todo
{
More types
use Business::ISSN qw( is_valid_checksum );
#verify the checksum
if( is_valid_checksum('01234567') ) { ... }
use Business::ISBN;
isbn = Business::ISBN->new( $BAD_CHECKSUM_ISBN );
isa_ok( $isbn, 'Business::ISBN' );
is( $isbn->is_valid
use Business::ISMN;
my $ismn = Business::ISMN->new( $GOOD_ISMN );
isa_ok( $ismn, 'Business::ISMN' );
is( $ismn->is_valid
use Business::DE::Konto;
my $konto = Business::DE::Konto->new(BLZ => 12345678, KONTO => 1234567890));
$konto->check();
use Business::ISIN;
my $isin = new Business::ISIN 'US459056DG91';
if ( $isin->is_valid ) {
print "$isin is valid!\n";
# or: print $isin->get() . " is valid!\n";
} else {
print "Invalid ISIN: " . $isin->error() . "\n";
print "The check digit I was expecting is ";
print Business::ISIN::check_digit('US459056DG9') . "\n";
}
use Business::UPC;
# Constructors:
# create a UPC object using standard (type-A) UPC
$upc = new Business::UPC('012345678905');
# create a UPC object using zero-supressed (type-E) UPC
$upc = type_e Business::UPC('01201303');
# is the UPC valid (correct check digit)?
$upc->is_valid;
use Business::CINS;
$cn = Business::CINS->new('035231AH2');
print "Looks good.\n" if $cn->is_valid;
use Business::IBAN;
use Locale::Country;
my $cc = country2code('Germany');
my $iban = Business::IBAN->new();
my $ib = $iban->getIBAN(
{
ISO => $cc, # or "DE", etc.
BIC => 12345678, # Bank Identifier Code, meaning the BLZ
# in Germany
AC => "1234567890",
});
# or
my $ib = $iban->getIBAN(
{
ISO => "DE",
BBAN => 123456781234567890,
});
if ($ib) {
print "IBAN is $ib\n";
}
else {
$iban->printError();
# or
my @errors = $iban->getError();
# print your own error messages (for description of error-
# codes see section ERROR-CODES
}
if ($iban->valid($ib)) {
print "$iban is valid\n";
}
else {
$iban->printError();
}
MARKUP( XML; HTML )
SOURCE( Perl, Java etc. )
}
partial
{
win32 compatibles are inserted, but MARKUP and SOURCE still missing.
}
partial
{
XML added.
}
todo
{
refactor names to terminus technicii from http://www.w3.org/TR/2001/REC-xmlschema-2-20010502/#datatype
}
done
{
sufficiently done.
}
todo
{
recognize new types via their VERSION in release.pl, so we could place them via templates
}
done
bug
{
dverify doesn't work, seee basic.t.
Somehow the try catch block isn't entered.
}
done
{
was silly ';' missing after catch block ! This is a FOP (frequently occuring problem) with the
Error module syntax sugar.
todo
{
FOP should be included into the FAQ.
}
done
}
todo
{
HTML doesn't validate.
In future via HMTL::Tidy. But requires a Perl Wrapper copyright by Charles Reitzel.
Tidy available at http://tidy.sourceforge.net/
}
partial
{
crashes on my system. commented out until someone comes with a better HTML validating module.
HTML::Validator looks a bit old to me.
}
partial
{
HTML::Lint's Test::HTML::Lint seems to be the solution.
}
done
todo
{
- guard.t and typ.t test fail. Looks like a change in Class::Maker default handler return value
behaviour is the cause.
}
done
{
fixed bug in Class::Maker after 0.05.17
}
todo
{
Type::Exception's new isn't called when Failure::Type is thrown/created ?
}
partial
{
But now the caller line is wrong. This should be fixed via Error::Depth modification.
}
partial
{
no general remedy for that. Seems like the user has to set $Error::Depth.
}
todo
{
Add DYNAMIC( 'soap://mytype.wsdl?p1=bla;p2=bli' ) type, which is just a proxy for a soap-service
propagated type testing.
}
todo(CONTRIBUTIONS)
{
* This module is extendable, but this needs documenation and a good introduction.
+ Tutorials / round trip (everyday problems)
- Validation of form data
- Validating class properties
- Database type verification
* HOWTO's
- Add custom datatype
* Publication
- Introductional article with the focus on perl.com and "the perl review"
}
todo
{
_translate should really translate "back and forth" type pkgname <> imported name (respecting
export).
}
comment
{
_translate renamed. Still uptodate?
}
todo
{
Add new type
OS::PATH( 'UNIX|OSX|WIN' ) - a valid path string
}
comment
{
with option mdircheck => 1|0 will try to create and remove
the directory in a temporary dir to see if it is really
creatable.
}
comment
{
File::Spec->catfile( qw(dir file) ) returns dir/file, where the slash is automatically
the systems correct path separator. This could be used to get OS::PATH( 'CURRENT' ) working.
Also Path::Class should be helpfull then.
}
todo
{
catalog/toc with grouping for localization
}
todo
{
overify() / Data::Type::Guard should report which member failed
}
todo
{
Add some localization. Like DE::PHONE which is Type::DE::phone etc.
Should use Locale::Maketext
}
comment
{
PHONE should be default lang, and PHONE::DE is explicit localized version.
}
partial
{
<type>::DE is implemented
}
todo
{
types ->info() should give usable info even we have no object instance, but
package names as $this.
}
done
{
->desc() static text
->info() instance information
->doc() extensive documentation of the type
->usage() complete type invocation usage
->default() default parameter list used instead of empty @_'s_
}
todo
{
summary() is returning confusing describtion in DOMAIN, because
the test's ok( ) procedure is not linear. Dont know how to solve
that easily.
}
comment
{
may be subst multiple regex with only one.
}
comment
{
DOMAIN type regex matching should be stuffed in a facet (otherwise shitty summaries)
}
done
{
Regexp::Common *again* had a excellent RFC regexp for that.
}
todo
{
dverify/is wasnt creating any correct errors, but returned correctly
true or false.
}
done
{
@Data::Type::err is holding the list of negative exceptions.
}
MILESTONE( "W3C XML SCHEMA TYPES" )
{
Now we have a CPAN RC.
}
reached
todo
{
Geneerate for each collection a new namespace instead of using everywhere Data::Type::Object::xxx, because
leads to name conflicts when different collection have the same package name.
}
done
todo
{
Somehow detecting of failures in export names. This is the confusing output when codegen isnt generating the
proxy sub:
"====================================================================================================
Use of inherited AUTOLOAD for non-method Data::Type::VARCHAR() is deprecated at test.pl line 91, <DATA> line 4454.
Can't locate class method 'Data::Type::VARCHAR' via package 'Data::Type' at test.pl line 91"
}
todo
{
add a C<Collection>-wide export prefix, so $Data::Type::Collection::W3C::export_prefix = 'W3C' would lead to
all type export names starting with "W3C::".
}
done
todo
{
add EMAIL( alias_for_at => qr/@|\sat\s|\(at\)|<at>/ ) option, so 'obfuscated' emails are detectable.
}
comment
{
Not sure if this isnt just good for spam-dogs.
}
comment
{
see Email::Valid::Loose etc.
}
todo
{
Option to apply codefilter to translate
my VARCHAR(80) $e, $e2, $ex;
to
typ VARCHAR(80), \( my $e, $e2, $ex );
for mimicing a "builtin" type support via the tie interface.
}
comment
{
aware conflicts with "use types;"
}
comment
{
contact Arthur bergman if C<Data::Type> could marry the C<types> pragma.
}
todo
{
delayed loading of type dependencies (modules) first time the type is used.
}
comment
{
There were this cut+autoload+eval modules on CPAN ?! Or use eval "use Module" unless $INC{Module};
}
done
{
implemented simple lazy loading of modules
}
todo
{
Api tests which use Class::Maker::Reflection to inspect packages
and validate if the abstract interface was correctly implemented.
}
todo
{
Use Pod::Select Pod::XML to generate the pod documentation of datatype/filters as an xml resource.
}
todo
{
Write an Data::Type->add( ) function which allows to easily add custom types.
->add( facets => qw(), interfaces => qw(), desc => , info => , usage => ...
}
todo
{
Eliminate tricks flagged with "dirty"
}
todo(BUG)
{
->test of datatypes requires invocation of facets via Data::Type::ok for correct summary generation.
THIS MEANS: ALL ->test() DATATYPE METHODS MUST BE CLEANED FROM DIRECT C<throw Exception>. POSSIBLE
SOLUTION: EVERYTHING STUFFED IN FACETS:
OR ANY IDEA ?
}
comment
{
For now use of "private facets" is recommended.
}
todo(PRIO => 10)
{
Extend Data::Type::IType with a 'exporter' method which des the namespace/name resolution and takes
the part of exporting (which is bulkly done by C<codegen()>).
NOTE This will fix the $Data::Type::Collection::*::prefix."ID" parts from C<export> method.
}
comment
{
DEFINITION Collection datatypes should implement its own interface.
Example
package Data::Type::Collection::Bio;
package Data::Type::ICBio;
our @ISA = qw(Data::Type::IType);
sub prefix : method { 'BIO::' }
sub exported : method
{
map { __PACKAGE__->prefix().$_ } $this->export;
}
package Data::Type::Object::rna;
our @ISA = qw(Data::Type::ICBio);
}
todo
{
Create a type for strong passwords with Data::Password or whatever.
Articles:
See http://www.microsoft.com/security/articles/password.asp.
See http://www.perlmonks.org/index.pl?node_id=278236
http://www.perlmonks.org/index.pl?node_id=278798
Make sure you create a password that:
Is at least seven characters in length, and the longer the better.
Includes upper and lower case letters, numerals, symbols
Has at least one symbol character in the second through sixth position
Has at least four different characters in your password (no repeats)
Looks like a sequence of random letters and numbers
Make sure you:
Don't use ANY PART of your logon name for your password
Don't use any actual word or name in ANY language
Don't use numbers in place of similar letters
Don't reuse any portion of your old password
Don't use consecutive letters or numbers like "abcdefg" or "234567"
Don't use adjacent keys on your keyboard like "qwerty"
Usage:
SEC::PASSWD( minlength => 10, symbolchars => 1, diversity => 4, random => 1, ... )
SEC::PASSWD( strength => 'strong|medium|weak' );
}
comment
{
use from CPAN Data::Password::BasicCheck. It looks good.
}
todo
{
detected SSN::Validate on CPAN to be next candidate to be supported in this suite.
}
todo
{
also Data::Types should be included. Well, we have a 'cast'able value type now.
}
todo
{
fill all the desc(), info(), usage() etc with comprehensive docu. Then CPAN release.
}
todo
{
CHEM::ATOM
See CPAN Chemistry::Atom:
@ELEMENTS = qw(
n
H He
Li Be B C N O F Ne
Na Mg Al Si P S Cl Ar
K Ca Sc Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br Kr
Rb Sr Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te I Xe
);
}
done
todo
{
cite http://www.perlmonks.org/index.pl?node_id=295306
}
todo
{
write GetOpt extension which adds Data::Type support to parameters.
}
todo
{
Create Regexp::Box
}
done
todo
{
Put in some mechanism to prefix all datatype export names. So one could
define a generale namespace like
use Data::Type from => 't::';
die unless is t::STD::EMAIL;
[Note] For people scarring the the namespace pollution could conflict somewhere.
die unless is __STD::EMAIL;
[Note] Or maybe just a prefixed string like '__' or '~' ?
}
done
{
See EXPORT / OPTIONS / MASTER PREFIX.
}
todo
{
Integrate Email::Valid::Loose into STD::EMAIL.
die unless is STD::EMAIL( loose => 1 );
}
todo
{
->summary with emptry parameters dies without informative message. There must be a "default summary" which can be used within the documentation.
Check the capibility of ->choice and ->default for that.
}
done
{
Not viable because the test procedure is dependant on parameters. Not alle procedures have a default parameter.
}
todo
{
sub _test
{
my $this = shift;
Data::Type->filter( [ 'uc' ] );
sucks!
----
Better add method
sub filters ( return ( [ 'lc' ], [ 'strip', ' ' ] ) }
and then call via
Data::Type::OBject::Interfacet::test
to filter $Data::Type::val.
---
Helps reflecting which filters are applied (automatically added to the docu).
}
done
todo
{
add <result> method which contains the 'parsing/extracting' results which
are for *free* because they are byproduct of the regex matching or whatever.
See L<Regexp::Common>.
}
comment
{
Needs some clarification/definition in the ::RFC.
}
todo
{
Documentaton:
* clean $scalar = bla() headers and change to
=head function( )
$b = function();
@a = function(1);
style.
Complete
* Type/Docs/RFC.pod
- Type/Facet.pm
- Type/Field.pm
- Type/Filter.pm
}
partial
{
still
- Type/Facet.pm
- Type/Field.pm
- Type/Filter.pm
}
partial
{
still
- Type/Field.pm
}
todo
{
shell> perldoc STD::VARCHAR
would be great but is very brute to namespaces etc.
shell> Data::Type::STD::VARCHAR
is better?
shell> perldoc Data::Type::Collection::Std;
is ok though?
}
todo(Data::Type::Docs::RFC)
{
Methods all _private or not? _test is but others?
}
todo
{
write test for L<Data::Type/summary()>.
}
todo
{
allow aliases DB_VARCHAR instead of DB::VARCHAR.
}
partial
{
But exports doesnt work now.
sub DB::VARCHAR ...
is now
sub DB_VARCHAR
[Note] in the first case we have 'export' for free. In the second case it now gets eval'ed within Data::Type:: and
DB_VARCHAR is accessible via Data::Type::DB_VARCHAR.
HMmmmmm..... ? Is this bad ?
}
partial
{
No full blown control via the L<Data::Type/EXPORT/MASTER PREFIX> option.
}
todo
{
Somewhat i would love to write
$_ = cgi->param->{input};
warn unless is( 'std.varchar.de', option1 => 1, option2 => 2, ... );
or
warn unless is( 'bio.dna', option1 => 1, option2 => 2, ... );
or
try
{
valid 'ATGC', 'bio.dna', option1 => 1, option2 => 2, ...;
}
catch Error with
{
};
My gut says is less error prone and easier to read. Syntactic sugar is sometimes too annoying.
}