silverstripe-framework/core/Convert.php

492 lines
14 KiB
PHP
Raw Normal View History

<?php
use SilverStripe\ORM\DB;
/**
* Library of conversion functions, implemented as static methods.
*
* The methods are all of the form (format)2(format), where the format is one of
2014-08-15 08:53:05 +02:00
*
* raw: A UTF8 string
* attr: A UTF8 string suitable for inclusion in an HTML attribute
* js: A UTF8 string suitable for inclusion in a double-quoted javascript string.
2014-08-15 08:53:05 +02:00
*
* array: A PHP associative array
* json: JavaScript object notation
*
* html: HTML source suitable for use in a page or email
* text: Plain-text content, suitable for display to a user as-is, or insertion in a plaintext email.
2014-08-15 08:53:05 +02:00
*
* Objects of type {@link ViewableData} can have an "escaping type",
2014-08-15 08:53:05 +02:00
* which determines if they are automatically escaped before output by {@link SSViewer}.
*
* @package framework
* @subpackage misc
*/
class Convert {
2014-08-15 08:53:05 +02:00
/**
* Convert a value to be suitable for an XML attribute.
2014-08-15 08:53:05 +02:00
*
* @param array|string $val String to escape, or array of strings
* @return array|string
*/
public static function raw2att($val) {
return self::raw2xml($val);
}
/**
* Convert a value to be suitable for an HTML attribute.
2014-08-15 08:53:05 +02:00
*
* @param string|array $val String to escape, or array of strings
* @return array|string
*/
public static function raw2htmlatt($val) {
return self::raw2att($val);
}
/**
2013-05-26 01:01:46 +02:00
* Convert a value to be suitable for an HTML ID attribute. Replaces non
* supported characters with a space.
2014-08-15 08:53:05 +02:00
*
* @see http://www.w3.org/TR/REC-html40/types.html#type-cdata
2013-05-26 01:01:46 +02:00
*
* @param array|string $val String to escape, or array of strings
2013-05-26 01:01:46 +02:00
*
* @return array|string
*/
public static function raw2htmlname($val) {
if(is_array($val)) {
2013-05-26 01:01:46 +02:00
foreach($val as $k => $v) {
$val[$k] = self::raw2htmlname($v);
}
return $val;
} else {
return self::raw2att($val);
}
}
/**
* Convert a value to be suitable for an HTML ID attribute. Replaces non
* supported characters with an underscore.
*
* @see http://www.w3.org/TR/REC-html40/types.html#type-cdata
*
* @param array|string $val String to escape, or array of strings
*
* @return array|string
*/
public static function raw2htmlid($val) {
if(is_array($val)) {
foreach($val as $k => $v) {
$val[$k] = self::raw2htmlid($v);
}
return $val;
} else {
2013-05-26 01:01:46 +02:00
return trim(preg_replace(
'/_+/', '_', preg_replace('/[^a-zA-Z0-9\-_:.]+/','_', $val)),
'_'
);
}
}
2014-08-15 08:53:05 +02:00
/**
* Ensure that text is properly escaped for XML.
2014-08-15 08:53:05 +02:00
*
* @see http://www.w3.org/TR/REC-xml/#dt-escape
* @param array|string $val String to escape, or array of strings
* @return array|string
*/
public static function raw2xml($val) {
if(is_array($val)) {
foreach($val as $k => $v) $val[$k] = self::raw2xml($v);
return $val;
} else {
return htmlspecialchars($val, ENT_QUOTES, 'UTF-8');
}
}
2014-08-15 08:53:05 +02:00
/**
* Ensure that text is properly escaped for Javascript.
*
* @param array|string $val String to escape, or array of strings
* @return array|string
*/
public static function raw2js($val) {
if(is_array($val)) {
foreach($val as $k => $v) $val[$k] = self::raw2js($v);
return $val;
} else {
return str_replace(
// Intercepts some characters such as <, >, and & which can interfere
array("\\", '"', "\n", "\r", "'", "<", ">", "&"),
array("\\\\", '\"', '\n', '\r', "\\'", "\\x3c", "\\x3e", "\\x26"),
$val
);
}
}
/**
* Encode a value as a JSON encoded string.
*
* @param mixed $val Value to be encoded
* @return string JSON encoded string
*/
public static function raw2json($val) {
return json_encode($val);
}
/**
* Encode an array as a JSON encoded string.
* THis is an alias to {@link raw2json()}
*
* @param array $val Array to convert
* @return string JSON encoded string
*/
public static function array2json($val) {
return self::raw2json($val);
}
/**
* Safely encodes a value (or list of values) using the current database's
* safe string encoding method
2014-08-15 08:53:05 +02:00
*
* @param mixed|array $val Input value, or list of values as an array
* @param boolean $quoted Flag indicating whether the value should be safely
2014-08-15 08:53:05 +02:00
* quoted, instead of only being escaped. By default this function will
* only escape the string (false).
* @return string|array Safely encoded value in the same format as the input
*/
public static function raw2sql($val, $quoted = false) {
if(is_array($val)) {
foreach($val as $k => $v) {
$val[$k] = self::raw2sql($v, $quoted);
}
return $val;
} else {
if($quoted) {
return DB::get_conn()->quoteString($val);
} else {
return DB::get_conn()->escapeString($val);
}
}
}
2014-08-15 08:53:05 +02:00
/**
* Safely encodes a SQL symbolic identifier (or list of identifiers), such as a database,
* table, or column name. Supports encoding of multi identfiers separated by
* a delimiter (e.g. ".")
2014-08-15 08:53:05 +02:00
*
* @param string|array $identifier The identifier to escape. E.g. 'SiteTree.Title' or list of identifiers
* to be joined via the separator.
* @param string $separator The string that delimits subsequent identifiers
* @return string The escaped identifier. E.g. '"SiteTree"."Title"'
*/
public static function symbol2sql($identifier, $separator = '.') {
return DB::get_conn()->escapeIdentifier($identifier, $separator);
}
/**
* Convert XML to raw text.
* @uses html2raw()
* @todo Currently &#xxx; entries are stripped; they should be converted
* @param mixed $val
* @return array|string
*/
public static function xml2raw($val) {
if(is_array($val)) {
foreach($val as $k => $v) $val[$k] = self::xml2raw($v);
return $val;
} else {
// More complex text needs to use html2raw instead
if(strpos($val,'<') !== false) return self::html2raw($val);
else return html_entity_decode($val, ENT_QUOTES, 'UTF-8');
}
}
/**
* Convert a JSON encoded string into an object.
*
* @param string $val
* @return object|boolean
*/
public static function json2obj($val) {
return json_decode($val);
}
/**
* Convert a JSON string into an array.
2014-08-15 08:53:05 +02:00
*
* @uses json2obj
* @param string $val JSON string to convert
* @return array|boolean
*/
public static function json2array($val) {
return json_decode($val, true);
}
2014-08-15 08:53:05 +02:00
/**
* Converts an XML string to a PHP array
2015-03-20 05:21:59 +01:00
* See http://phpsecurity.readthedocs.org/en/latest/Injection-Attacks.html#xml-external-entity-injection
*
2011-10-07 22:57:04 +02:00
* @uses recursiveXMLToArray()
2015-03-20 05:21:59 +01:00
* @param string $val
* @param boolean $disableDoctypes Disables the use of DOCTYPE, and will trigger an error if encountered.
* false by default.
* @param boolean $disableExternals Disables the loading of external entities. false by default.
* @return array
* @throws Exception
*/
2015-03-20 05:21:59 +01:00
public static function xml2array($val, $disableDoctypes = false, $disableExternals = false) {
// Check doctype
if($disableDoctypes && preg_match('/\<\!DOCTYPE.+]\>/', $val)) {
throw new InvalidArgumentException('XML Doctype parsing disabled');
}
// Disable external entity loading
$oldVal = null;
2015-03-20 05:21:59 +01:00
if($disableExternals) $oldVal = libxml_disable_entity_loader($disableExternals);
try {
$xml = new SimpleXMLElement($val);
$result = self::recursiveXMLToArray($xml);
} catch(Exception $ex) {
if($disableExternals) libxml_disable_entity_loader($oldVal);
throw $ex;
}
if($disableExternals) libxml_disable_entity_loader($oldVal);
return $result;
}
/**
2014-08-15 08:53:05 +02:00
* Convert a XML string to a PHP array recursively. Do not
* call this function directly, Please use {@link Convert::xml2array()}
2014-08-15 08:53:05 +02:00
*
* @param SimpleXMLElement
2014-08-15 08:53:05 +02:00
*
* @return mixed
*/
protected static function recursiveXMLToArray($xml) {
if($xml instanceof SimpleXMLElement) {
$attributes = $xml->attributes();
foreach($attributes as $k => $v) {
if($v) $a[$k] = (string) $v;
}
$x = $xml;
$xml = get_object_vars($xml);
}
if(is_array($xml)) {
if(count($xml) == 0) return (string) $x; // for CDATA
foreach($xml as $key => $value) {
$r[$key] = self::recursiveXMLToArray($value);
}
if(isset($a)) $r['@'] = $a; // Attributes
return $r;
}
2014-08-15 08:53:05 +02:00
return (string) $xml;
}
2014-08-15 08:53:05 +02:00
/**
* Create a link if the string is a valid URL
*
* @param string $string The string to linkify
* @return string A link to the URL if string is a URL
*/
public static function linkIfMatch($string) {
if( preg_match( '/^[a-z+]+\:\/\/[a-zA-Z0-9$-_.+?&=!*\'()%]+$/', $string ) )
return "<a style=\"white-space: nowrap\" href=\"$string\">$string</a>";
else
return $string;
}
2014-08-15 08:53:05 +02:00
/**
* Simple conversion of HTML to plaintext.
2014-08-15 08:53:05 +02:00
*
* @param string $data Input data
* @param bool $preserveLinks
* @param int $wordWrap
* @param array $config
* @return string
*/
public static function html2raw($data, $preserveLinks = false, $wordWrap = 0, $config = null) {
$defaultConfig = array(
'PreserveLinks' => false,
'ReplaceBoldAsterisk' => true,
'CompressWhitespace' => true,
'ReplaceImagesWithAlt' => true,
);
if(isset($config)) {
$config = array_merge($defaultConfig,$config);
} else {
$config = $defaultConfig;
}
$data = preg_replace("/<style([^A-Za-z0-9>][^>]*)?>.*?<\/style[^>]*>/is","", $data);
$data = preg_replace("/<script([^A-Za-z0-9>][^>]*)?>.*?<\/script[^>]*>/is","", $data);
if($config['ReplaceBoldAsterisk']) {
$data = preg_replace('%<(strong|b)( [^>]*)?>|</(strong|b)>%i','*',$data);
}
2014-08-15 08:53:05 +02:00
// Expand hyperlinks
if(!$preserveLinks && !$config['PreserveLinks']) {
$data = preg_replace_callback('/<a[^>]*href\s*=\s*"([^"]*)">(.*?)<\/a>/i', function($matches) {
return Convert::html2raw($matches[2]) . "[$matches[1]]";
}, $data);
$data = preg_replace_callback('/<a[^>]*href\s*=\s*([^ ]*)>(.*?)<\/a>/i', function($matches) {
return Convert::html2raw($matches[2]) . "[$matches[1]]";
}, $data);
}
2014-08-15 08:53:05 +02:00
// Replace images with their alt tags
if($config['ReplaceImagesWithAlt']) {
2012-02-27 22:14:02 +01:00
$data = preg_replace('/<img[^>]*alt *= *"([^"]*)"[^>]*>/i', ' \\1 ', $data);
$data = preg_replace('/<img[^>]*alt *= *([^ ]*)[^>]*>/i', ' \\1 ', $data);
}
2014-08-15 08:53:05 +02:00
// Compress whitespace
if($config['CompressWhitespace']) {
2012-02-27 22:14:02 +01:00
$data = preg_replace("/\s+/", " ", $data);
}
2014-08-15 08:53:05 +02:00
// Parse newline tags
2012-02-27 22:14:02 +01:00
$data = preg_replace("/\s*<[Hh][1-6]([^A-Za-z0-9>][^>]*)?> */", "\n\n", $data);
$data = preg_replace("/\s*<[Pp]([^A-Za-z0-9>][^>]*)?> */", "\n\n", $data);
$data = preg_replace("/\s*<[Dd][Ii][Vv]([^A-Za-z0-9>][^>]*)?> */", "\n\n", $data);
$data = preg_replace("/\n\n\n+/", "\n\n", $data);
$data = preg_replace("/<[Bb][Rr]([^A-Za-z0-9>][^>]*)?> */", "\n", $data);
$data = preg_replace("/<[Tt][Rr]([^A-Za-z0-9>][^>]*)?> */", "\n", $data);
2012-02-28 00:44:27 +01:00
$data = preg_replace("/<\/[Tt][Dd]([^A-Za-z0-9>][^>]*)?> */", " ", $data);
$data = preg_replace('/<\/p>/i', "\n\n", $data );
2014-08-15 08:53:05 +02:00
// Replace HTML entities
$data = html_entity_decode($data, ENT_QUOTES, 'UTF-8');
// Remove all tags (but optionally keep links)
2014-08-15 08:53:05 +02:00
// strip_tags seemed to be restricting the length of the output
// arbitrarily. This essentially does the same thing.
if(!$preserveLinks && !$config['PreserveLinks']) {
$data = preg_replace('/<\/?[^>]*>/','', $data);
} else {
$data = strip_tags($data, '<a>');
}
// Wrap
if($wordWrap) $data = wordwrap(trim($data), $wordWrap);
return trim($data);
}
2012-02-27 22:14:02 +01:00
/**
* There are no real specifications on correctly encoding mailto-links,
* but this seems to be compatible with most of the user-agents.
* Does nearly the same as rawurlencode().
* Please only encode the values, not the whole url, e.g.
* "mailto:test@test.com?subject=" . Convert::raw2mailto($subject)
2014-08-15 08:53:05 +02:00
*
* @param $data string
* @return string
* @see http://www.ietf.org/rfc/rfc1738.txt
*/
public static function raw2mailto($data) {
return str_ireplace(
array("\n",'?','=',' ','(',')','&','@','"','\'',';'),
array('%0A','%3F','%3D','%20','%28','%29','%26','%40','%22','%27','%3B'),
$data
);
}
2014-08-15 08:53:05 +02:00
2011-02-21 22:53:58 +01:00
/**
* Convert a string (normally a title) to a string suitable for using in
* urls and other html attributes. Uses {@link URLSegmentFilter}.
2011-02-21 22:53:58 +01:00
*
2014-08-15 08:53:05 +02:00
* @param string
2011-02-21 22:53:58 +01:00
* @return string
*/
public static function raw2url($title) {
$f = URLSegmentFilter::create();
return $f->filter($title);
2011-02-21 22:53:58 +01:00
}
2014-08-15 08:53:05 +02:00
/**
* Normalises newline sequences to conform to (an) OS specific format.
2014-08-15 08:53:05 +02:00
*
* @param string $data Text containing potentially mixed formats of newline
* sequences including \r, \r\n, \n, or unicode newline characters
* @param string $nl The newline sequence to normalise to. Defaults to that
* specified by the current OS
* @return string
*/
public static function nl2os($data, $nl = PHP_EOL) {
return preg_replace('~\R~u', $nl, $data);
}
/**
* Encode a value into a string that can be used as part of a filename.
* All string data must be UTF-8 encoded.
*
* @param mixed $val Value to be encoded
* @return string
*/
public static function base64url_encode($val) {
return rtrim(strtr(base64_encode(json_encode($val)), '+/', '~_'), '=');
}
/**
* Decode a value that was encoded with Convert::base64url_encode.
*
* @param string $val Value to be decoded
* @return mixed Original value
*/
public static function base64url_decode($val) {
return json_decode(
base64_decode(str_pad(strtr($val, '~_', '+/'), strlen($val) % 4, '=', STR_PAD_RIGHT)),
true
);
}
/**
* Converts upper camel case names to lower camel case,
* with leading upper case characters replaced with lower case.
* Tries to retain word case.
*
* Examples:
* - ID => id
* - IDField => idField
* - iDField => iDField
*
* @param $str
* @return string
*/
public static function upperCamelToLowerCamel($str) {
$return = null;
$matches = null;
if(preg_match('/(^[A-Z]{1,})([A-Z]{1})([a-z]+.*)/', $str, $matches)) {
// If string has trailing lowercase after more than one leading uppercase characters,
// match everything but the last leading uppercase character.
$return = implode('', [
strtolower($matches[1]),
$matches[2],
$matches[3]
]);
} else if(preg_match('/(^[A-Z]{1})([a-z]+.*)/', $str, $matches)) {
// If string has trailing lowercase after exactly one leading uppercase characters,
// match everything but the last leading uppercase character.
$return = implode('', [
strtolower($matches[1]),
$matches[2]
]);
} elseif(preg_match('/^[A-Z]+$/', $str)) {
// If string has leading uppercase without trailing lowercase,
// just lowerase the whole thing.
$return = strtolower($str);
} else {
// If string has no leading uppercase, just return.
$return = $str;
}
return $return;
}
2012-02-27 22:14:02 +01:00
}