=pod

=head1 NAME

Locale::Codes::Types - types of data sets supported

=head1 DESCRIPTION

This document contains a description of different types of code sets
supported by the B<Locale-Codes> distribution.

The following types are supported:

=over 4

=item L</"country">

=item L</"language">

=item L</"currency">

=item L</"script">

=item L</"langfam">

=item L</"langvar">

=item L</"langext">

=back

Any time you have to specify the type of data, use one of the values from
this list.  When using the OO interface, you have to specify the type of
data you are working with.  For example:

   use Locale::Codes;
   ...
   $obj->type('country');
   $obj->type('langext');

When using the traditional interfaces, the functions all have the data type
included in the function name.  For example:

   use Locale::Codes::Country;
   code2country(...);

   use Locale::Codes::LangExt;
   code2langext(...);

Each type of data may have any number of code sets.  Code sets may be
specified by name.  Traditionally, a perl constant was exported and could
also be used to specify the code set.

Both methods are available for both the OO and traditional interfaces, so
whenever a function or method takes an argument specifying a code set, either
the name or a constant can be used.

In the lists below, a code set is specified in the list by including the
name and the constant.  So, for example, the first country code set is
named C<'alpha-2'> and has a perl constant C<LOCALE_COUNTRY_ALPHA_2>
associated with it.  When using the OO interface, the constants are only
available if you import them by loading the module with:

   use Locale::Codes ':constants';

The constants are always available when using the traditional interfaces.

Some of the older perl constants names were not consistent, and in those
cases, two constants are available (a newer consistent name and the older
inconsistent one).  Either may be used.

The default code set for each type is marked with an asterisk (*).

=head1 country

Code sets for identifying countries are maintained by several different
agencies and standards.

The following code sets are maintained in the ISO 3166 standard.
The official home page for the ISO 3166 maintenance agency is:
L<http://www.iso.org/iso/home/standards/country_codes.htm> .

Only the officially assigned codes are included.

=over 4

=item B<* alpha-2, LOCALE_COUNTRY_ALPHA_2, LOCALE_CODE_ALPHA_2>

This is the set of two-letter (lowercase) codes from ISO 3166-1, such
as 'tv' for Tuvalu.

=item B<alpha-3, LOCALE_COUNTRY_ALPHA_3, LOCALE_CODE_ALPHA_3>

This is the set of three-letter (lowercase) codes from ISO 3166-1,
such as 'brb' for Barbados. These codes are actually defined and
maintained by the U.N. Statistics division.

=item B<numeric, LOCALE_COUNTRY_NUMERIC, LOCALE_CODE_NUMERIC>

This is the set of three-digit numeric codes from ISO 3166-1, such as
064 for Bhutan.

If a 2-digit code is entered, it is converted to 3 digits by prepending
a 0.

=back

A list of domain names are maintained by the IANA (Internet Assigned
Numbers Authority).  These are available at:
L<http://www.iana.org/domains/root/db/> .  Only the actual country
codes are used, and the country names come from ISO 3166.

=over 4

=item B<dom, LOCALE_COUNTRY_DOM, LOCALE_CODE_DOM>

The country domains assigned by IANA are usually the two-letter
(lowercase) codes from ISO 3166, but there are a few other additions.

=back

The United Nations also maintains country lists.  Their list is also
similar, but not identical, to the ISO 3166 list.

The data is available here:
L<https://unstats.un.org/unsd/methodology/m49/>

Previously, this table was treated as a source of the ISO 3166 data,
but I found that the table was incomplete, so I stopped using it.
Later, it was added back in as it's own list of codes.

=over 4

=item B<un-alpha-3, LOCALE_COUNTRY_UN_ALPHA_3, LOCALE_CODE_UN_ALPHA_3>

This is similar to the 'alpha-3' set from ISO 3166, except that the
codes are uppercase.

=item B<un-numeric, LOCALE_COUNTRY_UN_NUMERIC, LOCALE_CODE_UN_NUMERIC>

This is similar to the 'numeric' set from ISO 3166.

=back

The US Government also keeps a list of codes.  Originally, it maintained
the FIPS-11 code set, but this was deprecated and replaced by the GENC code
set.  The FIPS-11 code sets are no longer supported by B<Locale-Codes>.

The GENC code sets are available here:
L<https://nsgreg.nga.mil/genc/discovery> .  They are also similar, but
not identical, to the ISO 3166 code sets.

=over 4

=item B<genc-alpha-2, LOCALE_COUNTRY_GENC_ALPHA_2, LOCALE_CODE_GENC_ALPHA_2>

Similar to the 'alpha-2' set, but uppercase.

=item B<genc-alpha-3, LOCALE_COUNTRY_GENC_ALPHA_3, LOCALE_CODE_GENC_ALPHA_3>

Similar to the 'alpha-3' set, but uppercase.

=item B<genc-numeric, LOCALE_COUNTRY_GENC_NUMERIC, LOCALE_CODE_GENC_NUMERIC>

Similar to the 'numeric' set.

=back

There are other sources of codes that are not currently used in this
distribution.

ISO codes for country sub-divisions (states, counties, provinces,
etc), as defined in ISO 3166-2.  This module is not part of the
Locale-Codes distribution, but is available from CPAN in
CPAN/modules/by-module/Locale/

The World Factbook maintained by the CIA is a potential source of
the data.  Unfortunately, it adds/preserves non-standard codes, so it is
not used as a source of data.
L<https://www.cia.gov/library/publications/the-world-factbook/appendix/appendix-d.html>

Another unofficial source of data is the Statoids web site:
L<http://www.statoids.com/wab.html> . Currently, it is not used to get
data, but the notes and explanatory material were very useful for
understanding discrepancies between the sources.

=head1 language

Code sets for identifying languages come from a couple different locations.

The primary source is ISO 639 .  The ISO 639-2 codes are available here:
L<http://www.loc.gov/standards/iso639-2/>
and the ISO 639-5 codes are available here:
L<http://www.loc.gov/standards/iso639-5/> .

In addition, the IANA maintains a language registry which are added to the ISO
lists.  Because it is intended to supplement the ISO standard, the IANA list is
not separate.

The IANA data is available here:
L<http://www.iana.org/assignments/language-subtag-registry>

The code sets are:

=over 4

=item B<* alpha-2, LOCALE_LANGUAGE_ALPHA_2, LOCALE_LANG_ALPHA_2>

This is the set of two-letter (lowercase) codes from ISO 639-1, such
as 'he' for Hebrew.  It also includes additions to this set included
in the IANA language registry.

=item B<alpha-3, LOCALE_LANGUAGE_ALPHA_3, LOCALE_LANG_ALPHA_3>

This is the set of three-letter (lowercase) bibliographic codes from
ISO 639-2 and 639-5, such as 'heb' for Hebrew.  It also includes
additions to this set included in the IANA language registry.

=item B<term, LOCALE_LANGUAGE_TERM, LOCALE_LANG_TERM>

This is the set of three-letter (lowercase) terminologic codes from
ISO 639.

=back

=head1 currency

The source of currency codes is the ISO 4217 data available here:
L<http://www.iso.org/iso/support/currency_codes_list-1.htm>

The code sets are:

=over 4

=item B<* alpha, LOCALE_CURRENCY_ALPHA, LOCALE_CURR_ALPHA>

This is a set of three-letter (uppercase) codes from ISO 4217 such
as EUR for Euro.

Two of the codes specified by the standard (XTS which is reserved
for testing purposes and XXX which is for transactions where no
currency is involved) are omitted.

=item B<num, LOCALE_CURRENCY_NUMERIC, LOCALE_CURR_NUMERIC>

This is the set of three-digit numeric codes from ISO 4217.

=back

=head1 script

The source of script code sets is ISO 15924 available here:
L<http://www.unicode.org/iso15924/>

Additional data comes from the IANA language subtag registry:
L<http://www.iana.org/assignments/language-subtag-registry> .

Code sets are:

=over 4

=item B<* alpha, LOCALE_SCRIPT_ALPHA>

This is a set of four-letter (capitalized) codes from ISO 15924
such as 'Phnx' for Phoenician.  It also includes additions to this
set included in the IANA language registry.

The Zxxx, Zyyy, and Zzzz codes are not used.

=item B<num, LOCALE_SCRIPT_NUMERIC>

This is a set of three-digit numeric codes from ISO 15924 such as 115
for Phoenician.

=back

=head1 langfam

Language families are specified using codes from ISO 639-5 available here:
L<http://www.loc.gov/standards/iso639-5/id.php>

Code sets are:

=over 4

=item B<* alpha, LOCALE_LANGFAM_ALPHA>

This is the set of three-letter (lowercase) codes from ISO 639-5
such as 'apa' for Apache languages.

=back

=head1 langvar

Language variations are specified using codes from he IANA language
subtag registry available here:
L<http://www.iana.org/assignments/language-subtag-registry>

Code sets are:

=over 4

=item B<* alpha, LOCALE_LANGVAR_ALPHA>

This is the set of alphanumeric codes from the IANA
language registry, such as 'arevela' for Eastern Armenian.

=back

=head1 langext

Language extensions are specified using codes from he IANA language
subtag registry available here:
L<http://www.iana.org/assignments/language-subtag-registry>

Code sets are:

=over 4

=item B<* alpha, LOCALE_LANGEXT_ALPHA>

This is the set of three-letter (lowercase) codes from the IANA
language registry, such as 'acm' for Mesopotamian Arabic.

=back

=head1 NEW CODE SETS

I'm always open to suggestions for new code sets.

In order for me to add a code set, I want the following criteria
to be met:

=over 4

=item B<General-use code set>

If a code set is not general use, I'm not likely to spend the time
to add and support it.

=item B<An official source of data>

I require an official (or at least, a NEARLY official) source where I
can get the data on a regular basis.

Ideally, I'd only get data from an official source, but sometimes that
is not possible. For example the ISO standards are not typically
available for free, so I may have to get some of that data from
alternate sources that I'm confident are getting their data from the
official source.  However, I will always be hesitant to accept a
non-official source.

As an example, I used to get some country data from the CIA World
Factbook. Given the nature of the source, I'm sure they're updating
data from the official sources and I consider it "nearly" official.
However, even in this case, I found that they were adding codes that
were not part of the standard, so I have stopped using them as a
source.

There are many 3rd party sites which maintain lists (many of which are
actually in a more convenient form than the official sites).
Unfortunately, I will reject most of them since I have no feel for how
"official" they are.

=item B<A free source of the data>

Obviously, the data must be free-of-charge. I'm not interested in
paying for the data (and I'm not interested in the overhead of having
someone else pay for the data for me).

=item B<A reliable source of data>

The source of data must come from a source that I can reasonably expect
to exist for the foreseeable future since I will be extremely reluctant
to drop support for a data set once it's included.

I am also reluctant to accept data sent to me by an individual.
Although I appreciate the offer, it is simply not practical to consider
an individual contribution as a reliable source of data. The source
should be an official agency of some sort.

=back

These requirements are open to discussion. If you have a code set
you'd like to see added, but which may not meet all of the above
requirements, feel free to email me and we'll discuss it.  Depending
on circumstances, I may be willing to waive some of these criteria.

=head1 SEE ALSO

=over 4

=item L<Locale::Codes>

The Locale-Codes distribution.

=back

=head1 AUTHOR

See Locale::Codes for full author history.

Currently maintained by Sullivan Beck (sbeck@cpan.org).

=head1 COPYRIGHT

   Copyright (c) 1997-2001 Canon Research Centre Europe (CRE).
   Copyright (c) 2001-2010 Neil Bowers
   Copyright (c) 2010-2020 Sullivan Beck

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

=cut