Message-ID: <17187.63016.726744.91386@celebrian.ci.tuwien.ac.at>
Date: 2005-09-11T09:17:28Z
From: Friedrich Leisch
Subject: Issue tracking in packages [was: Re: [R] change in read.spss, package foreign?]
In-Reply-To: <17187.4779.96104.903626@stat.math.ethz.ch>
>>>>> On Sat, 10 Sep 2005 19:06:51 +0200,
>>>>> Martin Maechler (MM) wrote:
>>>>> "TL" == Thomas Lumley <tlumley at u.washington.edu>
>>>>> on Sat, 10 Sep 2005 09:32:29 -0700 (PDT) writes:
>>> Standard location or a mechachanism like the one you
>>> describe are both similar amount of work (and not much at
>>> all), the HTML pages are generated by perl and I have the
>>> parsed DESCRIPTION file there, i.e., using a fixed name
>>> or the value of the Changelog field is basically the
>>> same.
>>>
TL> In which case a Changlog entry in DESCRIPTION would be a
TL> very nice addition, and would have the advantage of not
TL> requiring changes to packages.
> yes *and* does allow slightly more flexibility with almost
> no cost, as Fritz confirmed.
Well, as Kurt pointed out in another (?) thread "CRAN is not the R
universe", and, e.g., Seth might have another opinion when it comes to
BioC administration. But I don't think you can (or should) do too much
sensible computations on packages without having parsed the
DESCRIPTION file, so the "almost no cost" statement should be pretty
safe.
> And, BTW, Gabor, NEWS and ChangeLog are not at all the same
> thing and it would be silly to urge users to one of them.
> At least 'ChangeLog' is a well defined format for emacs users
> that can very quickly be updated semi-automagically
> ("C-x 4 a" when you're in file foo.R with function myfun(.)
> autogenerates a neat entry in a ChangeLog file);
> but then really people should be allowed to use other formats
> for good reasons.
I fully agree.
.f