MySQL 5.0 Reference Manual  /  ...  /  YEAR(2) Limitations and Migrating to YEAR(4)

11.3.4 YEAR(2) Limitations and Migrating to YEAR(4)

This section describes problems that can occur when using YEAR(2) and provides information about converting existing YEAR(2) columns to YEAR(4).

Although the internal range of values for YEAR(4) and YEAR(2) is the same (1901 to 2155, and 0000), the display width for YEAR(2) makes that type inherently ambiguous because displayed values indicate only the last two digits of the internal values and omit the century digits. The result can be a loss of information under certain circumstances. For this reason, consider avoiding YEAR(2) throughout your applications and using YEAR(4) wherever you need a YEAR data type. Note that conversion will become necessary at some point because support for YEAR data types with display values other than 4, most notably YEAR(2), is reduced as of MySQL 5.6.6 and will be removed entirely in a future release.

YEAR(2) Limitations

Issues with the YEAR(2) data type include ambiguity of displayed values, and possible loss of information when values are dumped and reloaded or converted to strings.

  • Displayed YEAR(2) values can be ambiguous. It is possible for up to three YEAR(2) values that have different internal values to have the same displayed value, as the following example demonstrates:

    mysql> CREATE TABLE t (y2 YEAR(2), y4 YEAR(4));
    Query OK, 0 rows affected (0.01 sec)
    mysql> INSERT INTO t (y2) VALUES(1912),(2012),(2112);
    Query OK, 3 rows affected (0.00 sec)
    Records: 3  Duplicates: 0  Warnings: 0
    mysql> UPDATE t SET y4 = y2;
    Query OK, 3 rows affected (0.00 sec)
    Rows matched: 3  Changed: 3  Warnings: 0
    mysql> SELECT * FROM t;
    | y2   | y4   |
    |   12 | 1912 |
    |   12 | 2012 |
    |   12 | 2112 |
    3 rows in set (0.00 sec)
  • If you use mysqldump to dump the table created in the preceding item, the dump file represents all y2 values using the same 2-digit representation (12). If you reload the table from the dump file, all resulting rows have internal value 2012 and display value 12, thus losing the distinctions among them.

  • Conversion of a YEAR(2) or YEAR(4) data value to string form uses the display width of the YEAR type. Suppose that YEAR(2) and YEAR(4) columns both contain the value 1970. Assigning each column to a string results in a value of '70' or '1970', respectively. That is, loss of information occurs for conversion from YEAR(2) to string.

  • Values outside the range from 1970 to 2069 are stored incorrectly when inserted into a YEAR(2) column in a CSV table. For example, inserting 2111 results in a display value of 11 but an internal value of 2011.

To avoid these problems, use YEAR(4) rather than YEAR(2). Suggestions regarding migration strategies appear later in this section.

Migrating from YEAR(2) to YEAR(4)

To convert YEAR(2) columns to YEAR(4), use ALTER TABLE. Suppose that a table t1 has this definition:


Modify the column using ALTER TABLE as follows. Remember to include any column attributes such as NOT NULL or DEFAULT:


The ALTER TABLE statement converts the table without changing YEAR(2) values. If the server is a replication master, the ALTER TABLE statement replicates to slaves and makes the corresponding table change on each one.

One migration method should be avoided: Do not dump your data with mysqldump and reload the dump file after upgrading. This has the potential to change YEAR(2) values, as described previously.

A migration from YEAR(2) to YEAR(4) should also involve examining application code for the possibility of changed behavior under conditions such as these:

  • Code that expects selecting a YEAR column to produce exactly two digits.

  • Code that does not account for different handling for inserts of numeric 0: Inserting 0 into YEAR(2) or YEAR(4) results in an internal value of 2000 or 0000, respectively.

Download this Manual
User Comments
Sign Up Login You must be logged in to post a comment.