Fix for #2613, when using php < 5.4 there was an unnecessary backslash before each utf-8 char. The problem was that the regexp matched all utf-8 encoded chars (included the ones that where escaped). The new regexp uses the lookbehind feature to check if the backslash isn't prefixed with an other backslash.
Sandy Pleyte
11 years ago
parent
7a902ed96d
commit
f92f2f45a9
@ -0,0 +1,123 @@
|
||||
<?php
|
||||
|
||||
/*
|
||||
* This file is part of Composer.
|
||||
*
|
||||
* (c) Nils Adermann <naderman@naderman.de>
|
||||
* Jordi Boggiano <j.boggiano@seld.be>
|
||||
*
|
||||
* For the full copyright and license information, please view the LICENSE
|
||||
* file that was distributed with this source code.
|
||||
*/
|
||||
|
||||
namespace Composer\Json;
|
||||
|
||||
/**
|
||||
* Formats json strings used for php < 5.4 because the json_encode doesn't
|
||||
* supports the flags JSON_UNESCAPED_SLASHES | JSON_PRETTY_PRINT | JSON_UNESCAPED_UNICODE
|
||||
* in these versions
|
||||
*
|
||||
* @author Konstantin Kudryashiv <ever.zet@gmail.com>
|
||||
* @author Jordi Boggiano <j.boggiano@seld.be>
|
||||
*/
|
||||
class JsonFormatter
|
||||
{
|
||||
/**
|
||||
*
|
||||
* This code is based on the function found at:
|
||||
* http://recursive-design.com/blog/2008/03/11/format-json-with-php/
|
||||
*
|
||||
* Originally licensed under MIT by Dave Perrett <mail@recursive-design.com>
|
||||
*
|
||||
*
|
||||
* @param string $json
|
||||
* @param bool $unescapeUnicode Un escape unicode
|
||||
* @param bool $unescapeSlashes Un escape slashes
|
||||
* @return string
|
||||
*/
|
||||
public static function format($json, $unescapeUnicode, $unescapeSlashes)
|
||||
{
|
||||
$result = '';
|
||||
$pos = 0;
|
||||
$strLen = strlen($json);
|
||||
$indentStr = ' ';
|
||||
$newLine = "\n";
|
||||
$outOfQuotes = true;
|
||||
$buffer = '';
|
||||
$noescape = true;
|
||||
|
||||
for ($i = 0; $i < $strLen; $i++) {
|
||||
// Grab the next character in the string
|
||||
$char = substr($json, $i, 1);
|
||||
|
||||
// Are we inside a quoted string?
|
||||
if ('"' === $char && $noescape) {
|
||||
$outOfQuotes = !$outOfQuotes;
|
||||
}
|
||||
|
||||
if (!$outOfQuotes) {
|
||||
$buffer .= $char;
|
||||
$noescape = '\\' === $char ? !$noescape : true;
|
||||
continue;
|
||||
} elseif ('' !== $buffer) {
|
||||
if ($unescapeSlashes) {
|
||||
$buffer = str_replace('\\/', '/', $buffer);
|
||||
}
|
||||
|
||||
if ($unescapeUnicode && function_exists('mb_convert_encoding')) {
|
||||
// http://stackoverflow.com/questions/2934563/how-to-decode-unicode-escape-sequences-like-u00ed-to-proper-utf-8-encoded-cha
|
||||
$buffer = preg_replace_callback('/(\\\\+)u([0-9a-f]{4})/i', function($match) {
|
||||
$l = strlen($match[1]);
|
||||
|
||||
if ($l%2)
|
||||
return str_repeat ('\\', $l-1).mb_convert_encoding(pack('H*', $match[2]), 'UTF-8', 'UCS-2BE');
|
||||
|
||||
return $match[0];
|
||||
}, $buffer);
|
||||
}
|
||||
|
||||
$result .= $buffer.$char;
|
||||
$buffer = '';
|
||||
continue;
|
||||
}
|
||||
|
||||
if (':' === $char) {
|
||||
// Add a space after the : character
|
||||
$char .= ' ';
|
||||
} elseif (('}' === $char || ']' === $char)) {
|
||||
$pos--;
|
||||
$prevChar = substr($json, $i - 1, 1);
|
||||
|
||||
if ('{' !== $prevChar && '[' !== $prevChar) {
|
||||
// If this character is the end of an element,
|
||||
// output a new line and indent the next line
|
||||
$result .= $newLine;
|
||||
for ($j = 0; $j < $pos; $j++) {
|
||||
$result .= $indentStr;
|
||||
}
|
||||
} else {
|
||||
// Collapse empty {} and []
|
||||
$result = rtrim($result)."\n\n".$indentStr;
|
||||
}
|
||||
}
|
||||
|
||||
$result .= $char;
|
||||
|
||||
// If the last character was the beginning of an element,
|
||||
// output a new line and indent the next line
|
||||
if (',' === $char || '{' === $char || '[' === $char) {
|
||||
$result .= $newLine;
|
||||
|
||||
if ('{' === $char || '[' === $char) {
|
||||
$pos++;
|
||||
}
|
||||
|
||||
for ($j = 0; $j < $pos; $j++) {
|
||||
$result .= $indentStr;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
return $result;
|
||||
}
|
||||
}
|
||||
@ -0,0 +1,45 @@
|
||||
<?php
|
||||
|
||||
/*
|
||||
* This file is part of Composer.
|
||||
*
|
||||
* (c) Nils Adermann <naderman@naderman.de>
|
||||
* Jordi Boggiano <j.boggiano@seld.be>
|
||||
*
|
||||
* For the full copyright and license information, please view the LICENSE
|
||||
* file that was distributed with this source code.
|
||||
*/
|
||||
|
||||
namespace Composer\Test\Json;
|
||||
|
||||
use Composer\Json\JsonFormatter;
|
||||
|
||||
class JsonFormatterTest extends \PHPUnit_Framework_TestCase
|
||||
{
|
||||
/**
|
||||
* Test if \u0119 (196+153) will get correctly formatted
|
||||
* See ticket #2613
|
||||
*/
|
||||
public function testUnicodeWithPrependedSlash()
|
||||
{
|
||||
$data = '"' . chr(92) . chr(92) . chr(92) . 'u0119"';
|
||||
$encodedData = JsonFormatter::format($data, true, true);
|
||||
$expected = '34+92+92+196+153+34';
|
||||
$this->assertEquals($expected, $this->getCharacterCodes($encodedData));
|
||||
}
|
||||
|
||||
/**
|
||||
* Convert string to character codes split by a plus sign
|
||||
* @param string $string
|
||||
* @return string
|
||||
*/
|
||||
protected function getCharacterCodes($string)
|
||||
{
|
||||
$codes = array();
|
||||
for ($i = 0; $i < strlen($string); $i++) {
|
||||
$codes[] = ord($string[$i]);
|
||||
}
|
||||
return implode('+', $codes);
|
||||
}
|
||||
|
||||
}
|
||||
Loading…
Reference in New Issue