mail at svenbrauch.de
Fri Jul 28 19:10:44 UTC 2017
On 28/07/17 19:44, Nicolás Alvarez wrote:
> We should either do a real effort to make kdevplatform usable and
> good as a "framework to make an IDE", or stop pretending it is.
I don't think we have the manpower to do that, tbh.
It's true, plasmate uses kdevplatform. Realistically, you want to
install kdevelop anyways though: the qmljs plugin is there and you
probably want to have that ...
So I'd say if we merge kdevplatform into kdevelop, plasmate can just
stay as it is, it just needs to say find_package(kdevelop) instead of
find_package(kdevplatform). IMO that's good enough for this kind of
thing. Outside of our community, I really don't see people starting to
write IDEs based on kdevplatform, they could have done that for the last
ten years and nobody ever did.
Regarding merging kdev-python and kdev-php into the kdevelop repo, for
me this is less a philosophical and more a pragmatic question:
effectively, we have released them all synchroneously for several years
now. I don't see much gain of having them separated, but merging them
has several advantages:
- users cannot not install one of them by accident and then ask what to
do to get e.g. python support working
- users cannot install incompatible versions, also when building themselves
- developers cannot break API in kdevplatform and forget about updating
- it makes creating release tarballs a good bit easier
> You're right, but do we have enough manpower?
We'll manage if we decide we want to do it. Picking a suitable time to
do it certainly helps. I'm fairly sure it saves lots of time in the long
That said, I don't have a super strong opinion on any of this, if many
people think it's a bad idea I'm also fine with everything staying as it is.
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 833 bytes
Desc: OpenPGP digital signature
More information about the KDevelop-devel