On 08/12/2012 10:41 AM, Nils Kneuper wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi everybody!
I would like to release the first 1.11.x release very soonish.
Nice :-)
This means I will soon (probably after the release of 1.10.4) run a pot-update
on trunk so
that those branches start to differ.
My question to all devs involved is: what is the current status of trunk and
is it in shape for a first dev release in say 2 weeks? Meaning Saturday,
August 25th.
Yes, as far as I am aware of, my stuff is ready for a release in two weeks.
I count on getting 20+ fresh bugs assigned to me shortly after the release,
while being confident that there is no show-stopper left and 1.11 is
playable.
Most remaining bugs are left in the new editor features which are not
critical to map making (as long as it happens "conventional").
This of course also considers the work done during GSoC 2012
since that one should be basically done by then (code wise) so the efforts
spent there will show up in a release, too.
No idea about that, wasn't much involved.
Still I would like to see more WML code being introduced (early on) into
the development versions that puts the new features of the gsoc projects
in use.
In the last years most good working features weren't very accessible for
the wml coder at first,
the persistent variables project is an example.
Please tell me what you think of a soon to happen dev release and which issues
you still see.
I don't see any difficulties, please release very soon.
But there is that awesome new map generator. It would be nice to have it
in, to be sure it is part of 1.12 later.
I will write another email to the list about it.
AI already told me in IRC that there might be some issues left
with the map format and (ongoing?) conversions.
Current trunk contains the new map format.
It's mostly bug free and good enough for a development release.
However, on a second thought and while working with the new features, I
came to the conclusion that
not changing the map format at all and keeping the old one is the best
solution.
That means I keep the newly introduced [map] tag referencing to a
savefile, while overwriting the files usage and border_size attributes
if given.
The map file (when handled exactly like it was before my commits on the
issue) itself can't hold extra WML that the new editor features rely on.
Thus the editor saves them in an extra file containing the referencing
[map] tag.
This file is a valid multiplayer scenario on its own but can also be
included by the wml preprocessor to be used in [scenario] or
[multiplayer] files.
Still you have a file containing the pure map for conventional usage.
Being able to save/load both files, the new editor will be able to
produce 1.10 backward compatible maps (if you only use terrain known to
1.10 of course),
a feature I was asked for by many core developers.
This can be done quite fast, I only need to revert my commits (except
for the code that handles [map]) and do a little extra stuff on how the
editor saves and loads files.
If the result is not stable in 2 weeks I will just revert it all and
merge in what others did to the editor's files
(I think most of that was people helping me, thus there is not much to
consider if the thing gets trashed).
Such an action would result in the new editor features not being
save-able but still usable (although only for testing purposes).
All of that is not a big deal in my opinion, the map format changes are
only a very small part of the editor features project.
The editor itself has still some bugs in some of the new tools while
multiple maps are opened.
This includes segfaults.
I am not sure about them.
Those tools don't need to be used to produce a valid map and I like to
get some feedback, especially about bugs.
I leave it to you to decide if 1.11.0 shall include them.
The code does not need to get reverted, it is enough to disable them gui
wise in the editor's theme (WML) file.
Yes, the 1.11.0 release by no means has to be perfect and stable, but it has
to work somehow.
Indeed.
Thanks to Espreon, I have a good summary of what is still going wrong.
I hope to get through that list during the week.
So please tell me about blockers you see and also give me
estimates about fixing them (if you got them).
No blockers left.
There are 3 important bugs whose will disappear when reverting the map
format changes,
which I prefer currently, there are only advantages by doing so.
Cheers,
Nils Kneuper aka Ivanovic
Best regards,
Fabian Müller aka fabi/fendrin
_______________________________________________
Wesnoth-dev mailing list
[email protected]
https://mail.gna.org/listinfo/wesnoth-dev
