## Definition

- plus sign
- minus sign
- digits 0 through 9
- decimal point

Numeric literals are also referred to as numeric constants.

## Types of Numeric Literals

- Integers (see Integer Literals and Hexadecimal Integer Literals)
- Decimals (see Decimal Literals)
- Floating point numbers (see Floating Point Literals)

## Examples: Valid Numeric Literals

The following are examples of valid numeric literals.

Type | Examples | ||
---|---|---|---|

BYTEINT | 127 | -36 | -128 |

SMALLINT | 32767 | -12000 | -32768 |

INTEGER | 32768 | -60400 | 2147483647 |

DECIMAL | 0.0 | -23554367273149967931. | 2147483650 |

FLOATING POINT | 1E1 | 1.4E6 | 18E-3 |

## Examples: Nonvalid Numeric Literals

The following are examples of nonvalid numeric literals.

This literal is not valid … | Because it contains … |
---|---|

123456789012345678901234567890123456789 | more than 38 digits. |

$20,000.00 | a dollar sign and a comma. |

-38.7E2945 | four digits following the E. |

## Determining the Data Type of a Numeric Literal

The data type of a numeric literal is determined by the range of the literal value. The type used is the smallest that can contain the value.

For example, the data type of the numeric literal 127 is BYTEINT because it is the smallest type that can fit the value 127.

For decimal literals, the total number of digits determine the precision and the number of digits to the right of the decimal point determine the scale.

FOR details on … | SEE … |
---|---|

the range of values of integer literals | Integer Literals. |

determining the scale and precision of a decimal literal | Decimal Literals. |

## Implicit Conversion of Numeric Literals

Depending on the kind of operation performed or the type of column in which a value is to be stored, SQL may convert numeric literal constants from one numeric data type to another.

If a literal is outside the range for its required type, an error is reported during conversion.

For more information on numeric to numeric data type conversions, see Teradata Vantage™ SQL Functions, Expressions, and Predicates, B035-1145.

## Converting Character Strings to Numeric Literals

Character data can contain a string that is intended to be interpreted as a numeric value (for example, '15'). In such cases, Teradata Database attempts to convert the string to a numeric value whenever the context makes such a conversion necessary. If the character string does not represent a valid numeric value, an error is reported.

When data in a character column must be compared with data in a numeric column, the character data and the numeric data are converted to FLOAT before the comparison is made. Note that this can result in repeated conversions during data access.

If numeric data is defined and stored as a character string in a character column, and a SELECT operation uses a full table scan to compare the character column with a numeric literal, then the character column is converted to numeric in every row of the table.

For more information on character to numeric data type conversions, see Teradata Vantage™ SQL Functions, Expressions, and Predicates, B035-1145.