Skins

[Frédéric Buclin]

Le 29. 12. 15 14:34, Gervase Markham a écrit :
>> "High cost" where? How often did you, as a developer, have to deal with
>> the fact that we ship 2 skins? 
> 
> Well, when Dylan is working on porting the bug modal UI, he's going to
> need to deal with it. :-)

This doesn't mean the cost will be high. ;)


> Give me a quick refresher: you are saying that if someone invokes a
> custom skin, we load _both_ the Classic and the custom skin's CSS?

Yes, because the custom skin only contains changes made against the
Classic skin. It doesn't need to contain all rules. This is why writing
a new skin is easy: you just edit the few rules you want to change, such
as the background color. For instance, standard/buglist.css is 9.6KB
while contrib/Dusk/buglist.css is only 386 bytes. And
standard/global.css is 19KB while contrib/Dusk/global.css is only 3.7KB.

So even if dylan has to rewrite CSS files heavily in standard/, he only
has to edit a few rules in contrib/ assuming they are still relevant.
This is much less painful than to have to maintain the whole set of
rules again.


> Let's say for a moment that the group concludes you are right. How does
> Sandstone fit in? Do we want it, or not? If we do, do we ship 3 skins,
> or do we drop one? If we drop one, which?

We certainly want Sandstone, and it certainly could become the new
default skin. As you said, Dusk replaced Classic as the default skin
since Bugzilla 3.2 and so we don't want Classic back as the default
skin. I gave a quick look at the 40 Bugzilla installations I'm watching,
and many of them use their own skin by default, or Dusk (for those
running 4.4 and newer) or Classic (for those running 4.2 and older, i.e.
"obsolete" installations). So if my understanding is correct, there is a
move towards Dusk. So my proposal would be: replace Classic by
Standstone, and make it the new default skin for Bugzilla 6.0 (if done
on time for 6.0, of course). This way 1) skins/standard/ would really be
the place for the default skin, which makes sense, and 2) we could still
offer an alternative to users, aka Dusk (but we would need to see how
the final Standstone implementation would look like to know what to
change in Dusk), and 3) we don't need to support an extra 3rd skin.

LpSolit