[kdevplatform] d3af706 Increase KDE requirement to 4.4 to fix the build f

[David Nolden]

2010/6/29 Christoph Cullmann <cullmann at absint.de>:
> On Saturday 19 June 2010 13:07:41 Andreas Pakulat wrote:
>> On 19.06.10 12:02:56, Niko Sams wrote:
>> > On Sat, Jun 19, 2010 at 11:45, Andreas Pakulat <apaku at gmx.de> wrote:
>> > > On 19.06.10 08:38:15, Niko Sams wrote:
>> > >> If it is really that easy to use kate git + kde 4.4 I'd say it's no
>> > >> problem to have that as a requirement for master.
>> > >
>> > > Problem is that there's no way to depend on a specific kte version
>> > > (afaik), other than kdelibs version.
>> >
>> > what about adding a KTE version (to git only) - that way kdevelop could
>> > depend on KDE 4.5 _or_ KDE 4.4 + KTE x.x?
>>
>> Well, that would mean kate devs need to maintain a kte-version.h file and
>> we'd have to maintain a FindKTE.cmake file that reads that header, parses
>> the version number and act accordingly. And thats IMHO more work than is
>> justifieable for this temporary situation (we'll depend on KDE 4.5 as soon
>> as its released anyway for master).
> ;) Any updates on the big smart => moving changes?
> If still stuff needs to be patched in kate part, would be good to have some
> feedback until tampere, as there I and Dominik and Co. have time to look after
> missing stuff, regressions and so on.
>
> Beside, other thing: It is required to use the CodeFolding interface version
> 3, the old one used Smart* and that won't work with KDE 4.6 then. (For KDE
> 4.5, it will still work and no stuff will change but just to not let that slip
> apart)

The port is going well, from what I see, there is no additional API
required, just the existing API has to work right. :-) However I'm
still in the "port the code" phase, and didn't have a chance to test
yet.

There is one important thing (in case this hasn't happend yet): All
the kate internal highlighting should use proper z-depth, according to
the depths we once defined on kwrite-devel, and then they should be
documented on the setZDepth function as reference.

Then we need the alpha-blending of overlapping ranges implemented, and
that should be enough.

Greetings, David