[rkward-devel] [rkward-users] Call for testing: RKWard 0.5.3-pre1 / Release schedule

[Thomas Friedrichsmeier]

On Wednesday 21 April 2010, Prasenjit Kapat wrote:
> BTW, is kate's 'Filesystem Browser' different then? This doesn't seem
> to happen there. (Its 'tab bar' is completely different, a bit weired,
> if I may.)

Good point. The solution is that kate simply refuses to open the same file 
twice (and in RKWard we were trying to do the same, but there was a bug in 
that). So that hides the bug in kdelibs, and of course we can do the same 
thing (that simply did had not occurred to me, when I wrote this was hard to 
work around).

There is a real bug with respect to that in kdelibs, though.

> Thanks. This looks and works great now, very much usable. The 'icon
> view' is gone now? (I won't miss it!)

Yes, as a side-effect of properly saving and reading the configuration.

> > Indeed. I see the problem, but don't know how to fix this, yet (other
> > than to assign different shortcuts). BTW, the katepart is not to blame.
> > It's the tab- bar that assigns shortcuts to the tabs, apparently without
> > taking existing shortcuts into account.
> 
> Hmm.. well... at least now the shortcuts are configurable.

Ok, I found a way to work around that (and I submitted a patch to kdelibs), so 
this will not be an issue in the release.

> It may get a bit confusing, but far better than messing up the code!
> It is one thing to see this happening in a blank script file and whole
> another thing when it happens is a big R code and you save it w/o
> realizing what was modified... A wild goose chase, when the code
> doesn't run the next day!  So, yeah fine by me.

Ok, I think I'll leave that for 0.5.4, though.

> Someday someone may ask why the two different (Alt & Ctrl)  window
> cycles? We'll have to come up with some answer, till then....

Well, by then, I hope to have implemented an Alt+Tab-like window cycling. Once 
we have that, at least the Alt-type cycle should be pretty much obsolete (and 
possibly the Ctrl-type cycle as well).

> Updated all_plugins.R using sink(). Do check it for any obvious errors.

Looks good. I re-arranged a bit to make it easier to see the result itself, 
and added some more versioning info.

> Although, for some reason I never liked the minimalistic menu of
> chooseCRANmirror(), so I took the liberty to modify the
> rk.select.CRAN.mirror() in public.R. This is basically, the
> chooseCRANmirror function, but with the only difference of paste(m[,
> 1L], m[, 5L], sep = " - ") instead of m[, 1L] (you can see it in the
> cvs list).

Of course another option would be to modify chooseCRANmirror() itself. Not 
sure that is worth the trouble, though.

> Of course, if you feel that this will not get due testing before the
> release then, I can revert back to the old chooseCRANmirror() and we
> can get back to this after the release.

Looks safe to me, so let's keep it. Besides, while testing your change, I 
found two neighbouring bugs, in internal.R... Time for a -pre3 release, I 
guess.

Regards
Thomas
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.kde.org/pipermail/rkward-devel/attachments/20100421/fe6cb9fb/attachment.sig>