructor::__construct */
+ public function testParseAFragmentWithBogusQuirksMode(): void {
+ $doc = new \DOMDocument();
+ $context = $doc->createElement("div");
+ $in = "hello world!";
+ $this->expectExceptionObject(new Exception(Exception::INVALID_QUIRKS_MODE));
+ Parser::parseFragment($context, -1, $in, "tex/html; charset=utf8");
+ }
+
public function testParseADocumentReportingErrors(): void {
$in = "hello world!";
$conf = new Config;
diff --git a/tests/cases/TestSerializer.php b/tests/cases/TestSerializer.php
new file mode 100644
index 0000000..522867d
--- /dev/null
+++ b/tests/cases/TestSerializer.php
@@ -0,0 +1,258 @@
+buildTree($data, $fragment);
+ $this->assertSame($exp, Serializer::serializeOuter($node));
+ }
+
+ public function provideStandardTreeTests(): iterable {
+ $blacklist = [];
+ $files = new \AppendIterator();
+ $files->append(new \GlobIterator(\MensBeam\HTML\Parser\BASE."tests/cases/serializer/*.dat", \FilesystemIterator::SKIP_DOTS | \FilesystemIterator::CURRENT_AS_PATHNAME));
+ foreach ($files as $file) {
+ if (!in_array(basename($file), $blacklist)) {
+ yield from $this->parseTreeTestFile($file);
+ }
+ }
+ }
+
+ /** @dataProvider provideTemplateTests */
+ public function testSerializeADecoratedTemplate(?string $ns, bool $content, bool $fragment, bool $text, string $exp): void {
+ $d = new \DOMDocument;
+ $t = $d->createElementNS($ns, "template");
+ $t->appendChild($d->createTextNode("EEK"));
+ if ($content) {
+ $t->content = null;
+ if ($fragment) {
+ $f = $d->createDocumentFragment();
+ $t->content = $f;
+ if ($text) {
+ $f->appendChild($d->createTextNode("OOK"));
+ }
+ }
+ }
+ $exp1 = $exp;
+ $exp2 = "$exp";
+ $this->assertSame($exp1, Serializer::serializeInner($t));
+ $this->assertSame($exp2, Serializer::serializeOuter($t));
+ }
+
+ public function provideTemplateTests(): iterable {
+ return [
+ [null, false, false, false, "EEK"],
+ [null, true, false, false, "EEK"],
+ [null, true, true, false, ""],
+ [null, true, true, true, "OOK"],
+ [Parser::HTML_NAMESPACE, false, false, false, "EEK"],
+ [Parser::HTML_NAMESPACE, true, false, false, "EEK"],
+ [Parser::HTML_NAMESPACE, true, true, false, ""],
+ [Parser::HTML_NAMESPACE, true, true, true, "OOK"],
+ ];
+ }
+
+ /** @dataProvider provideEmptyElementTests */
+ public function testInnerSerializeEmptyElement(string $tagName, ?string $ns, string $exp): void {
+ $d = new \DOMDocument;
+ $e = $d->createElementNS($ns, $tagName);
+ $e->appendChild($d->createTextNode("EEK"));
+ $this->assertSame($exp, Serializer::serializeInner($e));
+ }
+
+ public function provideEmptyElementTests(): iterable {
+ return [
+ ["basefont", null, ""],
+ ["bgsound", null, ""],
+ ["frame", null, ""],
+ ["keygen", null, ""],
+ ["area", null, ""],
+ ["base", null, ""],
+ ["br", null, ""],
+ ["col", null, ""],
+ ["embed", null, ""],
+ ["hr", null, ""],
+ ["img", null, ""],
+ ["input", null, ""],
+ ["link", null, ""],
+ ["meta", null, ""],
+ ["param", null, ""],
+ ["source", null, ""],
+ ["track", null, ""],
+ ["wbr", null, ""],
+ ["basefont", Parser::HTML_NAMESPACE, ""],
+ ["bgsound", Parser::HTML_NAMESPACE, ""],
+ ["frame", Parser::HTML_NAMESPACE, ""],
+ ["keygen", Parser::HTML_NAMESPACE, ""],
+ ["area", Parser::HTML_NAMESPACE, ""],
+ ["base", Parser::HTML_NAMESPACE, ""],
+ ["br", Parser::HTML_NAMESPACE, ""],
+ ["col", Parser::HTML_NAMESPACE, ""],
+ ["embed", Parser::HTML_NAMESPACE, ""],
+ ["hr", Parser::HTML_NAMESPACE, ""],
+ ["img", Parser::HTML_NAMESPACE, ""],
+ ["input", Parser::HTML_NAMESPACE, ""],
+ ["link", Parser::HTML_NAMESPACE, ""],
+ ["meta", Parser::HTML_NAMESPACE, ""],
+ ["param", Parser::HTML_NAMESPACE, ""],
+ ["source", Parser::HTML_NAMESPACE, ""],
+ ["track", Parser::HTML_NAMESPACE, ""],
+ ["wbr", Parser::HTML_NAMESPACE, ""],
+ ["basefont", Parser::SVG_NAMESPACE, "EEK"],
+ ["bgsound", Parser::SVG_NAMESPACE, "EEK"],
+ ["frame", Parser::SVG_NAMESPACE, "EEK"],
+ ["keygen", Parser::SVG_NAMESPACE, "EEK"],
+ ["area", Parser::SVG_NAMESPACE, "EEK"],
+ ["base", Parser::SVG_NAMESPACE, "EEK"],
+ ["br", Parser::SVG_NAMESPACE, "EEK"],
+ ["col", Parser::SVG_NAMESPACE, "EEK"],
+ ["embed", Parser::SVG_NAMESPACE, "EEK"],
+ ["hr", Parser::SVG_NAMESPACE, "EEK"],
+ ["img", Parser::SVG_NAMESPACE, "EEK"],
+ ["input", Parser::SVG_NAMESPACE, "EEK"],
+ ["link", Parser::SVG_NAMESPACE, "EEK"],
+ ["meta", Parser::SVG_NAMESPACE, "EEK"],
+ ["param", Parser::SVG_NAMESPACE, "EEK"],
+ ["source", Parser::SVG_NAMESPACE, "EEK"],
+ ["track", Parser::SVG_NAMESPACE, "EEK"],
+ ["wbr", Parser::SVG_NAMESPACE, "EEK"],
+ ];
+ }
+
+ public function testOuterSerializeAnInvalidNode(): void {
+ $d = new \DOMDocument;
+ $a = $d->createAttribute("oops");
+ $this->expectExceptionObject(new Exception(Exception::UNSUPPORTED_NODE_TYPE, [\DOMAttr::class]));
+ Serializer::serializeOuter($a);
+ }
+
+ public function testInnerSerializeAnInvalidNode(): void {
+ $d = new \DOMDocument;
+ $t = $d->createTextNode("OOPS");
+ $this->expectExceptionObject(new Exception(Exception::UNSUPPORTED_NODE_TYPE, [\DOMText::class]));
+ Serializer::serializeInner($t);
+ }
+
+ protected function buildTree(array $data, bool $fragment, bool $formatOutput = false): \DOMNode {
+ $document = new \DOMDocument;
+ $document->formatOutput = $formatOutput;
+ if ($fragment) {
+ $document->appendChild($document->createElement("html"));
+ $out = $document->createDocumentFragment();
+ } else {
+ $out = $document;
+ }
+ $cur = $out;
+ $pad = 2;
+ // process each line in turn
+ for ($l = 0; $l < sizeof($data); $l++) {
+ preg_match('/^(\|\s+)(.+)/', $data[$l], $m);
+ // pop any parents as long as the padding of the line is less than the expected padding
+ $p = strlen((string) $m[1]);
+ assert($p >= 2 && $p <= $pad && !($p % 2), new \Exception("Input data is invalid on line ".($l + 1)));
+ while ($p < $pad) {
+ $pad -= 2;
+ $cur = $cur->parentNode;
+ }
+ // act based upon what the rest of the line looks like
+ $d = $m[2];
+ if (preg_match('/^$/', $d, $m)) {
+ // comment
+ $cur->appendChild($document->createComment($m[1]));
+ } elseif (preg_match('/^]*)(?: "([^"]*)" "([^"]*)")?)?>$/', $d, $m)) {
+ // doctype
+ $name = strlen((string) ($m[1] ?? "")) ? $m[1] : " ";
+ $public = strlen((string) ($m[2] ?? "")) ? $m[2] : "";
+ $system = strlen((string) ($m[3] ?? "")) ? $m[3] : "";
+ $cur->appendChild($document->implementation->createDocumentType($name, $public, $system));
+ } elseif (preg_match('/^<\?([^ ]+) ([^>]*)>$/', $d, $m)) {
+ // processing instruction
+ $cur->appendChild($document->createProcessingInstruction($m[1], $m[2]));
+ } elseif (preg_match('/^<(?:([^ ]+) )?([^>]+)>$/', $d, $m)) {
+ // element
+ $ns = strlen((string) $m[1]) ? (array_flip(Parser::NAMESPACE_MAP)[$m[1]] ?? $m[1]) : null;
+ $cur = $cur->appendChild($document->createElementNS($ns, self::coerceName($m[2])));
+ $pad += 2;
+ } elseif (preg_match('/^(?:([^" ]+) )?([^"=]+)="((?:[^"]|"(?!$))*)"$/', $d, $m)) {
+ // attribute
+ $ns = strlen((string) $m[1]) ? (array_flip(Parser::NAMESPACE_MAP)[$m[1]] ?? $m[1]) : "";
+ $this->elementSetAttribute($cur, $ns, $m[2], $m[3]);
+ } elseif (preg_match('/^"((?:[^"]|"(?!$))*)("?)$/', $d, $m)) {
+ // text
+ $t = $m[1];
+ while (!strlen((string) $m[2])) {
+ preg_match('/^((?:[^"]|"(?!$))*)("?)$/', $data[++$l], $m);
+ $t .= "\n".$m[1];
+ }
+ $cur->appendChild($document->createTextNode($t));
+ } else {
+ throw new \Exception("Input data is invalid on line ".($l + 1));
+ }
+ }
+ return $out;
+ }
+
+ protected function parseTreeTestFile(string $file): \Generator {
+ $index = 0;
+ $l = 0;
+ $lines = array_map(function($v) {
+ return rtrim($v, "\n");
+ }, file($file));
+ while ($l < sizeof($lines)) {
+ $pos = $l + 1;
+ assert(in_array($lines[$l], ["#document", "#fragment"]), new \Exception("Test $file #$index does not start with #document or #fragment tag at line ".($l + 1)));
+ $fragment = $lines[$l] === "#fragment";
+ // collect the test input
+ $data = [];
+ for (++$l; $l < sizeof($lines); $l++) {
+ if (preg_match('/^#(script-(on|off)|output)$/', $lines[$l])) {
+ break;
+ }
+ $data[] = $lines[$l];
+ }
+ // set the script mode, if present
+ assert(preg_match('/^#(script-(on|off)|output)$/', $lines[$l]) === 1, new \Exception("Test $file #$index follows data with something other than script flag or output at line ".($l + 1)));
+ $script = null;
+ if ($lines[$l] === "#script-off") {
+ $script = false;
+ $l++;
+ } elseif ($lines[$l] === "#script-on") {
+ $script = true;
+ $l++;
+ }
+ // collect the output string
+ $exp = [];
+ assert($lines[$l] === "#output", new \Exception("Test $file #$index follows input with something other than output at line ".($l + 1)));
+ for (++$l; $l < sizeof($lines); $l++) {
+ if ($lines[$l] === "" && in_array(($lines[$l + 1] ?? ""), ["#document", "#fragment"])) {
+ break;
+ }
+ assert(preg_match('/^([^#]|$)/', $lines[$l]) === 1, new \Exception("Test $file #$index contains unrecognized data after output at line ".($l + 1)));
+ $exp[] = $lines[$l];
+ }
+ $exp = implode("\n", $exp);
+ if (!$script) {
+ yield basename($file)." #$index (line $pos)" => [$data, $fragment, $exp];
+ }
+ $l++;
+ $index++;
+ }
+ }
+}
diff --git a/tests/cases/TestTokenizer.php b/tests/cases/TestTokenizer.php
index f76f1a6..5771cf6 100644
--- a/tests/cases/TestTokenizer.php
+++ b/tests/cases/TestTokenizer.php
@@ -44,10 +44,10 @@ class TestTokenizer extends \PHPUnit\Framework\TestCase {
];
/** @dataProvider provideStandardTokenizerTests */
- public function testStandardTokenizerTests(string $input, array $expected, int $state, string $open = null, array $expErrors) {
+ public function testStandardTokenizerTests(string $input, array $expected, int $state, ?string $open, ?array $expErrors) {
$config = new Config;
$config->encodingFallback = "UTF-8";
- $errorHandler = new ParseError;
+ $errorHandler = ($expErrors !== null) ? new ParseError : null;
// initialize a stack of open elements, possibly with an open element
$stack = new OpenElementsStack(null);
if ($open) {
@@ -71,12 +71,31 @@ class TestTokenizer extends \PHPUnit\Framework\TestCase {
}
} finally {
$actual = $this->normalizeTokens($actual);
- $errors = $this->formatErrors($errorHandler->errors);
$this->assertEquals($expected, $actual, $tokenizer->debugLog);
+ $errors = ($expErrors !== null) ? $this->formatErrors($errorHandler->errors) : null;
$this->assertEquals($expErrors, $errors, $tokenizer->debugLog);
}
}
+ /**
+ * @dataProvider provideStandardTokenizerTests
+ * @depends testStandardTokenizerTests
+ */
+ public function testStandardTokenizerTestsWithoutErrorReporting(string $input, array $expected, int $state, ?string $open, array $expErrors) {
+ $this->testStandardTokenizerTests($input, $expected, $state, $open, null);
+ }
+
+ /** @dataProvider provideNonstandardTokenizerTests */
+ public function testNonstandardTokenizerTests(string $input, array $expected, int $state, ?string $open, array $expErrors) {
+ $this->testStandardTokenizerTests($input, $expected, $state, $open, $expErrors);
+ }
+
+ public function provideNonstandardTokenizerTests(): iterable {
+ return [
+ ["\xFF", [new CharacterToken("\u{FFFD}"), new EOFToken], Tokenizer::DATA_STATE, "", [['code' => "noncharacter-in-input-stream", 'line' => 1, 'col' => 1]]],
+ ];
+ }
+
public function provideStandardTokenizerTests() {
$tests = [];
$blacklist = ["xmlViolation.test"];
diff --git a/tests/cases/TestTreeConstructor.php b/tests/cases/TestTreeConstructor.php
index bf377d4..fcd06e6 100644
--- a/tests/cases/TestTreeConstructor.php
+++ b/tests/cases/TestTreeConstructor.php
@@ -16,6 +16,7 @@ use MensBeam\HTML\Parser\Tokenizer;
use MensBeam\HTML\Parser\TreeConstructor;
/**
+ * @covers \MensBeam\HTML\Parser\Data
* @covers \MensBeam\HTML\Parser\Tokenizer
* @covers \MensBeam\HTML\Parser\TreeConstructor
* @covers \MensBeam\HTML\Parser\ActiveFormattingElementsList
@@ -122,11 +123,11 @@ class TestTreeConstructor extends \PHPUnit\Framework\TestCase {
}
protected function patchTest(string $data, $fragment, array $errors, array $exp): array {
- // When using the HTML namespace, xmlns attribute cannot be inserted due to a PHP limitation
+ // When using the HTML namespace, xmlns attributes lose their namespace due to a PHP limitation
if ($this->ns) {
for ($a = 0; $a < sizeof($exp); $a++) {
if (preg_match('/^\|\s+xmlns xmlns=/', $exp[$a])) {
- array_splice($exp, $a--, 1);
+ $exp[$a] = preg_replace('/^\|(\s+)xmlns xmlns=/', "|$1xmlns=", $exp[$a]);
}
}
}
@@ -180,7 +181,7 @@ class TestTreeConstructor extends \PHPUnit\Framework\TestCase {
$prefix = "null ";
}
}
- $localName = $this->uncoerceName($e->localName);
+ $localName = self::uncoerceName($e->localName);
$this->push("<".$prefix.$localName.">");
$this->depth++;
$attr = [];
@@ -191,7 +192,7 @@ class TestTreeConstructor extends \PHPUnit\Framework\TestCase {
assert((bool) $prefix, new \Exception("Prefix for namespace {$a->namespaceURI} is not defined"));
$prefix .= " ";
}
- $attr[$prefix.$this->uncoerceName($a->name)] = $a->value;
+ $attr[$prefix.self::uncoerceName($a->name)] = $a->value;
}
ksort($attr, \SORT_STRING);
foreach ($attr as $k => $v) {
diff --git a/tests/cases/serializer/README.md b/tests/cases/serializer/README.md
new file mode 100644
index 0000000..824bf7e
--- /dev/null
+++ b/tests/cases/serializer/README.md
@@ -0,0 +1,99 @@
+HTML DOM serialization tests
+============================
+
+The format of these tests is essentially the format of html5lib's tree
+construction tests in reverse. There are, however, important differences,
+so the format is documented in full here.
+
+Each file containing tree construction tests consists of any number of
+tests separated by two newlines (LF) and a single newline before the end
+of the file. For instance:
+
+ [TEST]LF
+ LF
+ [TEST]LF
+ LF
+ [TEST]LF
+
+Where [TEST] is the following format:
+
+Each test begins with a line reading `#document` or `#fragment`; subsequent
+lines represent the document or document fragment (respectively) used as
+input, until a line is encountered which reads `#output`, `#script-on`,
+or `#script-off`.
+
+Each DOM node in the input is written on its own line beginning with the
+characters "| " (a vertical bar followed by a single space); lines which begin
+with other characters are a continuation of the previous line. Attributes
+are treated as distinct nodes and have their own entries. There is no escape
+mechanism: all input is literal, including newlines and quotation marks. Two
+spaces are used to denote each level of nesting. For example:
+
+ | node
+ | child node
+ continuation of child node
+ | grandchild node
+ | child node
+ | attribute node of child
+ | grandchild node
+
+The different types of nodes are:
+
+- Element nodes in the form `` for an element in the HTML namespace,
+ or `