Follow-up Comment #11, patch #1134 (project wesnoth):
Attached the raw results of my script with the animation filter activated
(but not the suffix filter, because few base frames use one)
See out-of-hex.txt
(file #5508)
___
Additional
URL:
http://gna.org/bugs/?13262
Summary: castle_locs function not found
Project: Battle for Wesnoth
Submitted by: seahen
Submitted on: Thursday 03/26/2009 at 07:32
Category: Bug
Severity: 3 - Normal
Follow-up Comment #9, patch #1139 (project wesnoth):
Don't worry, I have read your original submission. It explains why you did it
that way, but it doesn't explain why it couldn't be done another way.
For instance, there is an ambiguity caused by t_string::operator=(std::string
const ) and you
URL:
http://gna.org/patch/?1141
Summary: Graphics optimization
Project: Battle for Wesnoth
Submitted by: cjhopman
Submitted on: Thursday 03/26/2009 at 07:51
Priority: 5 - Normal
Status: None
Update of patch #1141 (project wesnoth):
Assigned to:None = cjhopman
___
Reply to this item at:
http://gna.org/patch/?1141
___
Message sent
Follow-up Comment #1, patch #1141 (project wesnoth):
Overall, the patch looks fine, a quadtree is indeed the way to go in my
opinion. A few things I have noticed:
First (but I guess this one you already know), use a more compact
representation for quadtree::tree.
Second, your halting condition
Update of bug #13261 (project wesnoth):
Assigned to:None = dragonking
___
Reply to this item at:
http://gna.org/bugs/?13261
___
Nachricht
Follow-up Comment #1, bug #13262 (project wesnoth):
castle_locs appeared in r33712.
You've got r33557. So 'it is not in your version'.
___
Reply to this item at:
http://gna.org/bugs/?13262
Follow-up Comment #2, bug #13262 (project wesnoth):
Documentation in the wiki
http://www.wesnoth.org/wiki/FormulaAI_Functions#.27castle_locs.27_function is
based on 1.6a - so the wiki is correct (castle_locs is already present there),
you are using a slightly outdated version.
Update of bug #13258 (project wesnoth):
Status:None = Need Info
___
Follow-up Comment #1:
Hmm, this one is *really* problematic. That is:
1) At least in my version of glibc (version
Follow-up Comment #10, patch #1139 (project wesnoth):
I'll try removing t_string::operator=(std::string const ) ... I just wanted
to try to not touch t_string, if possible. If that won't work I have at least
one other I want to try to get rid of the suffixes.
Follow-up Comment #2, patch #1140 (project wesnoth):
Now, I couldn't find a translated string (in game or in help with
a hypertext link) where the word wrapping was broken (tested
Japanese and Korean). I know that the bug is there and I suppose
that I could recreate a case, but it would be
Follow-up Comment #2, patch #1141 (project wesnoth):
Yeah, there is a lot of wasted space int quadtree::tree. It was just easier
to do it this way for testing. That will definitely be changed.
You are right that + 1 would make much more sense for the halting
condition. And why 48? No reason,
Follow-up Comment #3, patch #1141 (project wesnoth):
Not sure how you got the idea that SDL rectangles were inclusive, but still
you made me doubt, so I went to check SDL's source code. Therefore I can
confirm: SDL rectangles are not inclusive and you are losing the last row and
last column. No
URL:
http://gna.org/bugs/?13267
Summary: Campaign - 'Towards Mountains of Haag':
Project: Battle for Wesnoth
Submitted by: None
Submitted on: Thursday 03/26/2009 at 18:56 CET
Category: Bug
Severity: 4 -
Follow-up Comment #4, patch #1141 (project wesnoth):
You're right, what I was saying about SDL rectangles didn't really make any
sense.
Doing some testing, it seems that it is very slow when we have the rectangle
{0, 0, w, h}. I do not know why. Making it a pixel smaller in any direction
makes
Update of bug #13252 (project wesnoth):
Status: Ready For Test = Fixed
___
Follow-up Comment #6:
Ok, I see now that transfering AI sides from non-host players was not
implemented yet which
Follow-up Comment #5, patch #1141 (project wesnoth):
I can at least explain why SDL_Flip doesn't make it any faster; it's because
it's equivalent code-wise to updating the whole rectangle when there is no
double buffering, which there usually isn't.
I guess it means you should split the screen
Follow-up Comment #6, patch #1137 (project wesnoth):
(21:49:10) Crab_: boucman, Mordante said two days ago: Crab_, patch looks
good to me, only the doxygen comment should have a @params part, but I can do
that. I let Dragonking have a look since he's more familiar with that code and
then I'll
Update of bug #13267 (project wesnoth):
Status:None = Need Info
___
Follow-up Comment #1:
Where's the screenshot? Afraid this report isn't much help without it, or at
least without the
URL:
http://gna.org/bugs/?13268
Summary: save corruption through undo/redo of recalls
Project: Battle for Wesnoth
Submitted by: soliton
Submitted on: Thursday 03/26/2009 at 23:34
Category: Bug
Severity: 3
Update of bug #13268 (project wesnoth):
Assigned to:None = jhinrichs
___
Reply to this item at:
http://gna.org/bugs/?13268
___
Nachricht
Follow-up Comment #11, patch #1139 (project wesnoth):
Apparently t_string::operator=(std::string const ) wasn't used anywhere
since there was no problem removing it.
Thus t_str() and it's uses were also removed.
The proxy string internally now stores a simple ref; so runtime overhead
shouldn't
Additional Item Attachment, bug #13267 (project wesnoth):
File name: ScreenCantSave1.pngSize:971 KB
___
Reply to this item at:
http://gna.org/bugs/?13267
___
Message sent
Follow-up Comment #2, bug #13267 (project wesnoth):
File as been attached.
___
Reply to this item at:
http://gna.org/bugs/?13267
___
Message sent via/by Gna!
http://gna.org/
25 matches
Mail list logo
