Fix parsing in Header::normalize() (#476)

Co-Authored-By: Tim Düsterhus <209270+TimWolla@users.noreply.github.com> Co-authored-by: Tim Düsterhus <209270+TimWolla@users.noreply.github.com>
guzzle · Mar 13, 2022 · d55bd56 · d55bd56
1 parent ff76777
commit d55bd56
Show file tree

Hide file tree

Showing 2 changed files with 94 additions and 11 deletions.
diff --git a/src/Header.php b/src/Header.php
@@ -47,19 +47,21 @@ public static function parse($header): array
      */
     public static function normalize($header): array
     {
-        if (!is_array($header)) {
-            return array_map('trim', explode(',', $header));
-        }
-
         $result = [];
-        foreach ($header as $value) {
+        foreach ((array) $header as $value) {
             foreach ((array) $value as $v) {
                 if (strpos($v, ',') === false) {
-                    $result[] = $v;
+                    $trimmed = trim($v);
+                    if ($trimmed !== '') {
+                        $result[] = $trimmed;
+                    }
                     continue;
                 }
-                foreach (preg_split('/,(?=([^"]*"[^"]*")*[^"]*$)/', $v) as $vv) {
-                    $result[] = trim($vv);
+                foreach (preg_split('/,(?=([^"]*"([^"]|\\\\.)*")*[^"]*$)/', $v) as $vv) {
+                    $trimmed = trim($vv);
+                    if ($trimmed !== '') {
+                        $result[] = $trimmed;
+                    }
                 }
             }
         }

diff --git a/tests/HeaderTest.php b/tests/HeaderTest.php
@@ -63,9 +63,90 @@ public function testParseParams($header, $result): void
         self::assertSame($result, Psr7\Header::parse($header));
     }
 
-    public function testParsesArrayHeaders(): void
+    public function normalizeProvider(): array
     {
-        $header = ['a, b', 'c', 'd, e'];
-        self::assertSame(['a', 'b', 'c', 'd', 'e'], Psr7\Header::normalize($header));
+        return [
+            [
+                '',
+                [],
+            ],
+            [
+                ['a, b', 'c', 'd, e'],
+                ['a', 'b', 'c', 'd', 'e'],
+            ],
+            // Example 'accept-encoding'
+            [
+                'gzip, br',
+                ['gzip', 'br'],
+            ],
+            // https://httpwg.org/specs/rfc7231.html#rfc.section.5.3.2
+            [
+                'text/plain; q=0.5, text/html, text/x-dvi; q=0.8, text/x-c',
+                ['text/plain; q=0.5', 'text/html', 'text/x-dvi; q=0.8', 'text/x-c'],
+            ],
+            // Example 'If-None-Match' with comma within an ETag
+            [
+                '"foo", "foo,bar", "bar"',
+                ['"foo"', '"foo,bar"', '"bar"'],
+            ],
+            // https://httpwg.org/specs/rfc7234.html#cache.control.extensions
+            [
+                'private, community="UCI"',
+                ['private', 'community="UCI"'],
+            ],
+            // The Cache-Control example with a comma within a community
+            [
+                'private, community="Guzzle,Psr7"',
+                ['private', 'community="Guzzle,Psr7"'],
+            ],
+            // The Cache-Control example with an escaped space (quoted-pair) within a community
+            [
+                'private, community="Guzzle\\ Psr7"',
+                ['private', 'community="Guzzle\\ Psr7"'],
+            ],
+            // The Cache-Control example with an escaped quote (quoted-pair) within a community
+            [
+                'private, community="Guzzle\\"Psr7"',
+                ['private', 'community="Guzzle\\"Psr7"'],
+            ],
+            // The Cache-Control example with an escaped quote (quoted-pair) and a comma within a community
+            [
+                'private, community="Guzzle\\",Psr7"',
+                ['private', 'community="Guzzle\\",Psr7"'],
+            ],
+            // The Cache-Control example with an escaped backslash (quoted-pair) within a community
+            [
+                'private, community="Guzzle\\\\Psr7"',
+                ['private', 'community="Guzzle\\\\Psr7"'],
+            ],
+            // The Cache-Control example with an escaped backslash (quoted-pair) within a community
+            [
+                'private, community="Guzzle\\\\", Psr7',
+                ['private', 'community="Guzzle\\\\"', 'Psr7'],
+            ],
+            // https://httpwg.org/specs/rfc7230.html#rfc.section.7
+            [
+                'foo ,bar,',
+                ['foo', 'bar'],
+            ],
+            // https://httpwg.org/specs/rfc7230.html#rfc.section.7
+            [
+                'foo , ,bar,charlie   ',
+                ['foo', 'bar', 'charlie'],
+            ],
+            [
+                "<https://example.gitlab.com>; rel=\"first\",\n<https://example.gitlab.com>; rel=\"next\",\n<https://example.gitlab.com>; rel=\"prev\",\n<https://example.gitlab.com>; rel=\"last\",",
+                ['<https://example.gitlab.com>; rel="first"', '<https://example.gitlab.com>; rel="next"', '<https://example.gitlab.com>; rel="prev"', '<https://example.gitlab.com>; rel="last"'],
+            ],
+        ];
+    }
+
+    /**
+     * @dataProvider normalizeProvider
+     */
+    public function testNormalize($header, $result): void
+    {
+        self::assertSame($result, Psr7\Header::normalize([$header]));
+        self::assertSame($result, Psr7\Header::normalize($header));
     }
 }