Re: [Module::Build] Case-Insensitive POD

On Sep 30, 2004, at 5:56 PM, Ron Savage wrote:

> We should be able to discuss ideas without worrying whether or not 
> someone agrees with them. We only really find out about the agreement 
> or disagreement during the discussion. Prejudging the degree of 
> agreement is not really productive.

Yeah - if there are other points we haven't considered, or facts in 
error, do raise them.

  -Ken

-------------------------------------------------------
This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
Use IT products in your business? Tell us what you think of them. Give us
Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more
http://productguide.itmanagersjournal.com/guidepromo.tmpl

Re: [Module::Build] Case-Insensitive POD

Re: [Module::Build] Heretical ideas about M::B

On Thu, 30 Sep 2004, Randy W. Sims wrote:

> Austin Schutz wrote:
[ ... ]

> > M::B makes use of 'backlink' in Pod::Html, a newer
> > feature which doesn't exist in the versions included
> > with older perls. Possibly some other conflicts, haven't
> > gotten past this yet. IMO M::B should work with
> > reasonably old stable perls out of the box, because
> > people are still developing for them as targets.
>
> Backwards compatability is one of Ken's goals. Most of the
> code is compatable back to 5.005, I believe. However, it's
> possible that some sneaked in. I think Randy Kobes
> contributed the htmldoc support; perhaps he can help here
> if he's still on the list?

Does the following fix this?
==========================================================
Index: lib/Module/Build/Base.pm
===================================================================
RCS file: /cvsroot/module-build/Module-Build/lib/Module/Build/Base.pm,v
retrieving revision 1.334
diff -u -r1.334 Base.pm
--- lib/Module/Build/Base.pm	28 Sep 2004 21:24:26 -0000	1.334
+++ lib/Module/Build/Base.pm	1 Oct 2004 12:03:06 -0000
 <at>  <at>  -1723,13 +1723,14  <at>  <at> 
   }

   foreach my $pod (keys %$pods){
-    $self->_htmlify_pod(
-			path => $pod,
-			rel_path => $pods->{$pod},
-			htmldir => $html,
-			backlink => '__top',
-			css => ($^O =~ /Win32/) ? 'Active.css' : '',
-		       );
+    my %opts = (path => $pod,
+                rel_path => $pods->{$pod},
+                htmldir => $html,
+                css => ($^O =~ /Win32/) ? 'Active.css' : '',
+                eval {Pod::Html->VERSION(1.03); 1} ?
+                    (backlink => '__top') : (),
+               );
+    $self->_htmlify_pod(%opts);
   }
 }

================================================================

--

-- 
best regards,
randy

-------------------------------------------------------
This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
Use IT products in your business? Tell us what you think of them. Give us
Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more
http://productguide.itmanagersjournal.com/guidepromo.tmpl

Re: [Module::Build] Case-Insensitive POD

On Sep 30, 2004, at 8:31 AM, Ken Williams wrote:

> Yeah, I agree, I don't want M::B to dictate any POD formats.  But I do 
> think it should be compliant with the tools/specs that *do* dictate 
> those formats.  So I don't want to strike out on my own and start 
> loosening specs that I don't own.

Yes, but my point is that I don't believe the change would make any 
difference to them, an no one would ever notice.

Regards,

David

-------------------------------------------------------
This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
Use IT products in your business? Tell us what you think of them. Give us
Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more
http://productguide.itmanagersjournal.com/guidepromo.tmpl

Re: [Module::Build] Case-Insensitive POD

On Fri, Oct 01, 2004 at 09:22:33AM -0700, David Wheeler wrote:
> >Yeah, I agree, I don't want M::B to dictate any POD formats.  But I do 
> >think it should be compliant with the tools/specs that *do* dictate 
> >those formats.  So I don't want to strike out on my own and start 
> >loosening specs that I don't own.
> 
> Yes, but my point is that I don't believe the change would make any 
> difference to them, an no one would ever notice.

I was pondering this discussion some more the other day while driving.
As I belive another poster mentioned, the case-sensitivity of these
header sections could be problematic, esp. for folks like myself who
weren't even aware that of the issue.

If Ken is not willing to let M::B be case agnostic when gathering module
details, then perhaps it would be good to issue a warning when the
module is processed if M::B finds a header that is not in the proper
case.

Just a thought,
William

--

-- 
Knowmad Services Inc.
http://www.knowmad.com

-------------------------------------------------------
This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
Use IT products in your business? Tell us what you think of them. Give us
Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more
http://productguide.itmanagersjournal.com/guidepromo.tmpl

Re: [Module::Build] Case-Insensitive POD

On Oct 1, 2004, at 9:51 AM, William McKee wrote:

> If Ken is not willing to let M::B be case agnostic when gathering 
> module
> details, then perhaps it would be good to issue a warning when the
> module is processed if M::B finds a header that is not in the proper
> case.

Or no header at all.

Regards,

David

-------------------------------------------------------
This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
Use IT products in your business? Tell us what you think of them. Give us
Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more
http://productguide.itmanagersjournal.com/guidepromo.tmpl

[Module::Build] Module dependency expressions

Have you had a chance to look at any of the expression parsing solutions 
from the Perl QotW list yet? (links below for convenience) The number of 
responses was disappointing, but showed a variety of approaches. Should 
I pursue any of these solutions further? (No is a perfectly acceptable 
answer, but an answer is appreciated.)

Randy.

My Summary: (Manual shift-reduce parser)
<http://perl.plover.com/~alias/list.cgi?mss:108>

Tassilo von Parseval (impl my macro alt idea)
<http://perl.plover.com/~alias/list.cgi?1:mss:2276>

Mark Jason Dominus (Custom shift-reduce parser generator)
<http://perl.plover.com/~alias/list.cgi?1:mss:2277>

Marc Prewitt (Parse::RecDescent)
<http://perl.plover.com/~alias/list.cgi?1:mss:2279>

Ron Isaacson (Version discussion)
<http://perl.plover.com/~alias/list.cgi?1:mss:2278>

-------------------------------------------------------
This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
Use IT products in your business? Tell us what you think of them. Give us
Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more
http://productguide.itmanagersjournal.com/guidepromo.tmpl

Re: [Module::Build] Heretical ideas about M::B

On Fri, Oct 01, 2004 at 07:05:40AM -0500, Randy Kobes wrote:
> > Austin Schutz wrote:
> [ ... ]
> 
> > > M::B makes use of 'backlink' in Pod::Html, a newer
> > > feature which doesn't exist in the versions included
> > > with older perls. Possibly some other conflicts, haven't
> > > gotten past this yet. IMO M::B should work with
> > > reasonably old stable perls out of the box, because
> > > people are still developing for them as targets.
> >
> > Backwards compatability is one of Ken's goals. Most of the
> > code is compatable back to 5.005, I believe. However, it's
> > possible that some sneaked in. I think Randy Kobes
> > contributed the htmldoc support; perhaps he can help here
> > if he's still on the list?
> 
> Does the following fix this?

	No, because it still doesn't like having --backlink in the args even
without a value. It also doesn't like having --header there.

	This works for me, though I moved the change to _htmlify_pods().

	Thanks,
	Austin

*** /usr/perl5/site_perl/5.005/Module/Build/Base.pm     Tue Sep 28 14:42:01 2004
--- /opt/packages/rancid/lib/perl/Module/Build/Base.pm  Tue Oct  5 22:12:42 2004
***************
*** 1517,1525 ****

    my $blib = $self->blib;
    my  <at> opts = (
!             '--header',
              '--flush',
-             "--backlink=$args{backlink}",
              "--title=$title",
              "--podpath=$podpath",
              "--infile=$infile",
--- 1517,1528 ----

    my $blib = $self->blib;
    my  <at> opts = (
!               (
!                eval {Pod::Html->VERSION(1.03); 1} ?
!                ( '--header', "--backlink=$args{backlink}") :
!                ()
!               ),
              '--flush',
              "--title=$title",
              "--podpath=$podpath",
              "--infile=$infile",

-------------------------------------------------------
This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
Use IT products in your business? Tell us what you think of them. Give us
Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more
http://productguide.itmanagersjournal.com/guidepromo.tmpl

[Module::Build] Question/Suggestion

Hi

Is there any standard way to say that a file, say 'foo.dat' is a data 
file to be installed somewhere (don't ask me where) in the Perl library 
tree?

It would be cool to have such a way (specially if then we had a handy 
way to, from the given module, access that file).

Kind Regards,
Alberto
--

-- 
Alberto Simões

Much as I hate to say it, the Computer Science view of language design
has gotten too inbred in recent years. The Computer Scientists should
pay more attention to the Linguists, who have a much better handle on
how people prefer to communicate.
             --Larry Wall

-------------------------------------------------------
This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
Use IT products in your business? Tell us what you think of them. Give us
Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more
http://productguide.itmanagersjournal.com/guidepromo.tmpl

[Module::Build] find_perl_interpreter() Finds Wrong Perl

Module-Builders,

I've had a report that the shebang line is not getting properly set 
when someone installs my new module, SVN::Notify. The script that comes 
with it, svnnotify, has the shebang line of #!/usr/bin/perl. Robert 
Spier installed it on a Perl.org server, but the shebang line was not 
changed, even though he was installing it with a version of Perl living 
somewhere other than /usr/bin/perl.

I looked at the source for find_perl_interpreter(), but since it's 
comparing the output of Config::myconfig() to determine if it has found 
the right Perl, and since myconfig() has no date and therefore could be 
identical between installed versions of Perl (because they're the same 
versions and were configured identically), find_perl_interpreter() is 
going with the wrong Perl.

So, let me ask a simple question: Why not use the value of 
$Config{startperl} or $Config{perlpath} to fix the shebang line? It 
seems like the values of these config settings would be more 
appropriate for use in fix_shebang_line().

Would you accept a patch adding this functionality? Something like this:

-- Base.pm.~1.330.~	Tue Sep  7 19:18:15 2004
+++ Base.pm	Tue Oct  5 21:00:06 2004
 <at>  <at>  -177,6 +177,7  <at>  <at> 
    return $^X if File::Spec->file_name_is_absolute($^X);
    my $proto = shift;
    my $c = ref($proto) ? $proto->{config} : \%Config::Config;
+  return $c->{perlpath} if $c->{perlpath};
    my $exe = $c->{exe_ext};

    my $thisperl = $^X;

Thanks,

David

-------------------------------------------------------
This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
Use IT products in your business? Tell us what you think of them. Give us
Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more
http://productguide.itmanagersjournal.com/guidepromo.tmpl