SQL wildcards are special characters used in the LIKE operator to search for patterns in text-based columns. The two most common wildcards are the percent sign (%), which matches zero or more characters, and the underscore (_), which matches exactly one character. These wildcards are applied in WHERE clauses to filter rows based on flexible patterns rather than exact matches. For example, WHERE name LIKE 'A%' retrieves all names starting with “A,” while WHERE code LIKE 'ABC_123' matches codes like “ABCX123” or “ABC-123” where the underscore represents any single character in that position.
Wildcards are particularly useful for partial string matching. The % wildcard can be placed at the start, middle, or end of a pattern. For instance, LIKE '%error%' finds any occurrence of “error” in a column, regardless of its position. The _ wildcard is more precise, allowing you to account for variable single characters. A pattern like LIKE 'User_123' matches “UserA123” or “UserX123.” Some databases also support square brackets (e.g., [a-z]) to match a range of characters, such as LIKE 'File[0-9]' for “File1” to “File9,” though this syntax varies by database system (e.g., not supported in MySQL).
When using wildcards, performance and clarity are key considerations. Patterns starting with % (e.g., LIKE '%data') can lead to full table scans, slowing queries on large datasets. To mitigate this, combine wildcards with other filters (e.g., date ranges) or use full-text search for complex patterns. Escaping wildcards with an escape character (e.g., \) is necessary when searching for literal % or _ characters. For example, WHERE note LIKE '50\% off' ESCAPE '\' ensures the % is treated as a regular character. Always test patterns to avoid unintended matches, and use database-specific features like indexes for optimized wildcard searches.