kate folding marker patch

[Matthew Woehlke]

Christoph Cullmann wrote:
> On Thursday 08 January 2009 03:20:12 Matthew Woehlke wrote:
>> I'm sitting on a patch for https://bugs.kde.org/show_bug.cgi?id=173695.
>> I was intending to check it in this week, and Dirk just went and
>> branched on me :-). Well, the good news is I can start working on a
>> better fix now ;-).
>>
>> Anyway... there is no way at this point to fix this properly in 4.2
>> (needs UI, and hence string, changes). The a11y implications of the
>> current behavior are not so fun. (I guess what it currently does - i.e.
>> completely ignore kate's colors and use the system scheme - is okay, but
>> responsible for the bug in the first place.)
>>
>> What I'd like to do is make the original background color "correct".
>> However this has the potential to do Horrible Things depending on how
>> the color scheme is configured, so my inclination is to drop the rainbow
>> effect for now and use a straight-up blend from the base background in
>> the direction of the foreground color. This is what I'd check into 4.2,
>> if anything. The alternative is to leave it broken in 4.2.
>>
>> For 4.3 I intend to take a very large axe to the current config dialog
>> and add proper configurability for this, at which point the default will
>> be similar to the current code, but configurable.
>>
>> There is a patch at [1] that may or may not still apply cleanly (one
>> hunk, for a different but related bug, is already committed). It's less
>> eye-candy, but otherwise seems about equally useful (at least to me),
>> and avoids the potential a11y pitfalls.
>
> I can live with that patch, as long as a11y is happy with it, it is ok for me.

a11y I'm more worried about going forward, but if the current behavior 
is okay, I think what I have in mind will be okay also (the plan for 4.3 
is really pretty much "make it customizable"). The 4.2 change should be 
an a11y improvement since it doesn't fiddle with hue or match unrelated 
colors (the particular a11y dangers I'm worried about), but it could be 
considered a feature regression, which is why I'm asking the kate folk.

Anyway, thanks muchly for the reply! (Assuming no one steps up to 
object) I don't think I'll get to this today, but I will try to get it 
checked in by Sunday at the latest.

-- 
Matthew
Please do not quote my e-mail address unobfuscated in message bodies.
-- 
We apologize for the inconvenience. -- God's Last Message to Creation 
(as reported by Douglas Adams in the Hitchhiker's Guide series)