More PI tests

3 years ago · 6ee9f0ccc2
4 changed files with 55 additions and 4 deletions
--- a/lib/Parser/TreeBuilder.php
+++ b/lib/Parser/TreeBuilder.php
@ -3894,7 +3894,7 @@ class TreeBuilder {
        # 3. If the adjusted insertion location is inside a template element, let it
        # instead be inside the template element’s template contents, after its last
        # child (if any).
-        if ($insertionLocation instanceof Element && $insertionLocation->nodeName === 'template' && $insertionLocation->namespaceURI === null) {
+        if ($insertionLocation instanceof \DOMElement && $insertionLocation->nodeName === 'template' && $insertionLocation->namespaceURI === null) {
            // DEVIATION: We don't implement template contents in the parser itself
            $insertionLocation = $insertionLocation;
        }
@ -3967,7 +3967,12 @@ class TreeBuilder {
            // see https://www.w3.org/TR/xml/#d0e1188
            && preg_match('/^\?(?![Xx][Mm][Ll](?:[ \x{9}\x{D}\x{A}]|$))([:A-Z_a-z\x{C0}-\x{D6}\x{D8}-\x{F6}\x{F8}-\x{2FF}\x{370}-\x{37D}\x{37F}-\x{1FFF}\x{200C}-\x{200D}\x{2070}-\x{218F}\x{2C00}-\x{2FEF}\x{3001}-\x{D7FF}\x{F900}-\x{FDCF}\x{FDF0}-\x{FFFD}\x{10000}-\x{EFFFF}][:A-Z_a-z\x{C0}-\x{D6}\x{D8}-\x{F6}\x{F8}-\x{2FF}\x{370}-\x{37D}\x{37F}-\x{1FFF}\x{200C}-\x{200D}\x{2070}-\x{218F}\x{2C00}-\x{2FEF}\x{3001}-\x{D7FF}\x{F900}-\x{FDCF}\x{FDF0}-\x{FFFD}\x{10000}-\x{EFFFF}-\.0-9\x{B7}\x{0300}-\x{036F}\x{203F}-\x{2040}]*)(?:[ \x{9}\x{D}\x{A}](.*))?/suD', $token->data, $m)
        ) {
-            $node = $this->DOM->createProcessingInstruction($m[1], $m[2]);
+            try {
+                $node = $this->DOM->createProcessingInstruction($m[1], $m[2]);
+            } catch (\DOMException $e) {
+                // PHP's DOM does not implement the laxer Name production of the fifth edition of XML; if an otherwise valid PI target cannot be used, create a comment instead
+                $node = $this->DOM->createComment($token->data);
+            }
        } else {
            $node = $this->DOM->createComment($token->data);
        }
@ -4243,9 +4248,8 @@ class TreeBuilder {
                // 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
-                $element->ownerDocument->mangledAttributes = true;
                if ($namespaceURI !== null) {
-                    $qualifiedName = implode(":", array_map([$element, "coerceName"], explode(":", $qualifiedName, 2)));
+                    $qualifiedName = implode(":", array_map([$this, "coerceName"], explode(":", $qualifiedName, 2)));
                } else {
                    $qualifiedName = $this->coerceName($qualifiedName);
                }
--- a/tests/cases/TestEncodingChange.php
+++ b/tests/cases/TestEncodingChange.php
@ -13,6 +13,7 @@ use MensBeam\HTML\Parser\Config;
 /**
 * @covers \MensBeam\HTML\Parser
 * @covers \MensBeam\HTML\Parser\TreeBuilder
+ * @covers \MensBeam\HTML\Parser\TagToken
 * @covers \MensBeam\HTML\Parser\Data::changeEncoding
 */
 class TestEncodingChange extends \PHPUnit\Framework\TestCase {
--- a/tests/cases/TestTreeConstructor.php
+++ b/tests/cases/TestTreeConstructor.php
@ -25,6 +25,7 @@ use MensBeam\HTML\Parser\TreeBuilder;
 * @covers \MensBeam\HTML\Parser\OpenElementsStack
 * @covers \MensBeam\HTML\Parser\Stack
 * @covers \MensBeam\HTML\Parser\TagToken
+ * @covers \MensBeam\HTML\Parser\ProcessingInstructionToken
 */
 class TestTreeConstructor extends \PHPUnit\Framework\TestCase {
    use \MensBeam\HTML\Parser\NameCoercion;
--- a/tests/cases/tree-construction/pi01.dat
+++ b/tests/cases/tree-construction/pi01.dat
@ -19,3 +19,48 @@
 |   <!-- ?xml oops -->
 |   <head>
 |   <body>
+
+#data
+<!DOCTYPE html><html><?xml-stylesheet blah>
+#errors
+(1,23): unexpected-question-mark-instead-of-tag-name
+#document
+| <!DOCTYPE html>
+| <html>
+|   <?xml-stylesheet blah>
+|   <head>
+|   <body>
+
+#data
+<!DOCTYPE html><html><?xml-stylesheet blah ?>
+#errors
+(1,23): unexpected-question-mark-instead-of-tag-name
+#document
+| <!DOCTYPE html>
+| <html>
+|   <?xml-stylesheet blah ?>
+|   <head>
+|   <body>
+
+#data
+<!DOCTYPE html><html><?test
+moop>
+#errors
+(1,23): unexpected-question-mark-instead-of-tag-name
+#document
+| <!DOCTYPE html>
+| <html>
+|   <?test moop>
+|   <head>
+|   <body>
+
+#data
+<!DOCTYPE html><html><?💩 This PI target should be valid, but PHP does not accept it>
+#errors
+(1,23): unexpected-question-mark-instead-of-tag-name
+#document
+| <!DOCTYPE html>
+| <html>
+|   <!-- ?💩 This PI target should be valid, but PHP does not accept it -->
+|   <head>
+|   <body>