Nils Kneuper <[EMAIL PROTECTED]>:
> Yesterday I have tagged 1.3.16 (1.4-beta3) and I want this to be the
> last beta for 1.4. If nothing goes wrong, I want to tag rc1 on
> sunday, rc2 at the end of FOSDEM and about a week later the
> hopefully last release candidate (rc3) (maybe directly 1.4, not
> sure, depends on the amount of changes...). So basically I want to
> switch from the bi weekly schedule for beta releases to a weekly
> schedule for the rcs until 1.4 is out. Anyone against this? Does
> anyone see any blockers for such a schedule?
Seems good to me.
The only bug I'm really worried about at this point is the persistent
replay corruption. Does anyone think they have a clue about what's
causing this?
The only things I have left to do before 1.4 are cleanup. I have
only one pending unfixed bug, and the fix for that will be to deploy
a new build system in 1.5.
As you know, the last improvement in wmlscope flushed out a
significant number of animation bugs; I've still working on cleaning
up after that. (One side effect is that I now understand how to read
and compose animations.)
Otherwise, until end of feature freeze, I plan mainly to continue
working on tightening up my WML sanity checks in wmlscope and wmllint.
I figure every time I do that I prevent entire classes of bugs from
ever shipping (e.g. no ordinary user will ever again see an image missing
because of a developer typo) so that'a a more effective way for me
to bulletproof 1.4 than chasing any of the 42 remaining individual bugs.
If anyone else can suggest additional WML consistency and sanity
tests, I'm listening. Zookeeper did pretty well on the last round, I
was able to implement 3 of his 8 possibilities (others required things
like parsing after macroexpansion which is not practical). One of his
suggestions led to argument type checking for macro calls: 86%
of the macros in mainline are now typechecked (which means, for
example, that wmlscope can detect attempts to pass a string literal
where an integer is wanted, or vice-versa). I will continuew to improve
that percentage.
I will of course continue to fix glitches like bug #11044 (NR: Dwarf
lord Hamal has no crown under orb) in the campaigns I'm co-responsible
for.
--
<a href="http://www.catb.org/~esr/";>Eric S. Raymond</a>
_______________________________________________
Wesnoth-dev mailing list
[email protected]
https://mail.gna.org/listinfo/wesnoth-dev
