[Bioperl-l] github -- pull request for adding bp_ to scripts

[Chris Fields]

(Lincoln, maybe you can add to the below?)

On Jul 25, 2011, at 12:21 PM, Carnë Draug wrote:

> 2011/7/25 Chris Fields <cjfields at illinois.edu>:
>> I responded to that:
>> 
>> http://lists.open-bio.org/pipermail/bioperl-l/2011-July/035425.html
>> 
>> I haven't heard any arguments against it, will merge in today.  We will need to ensure Build.PL is set up so the scripts are installed correctly.
> 
> Oh, you're right. I somehow missed it. Thank you.
> 
> Another thing that I found weird is the extension. Why the .PLS
> extension? It's the first time I see it for a perl script and actually
> confuses my desktop.

I recall there being an argument for this at some point; there is an old post here:

http://thread.gmane.org/gmane.comp.lang.perl.bio.general/880
http://thread.gmane.org/gmane.comp.lang.perl.bio.general/1150/focus=1216

The way I read that, I think the reasoning was to indicate the scripts are templates and the proper perl version would be affixed during build/installation.  Maybe this is now outdated; we have since moved on from ExtUtils::MakeMakes to Module::Build; IIRC the tempting system mentioned in the latter thread still works regardless of the file extension.

> I understand that aside the .pm and .t extension, perl doesn't really
> care and there's no official rule. However, the canonical approach is
> to use .pl for scripts or nothing for applications (and these scripts
> end up being installed as individual applications anyway).
> 
> Carnë

Well, that's not completely true (the 'canonical' bit).  In general it's considered best practice to use *.pl for scripts, *.pm for modules, and *.pod for doc-only (helps with editors, default launch apps, etc), but there isn't any real steadfast rule saying one must use that, or that not doing so will break things. Some installed scripts leave off .pl entirely and just make the script executable. 

chris