PPM Catalyst/Manual.html clash.

There is an odd file clash with PPM (the ActivePerl module
installer).

The fact: Catalyst::Runtime provides the file
Catalyst/Manual.pod and
Catalyst::Manual provides the file Catalyst/Manual.pm, each
of which
causes PPM to create a Catalyst/Manual.html file under the
html docs
tree it generates.

This may seem a trivial problem, but it unfortunately causes
PPM to
fail miserably when installing Task::Catalyst (or
Catalyst::Runtime and
Catalyst::Manual anyway).

This is a real shame, since it prevents the casual Windows
Perl user 
from trying Catalyst, whereas installing Catalyst on Windows
could be 
nowadays as smooth as a couple of clicks (as required by
PPM4 to install 
Task::Catalyst).
(PPM aside, the Catalyst/Manual.(pm|pod) duplication should
probably be
avoided anyway).

Could you please remove (or rename) Manual.pod from
Catalyst::Runtime or
adopt any other solution you see fit?

Many thanks,
Emanuele.

_______________________________________________
Catalyst-dev mailing list
Catalyst-devlists.rawmode.org
http://lists.rawmode.org/mailman/listinfo/catalyst-dev

On Sat, Jul 14, 2007 at 02:09:50PM +0200, Emanuele Zeppieri
wrote:
> There is an odd file clash with PPM (the ActivePerl
module installer).
> 
> The fact: Catalyst::Runtime provides the file
Catalyst/Manual.pod and
> Catalyst::Manual provides the file Catalyst/Manual.pm,
each of which
> causes PPM to create a Catalyst/Manual.html file under
the html docs
> tree it generates.
> 
> This may seem a trivial problem, but it unfortunately
causes PPM to
> fail miserably when installing Task::Catalyst (or
Catalyst::Runtime and
> Catalyst::Manual anyway).

Don't install the manual dist then, you can read it online
via search.cpan
anyway.
 
> Could you please remove (or rename) Manual.pod from
Catalyst::Runtime or
> adopt any other solution you see fit?

Before we start implementing workarounds for PPM brokenness,
could you
file a bug against PPM and see if active state will resolve
it?

I'm more and more seeing windows perl users using Vanilla
Perl due to just
this sort of silly problem that's ActiveState-specific; I'd
rather not
modify our distros just to support one vendor-controlled
perl distro with
a stupid packaging system if we can possibly avoid it.

-- 
      Matt S Trout       Need help with your Catalyst or
DBIx::Class project?
   Technical Director    Want a managed development or
deployment platform?
 Shadowcat Systems Ltd.  Contact mst (at)
shadowcatsystems.co.uk for a quote
http://chainsawblues.vo
x.com/             http://www.shadowc
atsystems.co.uk/ 

_______________________________________________
Catalyst-dev mailing list
Catalyst-devlists.rawmode.org
http://lists.rawmode.org/mailman/listinfo/catalyst-dev

Matt S Trout wrote:

> On Sat, Jul 14, 2007 at 02:09:50PM +0200, Emanuele
Zeppieri wrote:
>> There is an odd file clash with PPM (the ActivePerl
module installer).
>>
>> The fact: Catalyst::Runtime provides the file
Catalyst/Manual.pod and
>> Catalyst::Manual provides the file
Catalyst/Manual.pm, each of which
>> causes PPM to create a Catalyst/Manual.html file
under the html docs
>> tree it generates.
>>
>> This may seem a trivial problem, but it
unfortunately causes PPM to
>> fail miserably when installing Task::Catalyst (or
Catalyst::Runtime and
>> Catalyst::Manual anyway).
> 
> Don't install the manual dist then, you can read it
online via search.cpan
> anyway.

In PPM, Catalyst::Manual is installed by Catalyst:evel
which in turn 
is installed by Task::Catalyst.

I was not asking for a workaround anyway, I was just
pointing out that 
installing Catalyst under Windows could be a smooth and
effortless 
operation for everyone, even the most naive user, if we only
didn't have 
this (useless) file (quasi-)duplication.

>> Could you please remove (or rename) Manual.pod from
Catalyst::Runtime or
>> adopt any other solution you see fit?
> 
> Before we start implementing workarounds for PPM
brokenness, could you
> file a bug against PPM and see if active state will
resolve it?

I don't think PPM is to blame here (though it could signal
the file 
conflict and go on, rather than aborting the installation).

You should consider my suggestion also because, aside from
PPM, there is 
not much sense in having Catalyst/Manual.pm and then also 
Catalyst/Manual.pod from another dist (Catalyst::Runtime)
which 
shouldn't provide any /manual/.

Just drop Manual.pod from Catalyst::Runtime, because
Catalyst will be 
cleaner 

> I'm more and more seeing windows perl users using
Vanilla Perl due to just
> this sort of silly problem that's ActiveState-specific;
I'd rather not
> modify our distros just to support one
vendor-controlled perl distro with
> a stupid packaging system if we can possibly avoid it.


Though I could agree in general, this is not the case.

Cheers,
Emanuele.

_______________________________________________
Catalyst-dev mailing list
Catalyst-devlists.rawmode.org
http://lists.rawmode.org/mailman/listinfo/catalyst-dev

On Sat, Jul 14, 2007 at 04:24:45PM +0200, Emanuele Zeppieri
wrote:
> In PPM, Catalyst::Manual is installed by Catalyst:evel
which in turn 
> is installed by Task::Catalyst.
> 
> I was not asking for a workaround anyway, I was just
pointing out that 
> installing Catalyst under Windows could be a smooth and
effortless 
> operation for everyone, even the most naive user, if we
only didn't have 
> this (useless) file (quasi-)duplication.

Or if PPM was fixed to not be so fscking stupid.
 
> You should consider my suggestion also because, aside
from PPM, there is 
> not much sense in having Catalyst/Manual.pm and then
also 
> Catalyst/Manual.pod from another dist
(Catalyst::Runtime) which 
> shouldn't provide any /manual/.
> 
> Just drop Manual.pod from Catalyst::Runtime, because
Catalyst will be 
> cleaner 

We tried that. New users didn't realise the manual existed.

And we can't drop the .pm from ::Manual because then CPAN.pm
wouldn't know
to install it.

And there are far, far more Catalyst users on non-win32
platforms since it's
a fairly lousy platform for server-side development work.

I'd love to make it easier for you to install Catalyst using
your chosen
closed source GUI tool, but not at the expense of everybody
using the standard
free software solutions for perl deployment.

Anybody got a suggestion for a third way here?

-- 
      Matt S Trout       Need help with your Catalyst or
DBIx::Class project?
   Technical Director    Want a managed development or
deployment platform?
 Shadowcat Systems Ltd.  Contact mst (at)
shadowcatsystems.co.uk for a quote
http://chainsawblues.vo
x.com/             http://www.shadowc
atsystems.co.uk/ 

_______________________________________________
Catalyst-dev mailing list
Catalyst-devlists.rawmode.org
http://lists.rawmode.org/mailman/listinfo/catalyst-dev

On Sat, Jul 14, 2007 at 06:47:42PM +0200, Emanuele Zeppieri
wrote:
> Matt S Trout wrote:
> 
> >>Just drop Manual.pod from Catalyst::Runtime,
because Catalyst will be 
> >>cleaner 
> >
> >We tried that. New users didn't realise the manual
existed.
> 
> Really?!
> 
> Wasn't "See L<Catalyst::Manual> for more
documentation." in 
> Catalyst.pm's POD sufficient?
> 
> I can't imagine a user who didn't understand that (not
even a Windows 
> user 

People often browse the search.cpan dist page to look for
documentation.

It's amazing how easily users can miss links in POD,
DBIx::Class::Relationship
links to Relationship::Base with a similar message and new
users miss that
all the time.

> >I'd love to make it easier for you to install
Catalyst using your chosen
> >closed source GUI tool, but not at the expense of
everybody using the 
> >standard
> >free software solutions for perl deployment.
> 
> Nor Windows nor ActivePerl are my platforms of choice,
though I feel no 
> particular perverse pleasure in boycotting them.

I don't believe that's what I'm suggesting.
 
> Seriously, placing a fake file in Catalyst::Runtime
just to say: "Hey 
> man, look, there is also a Catalyst Manual!" is
not a good idea IMO.

I don't desperately like it, but I don't have a better
solution.

> Then incidentally it also causes that problem with PPM,
so you 
> ultimately get two damages at once...

Have you filed a bug against PPM yet? 
 
> Believe it or not, I suggested that *only* pro Catalyst
and its spread 
> among the greatest possible number of users (regardless
of the platform 
> they use).

I believe you. I just disagree that what you're suggesting
would be an
improvement ovr the current situation, on balance.

Hmm.

Does anybody know if calling the Catalyst/Manual.pm
something else but
leaving the package line intact would DTRT? I have a feeling
it would, and
that would let us leave the .pod in the main dist, 'cpan
Catalyst::Manual'
would still DTRT, -and- ppm would (hopefully) not break
itself.

-- 
      Matt S Trout       Need help with your Catalyst or
DBIx::Class project?
   Technical Director    Want a managed development or
deployment platform?
 Shadowcat Systems Ltd.  Contact mst (at)
shadowcatsystems.co.uk for a quote
http://chainsawblues.vo
x.com/             http://www.shadowc
atsystems.co.uk/ 

_______________________________________________
Catalyst-dev mailing list
Catalyst-devlists.rawmode.org
http://lists.rawmode.org/mailman/listinfo/catalyst-dev

On Sat, Jul 14, 2007 at 08:11:26PM +0200, Emanuele Zeppieri
wrote:
> Matt S Trout wrote:
> 
> >Have you filed a bug against PPM yet? 
> 
> Done!
> htt
p://bugs.activestate.com/show_bug.cgi?id=71049
> 
> Let's see if it is ActiveState or Catalyst to arrive
first 

If we find an acceptable workaround we'll likely implement
it whether they
fix PPM or not - there'll always be people who don't
upgrade.

Thanks for the report, and thanks -very- much for filing the
bug.

-- 
      Matt S Trout       Need help with your Catalyst or
DBIx::Class project?
   Technical Director    Want a managed development or
deployment platform?
 Shadowcat Systems Ltd.  Contact mst (at)
shadowcatsystems.co.uk for a quote
http://chainsawblues.vo
x.com/             http://www.shadowc
atsystems.co.uk/ 

_______________________________________________
Catalyst-dev mailing list
Catalyst-devlists.rawmode.org
http://lists.rawmode.org/mailman/listinfo/catalyst-dev

On Tue, Jul 17, 2007 at 06:26:01AM +0200, A. Pagaltzis
wrote:
> * Matt S Trout <dbix-classtrout.me.uk> [2007-07-14
15:30]:
> > Before we start implementing workarounds for PPM
brokenness,
> > could you file a bug against PPM and see if active
state will
> > resolve it?
> 
> How do you expect them to resolve it? I don’t see how
a computer
> (as opposed to a human) should be able to tell that
Manual.pm
> supersedes Manual.pod without some mechanism to say so,
and such
> a mechanism simply does not exist in the CPAN
toolchain.
> 
> (If I had to come up with a heuristic, I would in fact
consider
> `.pod` to be more specific than `.pm` – exactly the
opposite of
> what Catalyst would need.)

Actually, that's exactly what perldoc does and it works
absolutely fine
- and is another reason we need to continue to ship a
Manual.pod.

> I suggest simply renaming the file to something
equally
> attractive-looking, like UserManual.pod. That would
remove the
> ambiguity and would still catch all the flies that
Manual.pod
> attracts. Is there a problem I am overlooking with
that?

Is therere a problem I'm overlooking with my proposol to
rename Manual.pm
but leave the package declaration intact that leads you to
suggest the exact
same thing in reverse but less pretty?

-- 
      Matt S Trout       Need help with your Catalyst or
DBIx::Class project?
   Technical Director    Want a managed development or
deployment platform?
 Shadowcat Systems Ltd.  Contact mst (at)
shadowcatsystems.co.uk for a quote
http://chainsawblues.vo
x.com/             http://www.shadowc
atsystems.co.uk/ 

_______________________________________________
Catalyst-dev mailing list
Catalyst-devlists.rawmode.org
http://lists.rawmode.org/mailman/listinfo/catalyst-dev

* Matt S Trout <dbix-classtrout.me.uk> [2007-07-17
14:40]:
> On Tue, Jul 17, 2007 at 06:26:01AM +0200, A. Pagaltzis
wrote:
> > I suggest simply renaming the file to something
equally
> > attractive-looking, like UserManual.pod. That
would remove
> > the ambiguity and would still catch all the flies
that
> > Manual.pod attracts. Is there a problem I am
overlooking with
> > that?
> 
> Is therere a problem I'm overlooking with my proposol
to rename
> Manual.pm but leave the package declaration intact that
leads
> you to suggest the exact same thing in reverse but less
pretty?

Hmm, I didn’t see that in the thread? In any case I’ve
gotten
somewhat confused about which is which and what your
proposal
would change, though it doesn’t really matter. The only
thing
that concerns me is to address this at all; *how* is not so
important as long as it works, and since you have a
proposal
then chances are it’s fine. Mine’s just the first thing
that
occured to me.

Regards,
-- 
Aristotle Pagaltzis // <http://plasmasturm.org/&g
t;

_______________________________________________
Catalyst-dev mailing list
Catalyst-devlists.rawmode.org
http://lists.rawmode.org/mailman/listinfo/catalyst-dev