New module author

[Luke Palmer]

I would like an author id, as I have a module to upload.

Name: Luke Palmer
Email: [EMAIL PROTECTED]
Homepage: http://fibonaci.babylonia.flations.org
User-ID: LPALMER

I'm going to contribute Parse::Earley, a neato implementation of Earley's 
algorithm... Parse::RecDescent style (but not as full-featured).


Luke

Naming a module...

[Luke Palmer]

I'd like assistance in naming a module that I'm working on.  It's a 
basic predicate logic solver (like prolog, but easier to access through 
perl and not as full-featured).  I call it Logic now, but I'd like to know 
where to put it in the tree. I would use Language::Prolog, except it is 
pre-pre-pre-alpha and doesn't do anything.

A sample use: (also some suggestions on the names or usage of these 
functions?)

$logic = new Logic;
$logic->parse( 'door(a, b)' );
$logic->parse( 'door(a, c)' );
$logic->parse( 'door(b, c)' );

#Find doors from a to some place X
$query = $logic->parse( 'door(a, X)?' );

#Find first one
$pad = $logic->match( $query );
$pad and print "Door from a to $pad->{X}\n";

#Find next one
$pad = $logic->match( $query, $pad );

# ... et cetera


Thanks,
Luke

Request for PAUSE ID

[Luke Palmer]

I resent this, because it's been a really long time and people after me in 
the queue (I assume you do a queue-like system) have been added.  Maybe it 
was missed?

Here's the original:

> I would like an author id, as I have a module to upload.
> 
> Name: Luke Palmer
> Email: [EMAIL PROTECTED]
> Homepage: http://fibonaci.babylonia.flations.org
> User-ID: LPALMER
> 
> I'm going to contribute Parse::Earley, a neato implementation of Earley's 
> Context-Free parsing algorithm.


Luke

Re: PAUSE permissions clash between PARENT and parent

[Luke Palmer]

I don't really care. I wouldn't want to break anybody's code, but I doubt
this package has many users. I haven't done any perl for many years, can't
say I even know what you're talking about. I defer to your judgment.

Luke

On Wed, Jun 22, 2016, 2:44 PM Neil Bowers  wrote:

> Hi Luke,
>
> > I’m emailing you wearing my PAUSE admin hat: I’m working on resolving
> conflicts caused by PAUSE now considering package names case insensitively.
> This has left us with some situations where people are owners of namespaces
> previously considered distinct, and now considered the same.
> >
> > You have ownership of the PARENT namespace, which is used by your
> Class::Role module (last released in 2003), in the Class-Role distribution.
> Your PARENT namespace now conflicts with the “parent” module, which has
> been shipped with Perl since 5.10.1.
> >
> > To resolve this I’d like to drop your ownership of the “PARENT”
> namespace. To fully clean this up you should really release a new version
> that doesn’t use the PARENT namespace, or delete Class-Role from CPAN
> entirely. But because this is a cuckoo package (not a stand-alone module),
> installing your Class::Role module can’t clobber “parent” on
> case-insensitive filesystems.
>
> Would you be happy for Class-Role to be deleted from CPAN?
>
> Or if you’d like, I’ll do a release of Class::Role with PARENT changed to
> something like PARENTCLASS, which hasn’t already been taken.
>
> Cheers,
> Neil
>
>
>
>

Module Naming

[Luke Palmer]

I have a module that performs parallel substitutions on a string, as
in:

   $str = subst $str,
  qr/foo/ => 'bar',
  qr/bar/ => 'baz',
  qr/baz/ => 'foo';

Just as if it were typed:

   $str =~ s/(foo|bar|baz)/$1 eq 'foo' ? 'bar' :
   $1 eq 'bar' ? 'baz' :
   $1 eq 'baz' ? 'foo'/g;

For any set of regexes you want, handling $1,$2 and all that
correctly.  I call it Regexp::Subst::Parallel at the moment.  Does
anyone have any better ideas for the name of such a module?

Additionally, can someone think of a better calling convention to use
than the one illustrated above?  I'm running into problems with other
kinds of interpolations:

  $str = substr,
 qr/(the contractee)/i => "\$1, $name, ",
 qr/(the contractor)/i => "\$1, $name2, ";

That works, replacing \$1 with "the contractee" in whatever case it
was.  The problem is, that for a literal $ with other interpolations,
it would have to be "\\\$", which is by no means pretty.  Any ideas?

Thanks,
Luke

Module naming/behavior question

[Luke Palmer]

I have a module that does multiple regex substitutions in parallel
safely, like this:

$str = subst($str,
   qr/([0-9]*[1-9])/ => '${1}0',
   qr/([0-9]+)0/ => '$1',
)

(Adding zeroes to numbers that don't have them, and taking them off
numbers that do).

I now call it Regexp::Subst::Parallel.  Would there be a better name
for this module?

Also, would there be a better interface?  Should it be object
oriented?

Thanks,
Luke