Re: Getting Cruft off of CPANTS

2006-09-25 Thread Thomas Klausner
Hi!

On Sat, Sep 23, 2006 at 03:30:06PM -0600, Randy J. Ray wrote:
 I asked this once before, but rather than getting an answer, someone just
 did it. There are some modules that I've deleted completely from CPAN; one
 was because I renamed the top-level of the namespace (per conversation with
 Adam Kennedy), and the other because someone else has a much better module
 for the task (I've dropped Devel::Coverage in favor of using and maybe one
 day contributing to Devel::Cover).
 
 Should there be a clearer way for module authors to communicate this? Does
 CPANTS recognize when modules disappear?
 
Until some months ago CPANTS had a bug (caused by bugs in the tools
used..) that prevented some dists from beeing deleted. 

In your case (Devel::Coverage) the reason was that there was no new
CPANTS run since you deleted the dist from CPAN.

I did a new run, and now it's gone...


 Over the last few months, CPANTS has been a great resource for me in finding
 shortcomings in my distributions. Test suites and CPAN testers do their
 parts, and CPANTS helps keep the bundling sane.

Nice to hear :-)

-- 
#!/usr/bin/perl   http://domm.zsi.at
for(ref bless{},just'another'perl'hacker){s-:+-$-gprint$_.$/}


Getting Cruft off of CPANTS

2006-09-23 Thread Randy J. Ray

I asked this once before, but rather than getting an answer, someone just
did it. There are some modules that I've deleted completely from CPAN; one
was because I renamed the top-level of the namespace (per conversation with
Adam Kennedy), and the other because someone else has a much better module
for the task (I've dropped Devel::Coverage in favor of using and maybe one
day contributing to Devel::Cover).

Should there be a clearer way for module authors to communicate this? Does
CPANTS recognize when modules disappear?

Over the last few months, CPANTS has been a great resource for me in finding
shortcomings in my distributions. Test suites and CPAN testers do their
parts, and CPANTS helps keep the bundling sane.

Randy
--
Randy J. Ray / [EMAIL PROTECTED]
Campbell, CA


Re: Getting Cruft off of CPANTS

2006-09-23 Thread David Golden

I assume the idea behind not taking things off is to keep the history
available for future reference.  Even if someone deletes an old
tarball of CPAN, I want to see the test history for it.  I look not
only at current versions, but at past versions for judging robustness.

If a modules is renamed or a project is really scrapped, then I can
see pulling it off CPANTS, and that's probably better done by hand
after communication.

Clearer communication instructions would help, though.

Regards,
David

On 9/23/06, Randy J. Ray [EMAIL PROTECTED] wrote:

I asked this once before, but rather than getting an answer, someone just
did it. There are some modules that I've deleted completely from CPAN; one
was because I renamed the top-level of the namespace (per conversation with
Adam Kennedy), and the other because someone else has a much better module
for the task (I've dropped Devel::Coverage in favor of using and maybe one
day contributing to Devel::Cover).

Should there be a clearer way for module authors to communicate this? Does
CPANTS recognize when modules disappear?

Over the last few months, CPANTS has been a great resource for me in finding
shortcomings in my distributions. Test suites and CPAN testers do their
parts, and CPANTS helps keep the bundling sane.

Randy
--
Randy J. Ray / [EMAIL PROTECTED]
Campbell, CA