On Sep 20, 2013, at 9:03 AM, Nicholas Clark <n...@ccl4.org> wrote:

> On Thu, Sep 19, 2013 at 10:29:37PM -0500, Craig A. Berry wrote:
>> 
>> On Sep 19, 2013, at 1:36 PM, Nicholas Clark <n...@ccl4.org> wrote:
>> 
>>> So, my question is, surely that file should be installed as
>>> "utils/perldoc.COM" not "utils/perldoc"?
>> 
>> Yes.
> 
> Would anyone notice if it wasn't installed as utils/perldoc?
> (This might be related to the question below)

I can imagine a plain perldoc being useful for anyone running bash on VMS, 
where I don't think perldoc.com would work.  But in that case perldoc should 
not be identical to perldoc.com.  The former should get an ordinary Unix-like 
shebang edit rather than the full monty startperl prependage.
> 
> So it looks like the FIXIN command would be the correct place to do such a
> renaming. Or duplicating.

Yes, I think so.

> So it looks like on Win32, both foo and foo.bat are installed, but foo and
> foo.bat differ.
> 
> Whereas, if I've got VMS figured out, foo and foo.com would be identical,

They are now by accident.  I don't think they should be the same if we install 
both (which seems sensible).

> So, I guess next VMS question is:
> 
> Is it actually necessary to install both (eg) perldoc and perldoc.com?
> 
> 
> In that, ignoring backwards compatibility with what people expect
> 
> 1) We seem to have a discrepancy between what the core installs for perldoc,
>   and what a CPAN update installs
> 2) We seem to have a discrepancy between scripts installed by the core
>   ("foo.com") and scripts installed by non-core modules (just "bar")
> 3) It would make the core build process a lot easier if ExtUtils::MakeMaker
>   did the ".com" trick, as it's already doing the ".bat" dance for Win32,
>   because this would permit a lot of simplification of the utils/ directory.


Making it more like what Win32 does makes sense at least on the grounds that 
having three different ways to do the same thing is more crazy-making than 
necessary.  Getting MakeMaker to do the work both inside and outside the core 
also makes sense.  Thanks for peering into this mess.
___________________________________
Craig A. Berry
mailto:craigbe...@mac.com

"... getting out of a sonnet is much more
 difficult than getting in."
                 Brad Leithauser

Reply via email to