develooper Front page | perl.documentation | Postings from September 2010

Re: [PATCH] Start revamping perlipc.pod

Thread Previous | Thread Next
From:
Leon Timmermans
Date:
September 5, 2010 06:57
Subject:
Re: [PATCH] Start revamping perlipc.pod
Message ID:
AANLkTimvJKsk78Je7VZVBu047UKq6tY0sp6ncpU9oJ22@mail.gmail.com
On Sun, Sep 5, 2010 at 7:03 AM, Shlomi Fish <shlomif@iglu.org.il> wrote:
> Inspired by a message ot the perl documentation proejct, I started working on
> revamping perlipc.pod here:

Good idea :-)

> What I did so far is convert all tabs to spaces (as the indentation was very
> erratic) and started modernising the code (adding line spaces, declare
> variables with my, not cuddle else's, etc.). So far I've reached the named
> pipes section in my code coverage, and I'm planning to convert the socket
> examples to IO::Socket when I get to them.

Some of my ideas and questions for perlipc would be:

* Reorganize the order of chapters. Signals *really* should not be the
first thing on the list. In fact they should probably leave only SysV
IPC behind them. Named pipes should probably also be moved down.
* Convert all glob filehandles to lexical filehandles.
* Convert all open's to the 3+ argument form where possible.
* Make it use more core and CPAN modules instead of reinventing code
(e.g. autodie, IPC::System::Simple, IPC::Signal, IO::Pipe, etc…).
* Are there actually still operating systems out there where perl has
SysV behavior for signal handlers? Probably some rather old
installations of SysV derived unices, but I think this is no longer
relevant except maybe in perlport.
* Should SysV IPC be covered at all? I've never seen a Perl program
use it in the wild. Rightly so if you ask me, because using it from
Perl sucks even more than using it from C.

Leon

Thread Previous | Thread Next


nntp.perl.org: Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at ask@perl.org | Group listing | About