The MySQL data-types table in Open Document format (much better for printing).

Type {storage} Name Range Attributes Default
Numeric
{1 byte}
TINYINT[(M)] -128 TO 127
[0 to 255 if UNSIGNED]
AUTO_INCREMENT
UNSIGNED, ZEROFILL,
SERIAL DEFAULT VALUE
NULL
[0 if NOT NULL]
Numeric
{2 bytes}
SMALLINT[(M)] -32,768 to 32,767
[0 to 65,535]
AUTO_INCREMENT,
UNSIGNED, ZEROFILL,
SERIAL DEFAULT VALUE
NULL
[0 if NOT NULL]
Numeric
{3 bytes}
MEDIUMINT[(M)] -8,388,608 to 8,388,607
[0 to 16,777,215]
AUTO_INCREMENT,
UNSIGNED, ZEROFILL,
SERIAL DEFAULT VALUE
NULL
[0 if NOT NULL]
Numeric
{4 bytes}
INT[(M)] -/+2.147E+9
[0 to 4.294E+9]
AUTO_INCREMENT,
UNSIGNED, ZEROFILL,
SERIAL DEFAULT VALUE
NULL
[0 if NOT NULL]
Numeric
{8 bytes}
BIGINT[(M)] -/+9.223E+18
[0 to 18.45E+18]
AUTO_INCREMENT,
UNSIGNED, ZEROFILL,
SERIAL DEFAULT VALUE
NULL
[0 if NOT NULL]
Numeric
{4 or 8}
FLOAT(p) p=0-24  --> "FLOAT"
p=25-53  --> "DOUBLE"
UNSIGNED, ZEROFILL NULL
[0 if NOT NULL]
Numeric
{4 bytes}
FLOAT[(M,D)] Min=+/-1.175E-38
Max=+/-3.403E+38
UNSIGNED, ZEROFILL NULL
[0 if NOT NULL]
Numeric
{8 bytes}
DOUBLE[(M,D)] Min=+/-2.225E-308
Max=+/-1.798E+308
UNSIGNED, ZEROFILL NULL
[0 if NOT NULL]
Numeric
{M+2}
DECIMAL[(M,[D])]
Stored as string
Max Range = DOUBLE range
Fixed point vs. DOUBLE float
UNSIGNED, ZEROFILL NULL
[0 if NOT NULL]
Bit
{8 bytes}
BIT[(M)] Binary. Display by [add zero or converting with BIN()]. M=1-64 Prior to 5.03
TINYINT(1) Synonym
NULL
[0 if NOT NULL]
String
{M char's}
CHAR[(M)] M=0-255 Characters, FIXED.
Right padded with spaces.
BINARY, CHARACTER SET NULL
["" if NOT NULL]
String
{M char's1}
VARCHAR(M) M=0-65,535 Characters
M=0-255 <v5.0.3
BINARY, CHARACTER SET NULL
["" if NOT NULL]
String
{#char's1}
TINYTEXT2 0-255 Characters BINARY, CHARACTER SET NULL
["" if NOT NULL]
String
{#char's1}
TEXT2 0-65,535 Char's BINARY, CHARACTER SET NULL
["" if NOT NULL]
String
{#char's1}
MEDIUMTEXT2 0-16,777,215 Char's BINARY, CHARACTER SET NULL
["" if NOT NULL]
String
{#char's1}
LONGTEXT2 0-4,294,967,295 Char's BINARY, CHARACTER SET NULL
["" if NOT NULL]
String
{M bytes}
BINARY[(M)] M=0-255 bytes, FIXED. Global Only
(case sensitive)
NULL
["" if NOT NULL]
String
{M bytes}
VARBINARY(M) 0-65,535 bytes
M=0-255 <v5.0.3
Global Only
(case sensitive)
NULL
["" if NOT NULL]
String
{#bytes1}
TINYBLOB 0-255 bytes Global Only
(case sensitive)
NULL
["" if NOT NULL]
String
{#bytes1}
BLOB 0-65,535 bytes Global Only
(case sensitive)
NULL
["" if NOT NULL]
String
{#bytes1}
MEDIUMBLOB 0-16,777,215 bytes Global Only
(case sensitive)
NULL
["" if NOT NULL]
String
{#bytes1}
LONGBLOB 0-4,294,967,295 bytes Global Only
(case sensitive)
NULL
["" if NOT NULL]
String
{1-2 bytes}
ENUM2
("A1","A2",...)
Column is exactly 1 of 1-65,535 values CHARACTER SET NULL [1st value if NOT NULL]
String
{1-8 bytes}
SET2
("A1","A2",...)
Column is 0 or more values in list of 1-64 members CHARACTER SET NULL
["" if NOT NULL]
Date & Time
{3 bytes}
DATE "1000-01-01" - "9999-12-31" Global Only
(YYYY-MM-DD)
NULL
["0000-00-00" if NOT NULL]
Date & Time
{8 bytes}
DATETIME "1000-01-01 00:00:00" -
"9999-12-31 23:59:59"
Global Only
(YYYY-MM-DD hh:mm:ss)
NULL ["0000-00-00 00:00:00"
if NOT NULL]
Date & Time
{3 bytes}
TIME "-838:59:59" - "838:59:59" Global Only
(hh:mm:ss)
NULL
["00:00:00" if NOT NULL]
Date & Time
{4 bytes}
TIMESTAMP 19700101000000 -
2037+
Global Only
(YYYYMMDDhhmmss)
Current Date & Time
Date & Time
{1 bytes}
YEAR 1900 - 2155 Global Only
(YYYY)
NULL
["0000" if NOT NULL]

Notes:

  • 1 Storage will be # of characters or bytes, plus byte(s) to record length.
  • 2 These String data types are NOT case sensitive, unless given the "binary" attribute or have a case-sensitive CHARACTER SET collation.
  • "E" is an abbreviation for "exponent". E18 means move the decimal over 18 places (search "scientific notation").
  • SERIAL DEFAULT VALUE attribute is an alias for "AUTO_INCREMENT NOT NULL UNIQUE".
  • SERIAL data type is a synonym for "BIGINT UNSIGNED AUTO_INCREMENT NOT NULL UNIQUE".
  • BOOL and BOOLEAN data types are synonyms for TINYINT(1).
  • REAL[(M,D)] and DOUBLE PRECISION[(M,D)] datatypes are synonyms for DOUBLE[(M,D)].
  • REAL_AS_FLOAT system variable can make REAL[(M,D)] a synonym for FLOAT[(M,D)].
  • "UNSIGNED ZEROFILL" attributes: ZEROFILL means if you specify an M value for an integer, it will be padded with zeros to fill up the M spaces. Ex: M=6, integer=247, display="000247". UNSIGNED means no negative values and often expands your range.
  • Corresponding non-binary and binary string types:
    • CHAR vs. BINARY
    • VARCHAR vs. VARBINARY
    • TEXT vs. BLOB

The latest MySQL documentation is at the MySQL Web Site, but I highly recommend getting the book "MySQL" by Paul DuBois for a very readable and thorough book on the subject. There is a sample database that you can download to follow along with the examples in the book. Good Stuff. These notes are from his book on the 4.1 and 5.0 versions of MySQL.