1
0
Fork 0

apply a regex solution instead of tokenizer

pull/4208/head
Rob Bast 2015-07-03 10:15:17 +02:00
parent e4118385a0
commit b5d286e27b
2 changed files with 52 additions and 100 deletions

View File

@ -192,115 +192,69 @@ class SpdxLicense
* @param string $license * @param string $license
* *
* @return bool * @return bool
*
* @throws \RuntimeException * @throws \RuntimeException
*/ */
private function isValidLicenseString($license) private function isValidLicenseString($license)
{ {
$tokens = array( $licenses = array_map('preg_quote', array_keys($this->licenses));
'po' => '\(', sort($licenses);
'pc' => '\)', $licenses = array_reverse($licenses);
'op' => '(?:or|OR|and|AND)', $licenses = implode('|', $licenses);
'wi' => '(?:with|WITH)',
'lix' => '(?:NONE|NOASSERTION)',
'lir' => 'LicenseRef-\d+',
'lic' => '[-_.a-zA-Z0-9]{3,}\+?',
'ws' => '\s+',
'_' => '.',
);
$next = function () use ($license, $tokens) { $exceptions = array_map('preg_quote', array_keys($this->exceptions));
static $offset = 0; sort($exceptions);
$exceptions = array_reverse($exceptions);
$exceptions = implode('|', $exceptions);
if ($offset >= strlen($license)) { $regex = "{
return null; (?(DEFINE)
} # idstring: 1*( ALPHA / DIGIT / - / . )
(?<idstring>[\pL\pN\-\.]{1,})
foreach ($tokens as $name => $token) { # license-id: taken from list
if (false === $r = preg_match('{' . $token . '}', $license, $matches, PREG_OFFSET_CAPTURE, $offset)) { (?<licenseid>${licenses})
throw new \RuntimeException('Pattern for token %s failed (regex error).', $name);
}
if ($r === 0) {
continue;
}
if ($matches[0][1] !== $offset) {
continue;
}
$offset += strlen($matches[0][0]);
return array($name, $matches[0][0]); # license-exception-id: taken from list
} (?<licenseexceptionid>${exceptions})
throw new \RuntimeException( # license-ref: [DocumentRef-1*(idstring):]LicenseRef-1*(idstring)
'At least the last pattern needs to match, but it did not (dot-match-all is missing?).' (?<licenseref>(?:DocumentRef-(?&idstring):)?LicenseRef-(?&idstring))
);
};
$open = 0; # simple-expresssion: license-id / license-id+ / license-ref
$with = false; (?<simple_expression>(?&licenseid)\+? | (?&licenseid) | (?&licenseref))
$require = true;
$lastop = null;
while (list($token, $string) = $next()) { # compound expression: 1*(
switch ($token) { # simple-expression /
case 'po': # simple-expression WITH license-exception-id /
if ($open || !$require || $with) { # compound-expression AND compound-expression /
return false; # compound-expression OR compound-expression
} # ) / ( compound-expression ) )
$open = 1; (?<compound_head>
break; (?&simple_expression) ( \s+ (?:with|WITH) \s+ (?&licenseexceptionid))?
case 'pc': | \( \s* (?&compound_expression) \s*\)
if ($open !== 1 || $require || !$lastop || $with) { )
return false; (?<compound_expression>
} (?&compound_head) (?: \s+ (?:and|AND|or|OR) \s+ (?&compound_expression))?
$open = 2; )
break;
case 'op': # license-expression: 1*1(simple-expression / compound-expression)
if ($require || !$open || $with) { (?<license_expression>NONE | NOASSERTION | (?&compound_expression) | (?&simple_expression))
return false; ) # end of define
}
$lastop || $lastop = $string; ^(?&license_expression)$
if ($lastop !== $string) { }x";
return false;
} $match = preg_match($regex, $license);
$require = true;
break; if (0 === $match) {
case 'wi': return false;
$with = true;
break;
case 'lix':
if ($open || $with) {
return false;
}
goto lir;
case 'lic':
if ($with && $this->isValidExceptionIdentifier($string)) {
$require = true;
$with = false;
goto lir;
}
if ($with) {
return false;
}
if (!$this->isValidLicenseIdentifier(rtrim($string, '+'))) {
return false;
}
// Fall-through intended
case 'lir':
lir:
if (!$require) {
return false;
}
$require = false;
break;
case 'ws':
break;
case '_':
return false;
default:
throw new \RuntimeException(sprintf('Unparsed token: %s.', print_r($token, true)));
}
} }
return !($open % 2 || $require || $with); if (false === $match) {
throw new \RuntimeException('Regex failed to compile/run.');
}
return true;
} }
} }

View File

@ -39,6 +39,7 @@ class SpdxLicenseTest extends TestCase
"GPL-2.0 with Autoconf-exception-2.0", "GPL-2.0 with Autoconf-exception-2.0",
"GPL-2.0 WITH Autoconf-exception-2.0", "GPL-2.0 WITH Autoconf-exception-2.0",
"GPL-2.0+ WITH Autoconf-exception-2.0", "GPL-2.0+ WITH Autoconf-exception-2.0",
"(GPL-3.0 and GPL-2.0 or GPL-3.0+)",
), ),
$identifiers $identifiers
); );
@ -57,7 +58,6 @@ class SpdxLicenseTest extends TestCase
array(array()), array(array()),
array("The system pwns you"), array("The system pwns you"),
array("()"), array("()"),
array("(MIT)"),
array("(MIT"), array("(MIT"),
array("MIT)"), array("MIT)"),
array("MIT NONE"), array("MIT NONE"),
@ -66,8 +66,6 @@ class SpdxLicenseTest extends TestCase
array("(MIT and MIT) MIT"), array("(MIT and MIT) MIT"),
array(array("LGPL-2.0", "The system pwns you")), array(array("LGPL-2.0", "The system pwns you")),
array("and GPL-3.0+"), array("and GPL-3.0+"),
array("EUDatagrid and GPL-3.0+"),
array("(GPL-3.0 and GPL-2.0 or GPL-3.0+)"),
array("(EUDatagrid and GPL-3.0+ and )"), array("(EUDatagrid and GPL-3.0+ and )"),
array("(EUDatagrid xor GPL-3.0+)"), array("(EUDatagrid xor GPL-3.0+)"),
array("(MIT Or MIT)"), array("(MIT Or MIT)"),