Linter.php 5.1 KB
Newer Older
Hamza Arfaoui's avatar
Hamza Arfaoui committed
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179
<?php
/* vim: set expandtab sw=4 ts=4 sts=4: */
/**
 * Analyzes a query and gives user feedback.
 *
 * @package PhpMyAdmin
 */
namespace PhpMyAdmin;

use PhpMyAdmin\SqlParser\Lexer;
use PhpMyAdmin\SqlParser\Parser;
use PhpMyAdmin\SqlParser\UtfString;
use PhpMyAdmin\SqlParser\Utils\Error as ParserError;

/**
 * The linter itself.
 *
 * @package PhpMyAdmin
 */
class Linter
{
    /**
     * Gets the starting position of each line.
     *
     * @param string $str String to be analyzed.
     *
     * @return array
     */
    public static function getLines($str)
    {
        if ((!($str instanceof UtfString))
            && (defined('USE_UTF_STRINGS')) && (USE_UTF_STRINGS)
        ) {
            // If the lexer uses UtfString for processing then the position will
            // represent the position of the character and not the position of
            // the byte.
            $str = new UtfString($str);
        }

        // The reason for using the strlen is that the length
        // required is the length in bytes, not characters.
        //
        // Given the following string: `????+`, where `?` represents a
        // multi-byte character (lets assume that every `?` is a 2-byte
        // character) and `+` is a newline, the first value of `$i` is `0`
        // and the last one is `4` (because there are 5 characters). Bytes
        // `$str[0]` and `$str[1]` are the first character, `$str[2]` and
        // `$str[3]` are the second one and `$str[4]` is going to be the
        // first byte of the third character. The fourth and the last one
        // (which is actually a new line) aren't going to be processed at
        // all.
        $len = ($str instanceof UtfString) ?
            $str->length() : strlen($str);

        $lines = array(0);
        for ($i = 0; $i < $len; ++$i) {
            if ($str[$i] === "\n") {
                $lines[] = $i + 1;
            }
        }
        return $lines;
    }

    /**
     * Computes the number of the line and column given an absolute position.
     *
     * @param array $lines The starting position of each line.
     * @param int   $pos   The absolute position
     *
     * @return array
     */
    public static function findLineNumberAndColumn(array $lines, $pos)
    {
        $line = 0;
        foreach ($lines as $lineNo => $lineStart) {
            if ($lineStart > $pos) {
                break;
            }
            $line = $lineNo;
        }
        return array($line, $pos - $lines[$line]);
    }

    /**
     * Runs the linting process.
     *
     * @param string $query The query to be checked.
     *
     * @return array
     */
    public static function lint($query)
    {
        // Disabling lint for huge queries to save some resources.
        if (mb_strlen($query) > 10000) {
            return array(
                array(
                    'message' => __(
                        'Linting is disabled for this query because it exceeds the '
                        . 'maximum length.'
                    ),
                    'fromLine' => 0,
                    'fromColumn' => 0,
                    'toLine' => 0,
                    'toColumn' => 0,
                    'severity' => 'warning',
                )
            );
        }

        /**
         * Lexer used for tokenizing the query.
         *
         * @var Lexer
         */
        $lexer = new Lexer($query);

        /**
         * Parsed used for analysing the query.
         *
         * @var Parser
         */
        $parser = new Parser($lexer->list);

        /**
         * Array containing all errors.
         *
         * @var array
         */
        $errors = ParserError::get(array($lexer, $parser));

        /**
         * The response containing of all errors.
         *
         * @var array
         */
        $response = array();

        /**
         * The starting position for each line.
         *
         * CodeMirror requires relative position to line, but the parser stores
         * only the absolute position of the character in string.
         *
         * @var array
         */
        $lines = static::getLines($query);

        // Building the response.
        foreach ($errors as $idx => $error) {

            // Starting position of the string that caused the error.
            list($fromLine, $fromColumn) = static::findLineNumberAndColumn(
                $lines, $error[3]
            );

            // Ending position of the string that caused the error.
            list($toLine, $toColumn) = static::findLineNumberAndColumn(
                $lines, $error[3] + mb_strlen($error[2])
            );

            // Building the response.
            $response[] = array(
                'message' => sprintf(
                    __('%1$s (near <code>%2$s</code>)'),
                    htmlspecialchars($error[0]), htmlspecialchars($error[2])
                ),
                'fromLine' => $fromLine,
                'fromColumn' => $fromColumn,
                'toLine' => $toLine,
                'toColumn' => $toColumn,
                'severity' => 'error',
            );
        }

        // Sending back the answer.
        return $response;
    }

}