Data::Types - Validate and convert data types.

  use Data::Types qw(:all);

  my $whole = 4.5;
  $whole = to_whole($whole) unless is_whole($whole);

  my $int = 1.2;
  $int = to_int($int) unless is_int($int);

  my $decimal = '1.2foo';
  $decimal = to_decimal($decimal) unless is_decimal($decimal);

  my $real = '1.2foo';
  $real = to_real($real) unless is_real($real);

  my $float = '1.2foo';
  $float = to_float($float) unless is_float($float);

  my $string = [];
  $string = to_string($string) unless is_string($string);

This module exports a number of functions that are useful for validating and converting data types. It is intended for use in applications where data types are more important than they typically are in Perl e.g., database applications.

No functions are exported by default, though each function may be exported explicitly (see Functions, below, for a list of functions available for export). The following export tags are supported:

:whole

Exports is_whole() and to_whole().

:int

Exports is_int() and to_int().

:decimal

Exports is_decimal() and to_decimal().

:real

Exports is_real() and to_real().

:float

Exports is_float() and to_float().

:string

Exports is_string() and to_string().

:is

Exports all validation functions: is_whole(), is_int(), is_real(), is_decimal(), is_float(), and is_string().

:to

Exports all conversion functions: to_whole(), to_int(), to_real(), to_decimal(), to_float(), and to_string().

:all

Exports all functions.

  my $bool = is_whole($val);

Returns true if CW$val is a whole number (exclusive of 0), and false if it is not. The regular expression used to test the wholeness of CW$val is CW/^\d+$/.

  my $bool = is_whole(1); # Returns true.
  $bool = is_whole(-1);   # Returns false.
  $bool = is_whole(0);    # Returns false.

  my $whole = to_whole($val);

Converts CW$val to a whole number and returns it. Numbers will be rounded to the nearest whole. Note that since 0 (zero) is not considered a whole number by this module, it will not be returned. If CW$val is a mixture of numbers and letters, to_whole() will extract the first decimal number it finds and convert that number to a whole number.

  my $whole = to_whole(1);     # Returns 10.
  $whole = to_whole(0);         # Returns undef.
  $whole = to_whole(.22);       # Returns undef (rounded down to 0).
  $whole = to_whole(-2);        # Returns undef.
  $whole = to_whole('foo3.56'); # Returns 4.
  $whole = to_whole('foo');     # Returns undef.

  my $bool = is_int($val);

Returns true if CW$val is an integer, and false if it is not. Numbers may be preceded by a plus or minus sign. The regular expression used to test for an integer in CW$val is CW/^[+-]?\d+$/.

  my $bool = is_int(0); # Returns true.
  $bool = is_int(2);   # Returns true.
  $bool = is_int(-22);  # Returns false.
  $bool = is_int(3.2);  # Returns false.

  my $int = to_int($val);

Converts CW$val to an integer. If CW$val is a decimal number, it will be rounded to the nearest integer. If CW$val is a mixture of numbers and letters, to_int() will extract the first decimal number it finds and convert that number to an integer.

  my $int = to_int(10.5);  # Returns 10.
  $int = to_int(10.51);    # Returns 11.
  $int = to_int(-0.22);    # Returns 0.
  $int = to_int(-6.51);    # Returns 7.
  $int = to_int('foo');    # Returns undef.

  my $bool = is_decimal($val);

Returns true if CW$val is a decimal number, and false if it is not. Numbers may be preceded by a plus or minus sign. The regular expression used to test CW$val is CW/^[+-]?(?:\d+(?:\.\d*)?|\.\d+)$/.

  my $bool = is_decimal(1)    # Returns true.
  $bool = is_decimal(10.8)     # Returns true.
  $bool = is_decimal(-33.48)   # Returns true.
  $bool = is_decimal((1.23e99) # Returns false.

  my $dec = to_decimal($val);
  $dec = to_decimal($val, $precision);

Converts CW$val to a decimal number. The optional second argument sets the precision of the number. The default precision is 5. If CW$val is a mixture of numbers and letters, to_decimal() will extract the first decimal number it finds.

  my $dec = to_decimal(0);         # Returns 0.00000.
  $dec = to_decimal(10.5);         # Returns 10.5.
  $dec = to_decimal(10.500009);    # Returns 10.50001.
  $dec = to_decimal(10.500009, 7); # Returns 10.5000090.
  $dec = to_decimal('foo10.3')     # Returns 10.30000.
  $dec = to_decimal('foo-4.9')     # Returns -4.90000.
  $dec = to_decimal('foo')         # Returns undef.

  my $bool = is_real($val);

Returns true if CW$val is a real number, and false if it is not.

Note: This function is currently equivalent to is_decimal(), since this module cannot identify non-decimal real numbers (e.g., irrational numbers). This implementation may change in the future.

  my $real = to_real($val);
  $real = to_real($val, $precision);

Converts CW$val to a real number.

Note: Currently, this function is the equivalent of to_decimal(), since this module cannot identify non-decimal real numbers (e.g., irrational numbers). This implementation may change in the future.

  my $bool = is_real($val);

Returns true if CW$val is a float, and false if it is not. The regular expression used to test CW$val is CW/^([+-]?)(?=\d|\.\d)\d*(\.\d*)?([Ee]([+-]?\d+))?$/.

  my $bool = is_real(3);   # Returns true.
  $bool = is_real(1.23e99); # Returns true.
  $bool = is_real('foo');   # Returns false.

  my $dec = to_float($val);
  $dec = to_float($val, $precision);

Converts CW$val to a float. The optional second argument sets the precision of the number. The default precision is 5. If CW$val is a mixture of numbers and letters, to_float() will extract the first float it finds.

  my $float = to_float(1.23);          # Returns 1.23000.
  $float = to_float(1.23e99);          # Returns 1.23000e+99.
  $float = to_float(1.23e99, 1);       # Returns 1.2e+99.
  $float = to_float('foo-1.23');       # Returns -1.23000.
  $float = to_float('ick_1.23e99foo'); # Returns 1.23000e+99.

  my $bool = is_string($val);

Returns true if CW$val is a string, and false if it is not. All defined non-references are considered strings.

  my $bool = is_string('foo'); # Returns true.
  $bool = is_string(2);    # Returns true.
  $bool = is_string([]);       # Returns false.
  $bool = is_string(undef);    # Returns false.

  my $string = to_string($val);
  $string = to_string($val, $length);

Converts CW$val into a string. If CW$val is a reference, the string value of the reference will be returned. Such a value may be a memory address, or some other value, if the stringification operator has been overridden for the object stored in CW$val. If the optional second argument CW$length is passed, to_string() will truncate the string to that length. If CW$length is 0 (zero), it will not limit the length of the return string. If CW$val is undefined, to_string() will return undef.

  my $string = to_string('foo');   # Returns 'foo'.
  $string = to_string([]);         # Returns 'ARRAY(0)'.
  $string = to_string(undef);      # Returns undef.
  $string = to_string('hello', 4); # Returns 'hell'.

Hopefully none. Contact the author if you discover any.

David Wheeler <david@wheeler.net>

perlfaq4 lists the most of the regular expressions used to identify the different numeric types used in this module.

String::Checker also does some data type validation.

String::Scanf reimplements the C CWsscanf() function in perl, and also does data type validation and conversion.

Regexp::Common contains many useful common regular expressions (surprise!), including some that can be used to identify data types.

Arthur Bergman's types pragma, offers compile-time data types for Perl 5.8.0. The data types include int, float, and string. I highly recommend using this prgrma for fast, static data types.

Copyright (c) 2002-2003, David Wheeler. All Rights Reserved.

This module is free software; you can redistribute it and/or modify it under the same terms as Perl itself.