Merge pull request #194 from bytestream/anchor-target-query-param

goetas · web-flow · commit dafd1c0908d7 · 2020-09-01T11:06:08.000+02:00
fix: query parameter parsed as character entity
diff --git a/src/HTML5/Parser/Tokenizer.php b/src/HTML5/Parser/Tokenizer.php
@@ -1181,16 +1181,11 @@ protected function decodeCharacterReference($inAttribute = false)
             return $entity;
         }
 
-        // If in an attribute, then failing to match ; means unconsume the
-        // entire string. Otherwise, failure to match is an error.
-        if ($inAttribute) {
-            $this->scanner->unconsume($this->scanner->position() - $start);
-
-            return '&';
-        }
+        // Failing to match ; means unconsume the entire string.
+        $this->scanner->unconsume($this->scanner->position() - $start);
 
         $this->parseError('Expected &ENTITY;, got &ENTITY%s (no trailing ;) ', $tok);
 
-        return '&' . $entity;
+        return '&';
     }
 }
diff --git a/test/HTML5/Html5Test.php b/test/HTML5/Html5Test.php
@@ -480,4 +480,13 @@ public function testCDATA()
         $res = $this->cycleFragment('a<![CDATA[ This <is> a test. ]]>b');
         $this->assertRegExp('|<!\[CDATA\[ This <is> a test\. \]\]>|', $res);
     }
+
+    public function testAnchorTargetQueryParam()
+    {
+        $res = $this->cycle('<a href="https://domain.com/page.php?foo=bar&target=baz">https://domain.com/page.php?foo=bar&target=baz</a>');
+        $this->assertContains(
+            '<a href="https://domain.com/page.php?foo=bar&amp;target=baz">https://domain.com/page.php?foo=bar&amp;target=baz</a>',
+            $res
+        );
+    }
 }

Original file line number	Diff line number	Diff line change
`@@ -1181,16 +1181,11 @@ protected function decodeCharacterReference($inAttribute = false)`
`1181`	`1181`	`return $entity;`
`1182`	`1182`	`}`
`1183`	`1183`
`1184`		`- // If in an attribute, then failing to match ; means unconsume the`
`1185`		`- // entire string. Otherwise, failure to match is an error.`
`1186`		`- if ($inAttribute) {`
`1187`		`- $this->scanner->unconsume($this->scanner->position() - $start);`
`1188`		`-`
`1189`		`- return '&';`
`1190`		`- }`
	`1184`	`+ // Failing to match ; means unconsume the entire string.`
	`1185`	`+ $this->scanner->unconsume($this->scanner->position() - $start);`
`1191`	`1186`
`1192`	`1187`	`$this->parseError('Expected &ENTITY;, got &ENTITY%s (no trailing ;) ', $tok);`
`1193`	`1188`
`1194`		`- return '&' . $entity;`
	`1189`	`+ return '&';`
`1195`	`1190`	`}`
`1196`	`1191`	`}`
Original file line number	Diff line number	Diff line change
`@@ -480,4 +480,13 @@ public function testCDATA()`
`480`	`480`	`$res = $this->cycleFragment('a<![CDATA[ This <is> a test. ]]>b');`
`481`	`481`	`$this->assertRegExp('\|<!\[CDATA\[ This <is> a test\. \]\]>\|', $res);`
`482`	`482`	`}`
	`483`	`+`
	`484`	`+ public function testAnchorTargetQueryParam()`
	`485`	`+ {`
	`486`	`+ $res = $this->cycle('<a href="https://domain.com/page.php?foo=bar&target=baz">https://domain.com/page.php?foo=bar&target=baz</a>');`
	`487`	`+ $this->assertContains(`
	`488`	`+ '<a href="https://domain.com/page.php?foo=bar&target=baz">https://domain.com/page.php?foo=bar&target=baz</a>',`
	`489`	`+ $res`
	`490`	`+ );`
	`491`	`+ }`
`483`	`492`	`}`