--===============0638826156==
Content-Type: multipart/signed;
  boundary="nextPart2299439.KPkQRDKVAf";
  protocol="application/pgp-signature";
  micalg=pgp-sha1
Content-Transfer-Encoding: 7bit

--nextPart2299439.KPkQRDKVAf
Content-Type: text/plain;
  charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

Hi,

 the KDevelop Hackaton is almost over, we will the Trolltech office in=20
Munich soon. Here is my report about what has happened here, how we=20
felt. The report is mainly from Quanta's point of view. A dot story and=20
KDevelop report will follow, I'm sure. :)
 The gathering started with a two days conference, the first day was=20
more user oriented, the second is more technical, maybe a little=20
exception was my talk, as I was not following the KDevelop development=20
lately, so I could just talk about my plans and feeling about the=20
current status of Quanta.
 The hacking itself started on Monday, and will end today, Friday. There=20
were lot of changes made both inside the platform and the applications=20
using it. Work was done on the ui, on cpp support, debugger, plugins=20
and so on. In Quanta I tried to catch up with the latest KDevPlatform=20
changes and fixed the usertoolbars plugin (mostly, some bugs are still=20
present, I'm sure). The plugin loads the toolbars, and their actions=20
work, both from the toolbar and the new Actions menu.
I fixed the broken layout of some other plugins as well, but the=20
functionality was not changed.
 The biggest change happened when Jens Herden joined me for a few days.=20
With him we made some progress again with the parser. The new parser=20
that he and I designed and wrote at my home...almost two years ago, but=20
was never finished. It is a state machine based parser, and the nice=20
functionality is that you can modify the parser's behavior without=20
touching the C++ code (in some cases, especially when introducing new=20
features this might be needed though). This means that parsers for new=20
languages can be added by just writing an XML file. :) This is the=20
theory, right now only the state machine for XML languages is ready.=20
Now a developer who has good knowledge in parsers (at least I'm not an=20
expert in the theory behind the parser, altough I learned some thing=20
many years ago) might shoot us, probably we did some stupid thing even=20
starting with such a parser, but well, the end result is important,=20
not? At this moment I'm happy both with the code and how it works.
 It parses the files amazingly fast, but unfortunately doesn't parse yet=20
CSS, PHP or JavaScript. The node tree building is also very fast. We=20
use the Qt's model classes and build a model directly from the parser=20
result. This means that showing the node tree in the structure view is=20
very fast and does not require too much code. So here is what was also=20
done: the structure tree is now using the new parser. It cannot do=20
much, just clicking takes you to the selected tag, double clicking=20
selects the tag area. Just like in Quanta3. The groups view is not=20
working at this moment.
 The real big change was the removal of the old parser code. Although=20
the Quanta3's parser was ported to Qt/KDE4, it was never working=20
correctly, it would just crash the application. The code is not 100%=20
clean of the old parser, but the parser related files were removed, and=20
in other places where it was used, the code is commented out.=20
 This also means that Quanta right now cannot do anything fancy like=20
code completion, tag editing, error reporting or such. That's it, the=20
transition will be hard there is A LOT to do. To finish this will take=20
a lot of time, especially if I work on it alone. Now I had a motivation=20
to work on it, as we were in a team, but well...this can go away if I=20
will always need to hack alone on it, as it did some times in the past.
 But I'm sure that either we go this way, or there will be no Quanta for=20
KDE4. Keeping the old code and just making to work under KDE4 is=20
possible, but I won't do that. Quanta3 can still be used under KDE4, so=20
in that case, better stick with the latest 3.5.x release.
 What is the summary of this? Help needed. And in sense of C++ coders.
That's about Quanta4, where it is.=20
 KDevelop and KDevPlatform has also a long way to go, but they are=20
progressing well. :)=20
 And last, but not least thanks for Harald for organizing it, it was=20
well organized, I felt good here with him and the rest of KDevelop=20
guys.

Andras

=2D-=20
Quanta Plus developer - http://quanta.kdewebdev.org
K Desktop Environment - http://www.kde.org

--nextPart2299439.KPkQRDKVAf
Content-Type: application/pgp-signature; name=signature.asc 
Content-Description: This is a digitally signed message part.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.4-svn0 (GNU/Linux)

iD8DBQBICNsqTQdfac6L/08RAipuAJ9F7AqnGI8Y4ReuONvkZ7BObx6jegCfUGMk
jrc0HJC6py2BceSxal0aJkU=
=sPqx
-----END PGP SIGNATURE-----

--nextPart2299439.KPkQRDKVAf--

--===============0638826156==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
quanta-devel mailing list
quanta-devel@kde.org
https://mail.kde.org/mailman/listinfo/quanta-devel

--===============0638826156==--