NAME¶
Unicode::Stringprep - Preparation of Internationalized Strings (RFC 3454)
SYNOPSIS¶
use Unicode::Stringprep;
use Unicode::Stringprep::Mapping;
use Unicode::Stringprep::Prohibited;
my $prepper = Unicode::Stringprep->new(
3.2,
[ { 32 => '<SPACE>'}, ],
'KC',
[ @Unicode::Stringprep::Prohibited::C12, @Unicode::Stringprep::Prohibited::C22,
@Unicode::Stringprep::Prohibited::C3, @Unicode::Stringprep::Prohibited::C4,
@Unicode::Stringprep::Prohibited::C5, @Unicode::Stringprep::Prohibited::C6,
@Unicode::Stringprep::Prohibited::C7, @Unicode::Stringprep::Prohibited::C8,
@Unicode::Stringprep::Prohibited::C9 ],
1, 0 );
$output = $prepper->($input)
DESCRIPTION¶
This module implements the
stringprep framework for preparing Unicode
text strings in order to increase the likelihood that string input and string
comparison work in ways that make sense for typical users throughout the
world. The
stringprep protocol is useful for protocol identifier
values, company and personal names, internationalized domain names, and other
text strings.
The
stringprep framework does not specify how protocols should prepare
text strings. Protocols must create profiles of stringprep in order to fully
specify the processing options.
FUNCTIONS¶
This module provides a single function, "new", that creates a perl
function implementing a
stringprep profile.
This module exports nothing.
- new($unicode_version, $mapping_tables,
$unicode_normalization ,
$prohibited_tables, $bidi_check,
$unassigned_check )
- Creates a "bless"ed function reference that implements a
stringprep profile.
This function takes the following parameters:
- $unicode_version
- The Unicode version specified by the stringprep profile.
Currently, this parameter must be 3.2 (numeric).
- $mapping_tables
- The mapping tables used for stringprep.
The parameter may be a reference to a hash or an array, or
"undef". A hash must map Unicode codepoints (as integers,
e. g. 0x0020 for U+0020) to replacement strings (as perl strings).
An array may contain pairs of Unicode codepoints and replacement strings
as well as references to nested hashes and arrays.
Unicode::Stringprep::Mapping provides the tables from RFC 3454,
Appendix B.
For further information on the mapping step, see RFC 3454,
section 3.
- $unicode_normalization
- The Unicode normalization to be used.
Currently, "undef"/'' (no normalization) and 'KC' (compatibility
composed) are specified for stringprep.
For further information on the normalization step, see RFC 3454,
section 4.
Normalization form KC will also enable checks for some problem sequences for
which the normalization can't be implemented in an interoperable way.
For more information, see "CAVEATS" below.
- $prohibited_tables
- The list of prohibited output characters for stringprep.
The parameter may be a reference to an array, or "undef". The
array contains pairs of codepoints, which define the start
and end of a Unicode character range (as integers). The end
character may be "undef", specifying a single-character range.
The array may also contain references to nested arrays.
Unicode::Stringprep::Prohibited provides the tables from RFC 3454,
Appendix C.
For further information on the prohibition checking step, see
RFC 3454, section 5.
- $bidi_check
- Whether to employ checks for confusing bidirectional text. A boolean
value.
For further information on the bidi checking step, see RFC 3454,
section 6.
- $unassigned_check
- Whether to check for and prohibit unassigned characters. A boolean value.
The check must be used when creating stored strings. It should not be
used for query strings, increasing the chance that newly assigned
characters work as expected.
For further information on stored and query strings, see
RFC 3454, section 7.
The function returned can be called with a single parameter, the string to be
prepared, and returns the prepared string. It will die if the input string
cannot be successfully prepared because it would contain invalid output (so
use "eval" if necessary).
For performance reasons, it is strongly recommended to call the "new"
function as few times as possible, i. e. exactly once per
stringprep profile. It might also be better not to use this module
directly but to use (or write) a module implementing a profile, such as
Authen::SASL::SASLprep.
IMPLEMENTING PROFILES¶
You can easily implement a
stringprep profile without subclassing:
package ACME::ExamplePrep;
use Unicode::Stringprep;
use Unicode::Stringprep::Mapping;
use Unicode::Stringprep::Prohibited;
*exampleprep = Unicode::Stringprep->new(
3.2,
[ \@Unicode::Stringprep::Mapping::B1, ],
'',
[ \@Unicode::Stringprep::Prohibited::C12,
\@Unicode::Stringprep::Prohibited::C22, ],
1,
);
This binds "ACME::ExamplePrep::exampleprep" to the function created by
"Unicode::Stringprep->new".
Usually, it is not necessary to subclass this module. Sublassing this module is
not recommended.
DATA TABLES¶
The following modules contain the data tables from RFC 3454. These
modules are automatically loaded when loading "Unicode::Stringprep".
- •
- Unicode::Stringprep::Unassigned
@Unicode::Stringprep::Unassigned::A1 # Appendix A.1
- •
- Unicode::Stringprep::Mapping
@Unicode::Stringprep::Mapping::B1 # Appendix B.1
@Unicode::Stringprep::Mapping::B2 # Appendix B.2
@Unicode::Stringprep::Mapping::B2 # Appendix B.3
- •
- Unicode::Stringprep::Prohibited
@Unicode::Stringprep::Prohibited::C11 # Appendix C.1.1
@Unicode::Stringprep::Prohibited::C12 # Appendix C.1.2
@Unicode::Stringprep::Prohibited::C21 # Appendix C.2.1
@Unicode::Stringprep::Prohibited::C22 # Appendix C.2.2
@Unicode::Stringprep::Prohibited::C3 # Appendix C.3
@Unicode::Stringprep::Prohibited::C4 # Appendix C.4
@Unicode::Stringprep::Prohibited::C5 # Appendix C.5
@Unicode::Stringprep::Prohibited::C6 # Appendix C.6
@Unicode::Stringprep::Prohibited::C7 # Appendix C.7
@Unicode::Stringprep::Prohibited::C8 # Appendix C.8
@Unicode::Stringprep::Prohibited::C9 # Appendix C.9
- •
- Unicode::Stringprep::BiDi
@Unicode::Stringprep::BiDi::D1 # Appendix D.1
@Unicode::Stringprep::BiDi::D2 # Appendix D.2
CAVEATS¶
In Unicode 3.2 to 4.0.1, the specification of UAX #15: Unicode Normalization
Forms for forms NFC and NFKC is not logically self-consistent. This has been
fixed in Corrigendum #5
(<
http://unicode.org/versions/corrigendum5.html>).
Unfortunately, this yields two ways to implement NFC and NFKC in Unicode 3.2, on
which the Stringprep standard is based: one based on a literal interpretation
of the original specification and one based on the corrected specification.
The output of these implementations differs for a small class of strings, all
of which can't appear in meaningful text. See UAX #15, section 19
<
http://unicode.org/reports/tr15/#Stability_Prior_to_Unicode41> for
details.
This module will check for these strings and, if normalization is done, prohibit
them in output as it is not possible to interoperate under these
circumstandes.
Please note that due to this, the
normalization step may cause the
preparation to fail. That is, the preparation function may die even if there
are no prohibited characters and no checks for bidi sequences and unassigned
characters, which may be surprising.
AUTHOR¶
Claus Faerber <CFAERBER@cpan.org>
LICENSE¶
Copyright 2007-2009 Claus Faerber.
This library is free software; you can redistribute it and/or modify it under
the same terms as Perl itself.
SEE ALSO¶
Unicode::Normalize, RFC 3454
(<
http://www.ietf.org/rfc/rfc3454.txt>)