feature #17863 [Yaml] add support for parsing the !!binary tag (xabbuh)

This PR was merged into the 3.1-dev branch.

Discussion
----------

[Yaml] add support for parsing the !!binary tag

| Q             | A
| ------------- | ---
| Bug fix?      | no
| New feature?  | yes
| BC breaks?    | no
| Deprecations? | no
| Tests pass?   | yes
| Fixed tickets | #8094, #15587, #17599
| License       | MIT
| Doc PR        | TODO

Commits
-------

79a63d5 [Yaml] add support for the !!binary tag

Author: fabpot

src/Symfony/Component/Yaml/CHANGELOG.md

@@ -4,6 +4,12 @@ CHANGELOG
 3.1.0
 -----
 
+ * Added support for parsing base64 encoded binary data when they are tagged
+   with the `!!binary` tag.
+
+ * Added support for dumping binary data as base64 encoded strings by passing
+   the `Yaml::DUMP_BASE64_BINARY_DATA` flag.
+
  * Added support for parsing timestamps as `\DateTime` objects:
 
    ```php

src/Symfony/Component/Yaml/Inline.php

@@ -197,6 +197,8 @@ public static function dump($value, $flags = 0)
                 return $repr;
             case '' == $value:
                 return "''";
+            case Yaml::DUMP_BASE64_BINARY_DATA & $flags && self::isBinaryString($value):
+                return '!!binary '.base64_encode($value);
             case Escaper::requiresDoubleQuoting($value):
                 return Escaper::escapeWithDoubleQuotes($value);
             case Escaper::requiresSingleQuoting($value):
@@ -576,6 +578,8 @@ private static function evaluateScalar($scalar, $flags, $references = array())
                         return -log(0);
                     case '-.inf' === $scalarLower:
                         return log(0);
+                    case 0 === strpos($scalar, '!!binary '):
+                        return self::evaluateBinaryScalar(substr($scalar, 9));
                     case preg_match('/^(-|\+)?[0-9,]+(\.[0-9]+)?$/', $scalar):
                         return (float) str_replace(',', '', $scalar);
                     case preg_match(self::getTimestampRegex(), $scalar):
@@ -595,6 +599,33 @@ private static function evaluateScalar($scalar, $flags, $references = array())
         }
     }
 
+    /**
+     * @param string $scalar
+     *
+     * @return string
+     *
+     * @internal
+     */
+    public static function evaluateBinaryScalar($scalar)
+    {
+        $parsedBinaryData = self::parseScalar(preg_replace('/\s/', '', $scalar));
+
+        if (0 !== (strlen($parsedBinaryData) % 4)) {
+            throw new ParseException(sprintf('The normalized base64 encoded data (data without whitespace characters) length must be a multiple of four (%d bytes given).', strlen($parsedBinaryData)));
+        }
+
+        if (!preg_match('#^[A-Z0-9+/]+={0,2}$#i', $parsedBinaryData)) {
+            throw new ParseException(sprintf('The base64 encoded data (%s) contains invalid characters.', $parsedBinaryData));
+        }
+
+        return base64_decode($parsedBinaryData, true);
+    }
+
+    private static function isBinaryString($value)
+    {
+        return preg_match('/[^\x09-\x0d\x20-\xff]/', $value);
+    }
+
     /**
      * Gets a regex that matches a YAML date.
      *

src/Symfony/Component/Yaml/Parser.php

@@ -20,6 +20,7 @@
  */
 class Parser
 {
+    const TAG_PATTERN = '((?P<tag>![\w!.\/:-]+) +)?';
     const BLOCK_SCALAR_HEADER_PATTERN = '(?P<separator>\||>)(?P<modifiers>\+|\-|\d+|\+\d+|\-\d+|\d+\+|\d+\-)?(?P<comments> +#.*)?';
 
     private $offset = 0;
@@ -516,10 +517,16 @@ private function parseValue($value, $flags, $context)
             return $this->refs[$value];
         }
 
-        if (preg_match('/^'.self::BLOCK_SCALAR_HEADER_PATTERN.'$/', $value, $matches)) {
+        if (preg_match('/^'.self::TAG_PATTERN.self::BLOCK_SCALAR_HEADER_PATTERN.'$/', $value, $matches)) {
             $modifiers = isset($matches['modifiers']) ? $matches['modifiers'] : '';
 
-            return $this->parseBlockScalar($matches['separator'], preg_replace('#\d+#', '', $modifiers), (int) abs($modifiers));
+            $data = $this->parseBlockScalar($matches['separator'], preg_replace('#\d+#', '', $modifiers), (int) abs($modifiers));
+
+            if (isset($matches['tag']) && '!!binary' === $matches['tag']) {
+                return Inline::evaluateBinaryScalar($data);
+            }
+
+            return $data;
         }
 
         try {

src/Symfony/Component/Yaml/Tests/DumperTest.php

@@ -276,6 +276,22 @@ public function getEscapeSequences()
             'paragraph-separator' => array("\t\\P", '"\t\\\\P"'),
         );
     }
+
+    public function testBinaryDataIsDumpedAsIsWithoutFlag()
+    {
+        $binaryData = file_get_contents(__DIR__.'/Fixtures/arrow.gif');
+        $expected = "{ data: '".str_replace("'", "''", $binaryData)."' }";
+
+        $this->assertSame($expected, $this->dumper->dump(array('data' => $binaryData)));
+    }
+
+    public function testBinaryDataIsDumpedBase64EncodedWithFlag()
+    {
+        $binaryData = file_get_contents(__DIR__.'/Fixtures/arrow.gif');
+        $expected = '{ data: !!binary '.base64_encode($binaryData).' }';
+
+        $this->assertSame($expected, $this->dumper->dump(array('data' => $binaryData), 0, 0, Yaml::DUMP_BASE64_BINARY_DATA));
+    }
 }
 
 class A

src/Symfony/Component/Yaml/Tests/Fixtures/arrow.gif

@@ -532,4 +532,41 @@ public function getDateTimeDumpTests()
 
         return $tests;
     }
+
+    /**
+     * @dataProvider getBinaryData
+     */
+    public function testParseBinaryData($data)
+    {
+        $this->assertSame('Hello world', Inline::parse($data));
+    }
+
+    public function getBinaryData()
+    {
+        return array(
+            'enclosed with double quotes' => array('!!binary "SGVsbG8gd29ybGQ="'),
+            'enclosed with single quotes' => array("!!binary 'SGVsbG8gd29ybGQ='"),
+            'containing spaces' => array('!!binary  "SGVs bG8gd 29ybGQ="'),
+        );
+    }
+
+    /**
+     * @dataProvider getInvalidBinaryData
+     */
+    public function testParseInvalidBinaryData($data, $expectedMessage)
+    {
+        $this->setExpectedExceptionRegExp('\Symfony\Component\Yaml\Exception\ParseException', $expectedMessage);
+
+        Inline::parse($data);
+    }
+
+    public function getInvalidBinaryData()
+    {
+        return array(
+            'length not a multiple of four' => array('!!binary "SGVsbG8d29ybGQ="', '/The normalized base64 encoded data \(data without whitespace characters\) length must be a multiple of four \(\d+ bytes given\)/'),
+            'invalid characters' => array('!!binary "SGVsbG8#d29ybGQ="', '/The base64 encoded data \(.*\) contains invalid characters/'),
+            'too many equals characters' => array('!!binary "SGVsbG8gd29yb==="', '/The base64 encoded data \(.*\) contains invalid characters/'),
+            'misplaced equals character' => array('!!binary "SGVsbG8gd29ybG=Q"', '/The base64 encoded data \(.*\) contains invalid characters/'),
+        );
+    }
 }

src/Symfony/Component/Yaml/Tests/InlineTest.php

@@ -1120,6 +1120,87 @@ public function testAdditionallyIndentedLinesAreParsedAsNewLinesInFoldedBlocks()
             $this->parser->parse($yaml)
         );
     }
+
+    /**
+     * @dataProvider getBinaryData
+     */
+    public function testParseBinaryData($data)
+    {
+        $this->assertSame(array('data' => 'Hello world'), $this->parser->parse($data));
+    }
+
+    public function getBinaryData()
+    {
+        return array(
+            'enclosed with double quotes' => array('data: !!binary "SGVsbG8gd29ybGQ="'),
+            'enclosed with single quotes' => array("data: !!binary 'SGVsbG8gd29ybGQ='"),
+            'containing spaces' => array('data: !!binary  "SGVs bG8gd 29ybGQ="'),
+            'in block scalar' => array(
+                <<<EOT
+data: !!binary |
+    SGVsbG8gd29ybGQ=
+EOT
+    ),
+            'containing spaces in block scalar' => array(
+                <<<EOT
+data: !!binary |
+    SGVs bG8gd 29ybGQ=
+EOT
+    ),
+        );
+    }
+
+    /**
+     * @dataProvider getInvalidBinaryData
+     */
+    public function testParseInvalidBinaryData($data, $expectedMessage)
+    {
+        $this->setExpectedExceptionRegExp('\Symfony\Component\Yaml\Exception\ParseException', $expectedMessage);
+
+        $this->parser->parse($data);
+    }
+
+    public function getInvalidBinaryData()
+    {
+        return array(
+            'length not a multiple of four' => array('data: !!binary "SGVsbG8d29ybGQ="', '/The normalized base64 encoded data \(data without whitespace characters\) length must be a multiple of four \(\d+ bytes given\)/'),
+            'invalid characters' => array('!!binary "SGVsbG8#d29ybGQ="', '/The base64 encoded data \(.*\) contains invalid characters/'),
+            'too many equals characters' => array('data: !!binary "SGVsbG8gd29yb==="', '/The base64 encoded data \(.*\) contains invalid characters/'),
+            'misplaced equals character' => array('data: !!binary "SGVsbG8gd29ybG=Q"', '/The base64 encoded data \(.*\) contains invalid characters/'),
+            'length not a multiple of four in block scalar' => array(
+                <<<EOT
+data: !!binary |
+    SGVsbG8d29ybGQ=
+EOT
+                ,
+                '/The normalized base64 encoded data \(data without whitespace characters\) length must be a multiple of four \(\d+ bytes given\)/',
+            ),
+            'invalid characters in block scalar' => array(
+                <<<EOT
+data: !!binary |
+    SGVsbG8#d29ybGQ=
+EOT
+                ,
+                '/The base64 encoded data \(.*\) contains invalid characters/',
+            ),
+            'too many equals characters in block scalar' => array(
+                <<<EOT
+data: !!binary |
+    SGVsbG8gd29yb===
+EOT
+                ,
+                '/The base64 encoded data \(.*\) contains invalid characters/',
+            ),
+            'misplaced equals character in block scalar' => array(
+                <<<EOT
+data: !!binary |
+    SGVsbG8gd29ybG=Q
+EOT
+                ,
+                '/The base64 encoded data \(.*\) contains invalid characters/',
+            ),
+        );
+    }
 }
 
 class B

src/Symfony/Component/Yaml/Tests/ParserTest.php

@@ -26,6 +26,7 @@ class Yaml
     const PARSE_OBJECT_FOR_MAP = 8;
     const DUMP_EXCEPTION_ON_INVALID_TYPE = 16;
     const PARSE_DATETIME = 32;
+    const DUMP_BASE64_BINARY_DATA = 64;
 
     /**
      * Parses YAML into a PHP value.