* The key bindings now display (semi-)correctly.
- Here, "semi-" means the local keys display as a string under X and
a number everywhere else. The number is all but helpful and will
have to be replaced before it is usable, but at the moment, the
code is stable afaik.
* Fixes for compilation on 64-bit architectures :
string position-returning functions return size_t, not uint.
Fix that.
* Fixes warnings with new gcc about non-fixed string AND no arguments.
NOTE : 64-bit architectures still do not compile. There are 2 different places
where printf is used with a specifier that is not large enough on 64-bit because
size_t is now a ulong and not a uint. The solution on GNU systems is the %zu
specifier, but as I'm not sure it is supported on windows I don't want to
introduce a huge hard-to-notice bug, so I'll do it at a time when I got a
windowser to back me up.
* updated daily build
* Card spoiler now sorts by collector's number.
* Metadata looks for "[m" (for speed reasons), I've been using "[meta]" in files. No sets currently use metadata.
Eventually, the WDataSource stuff will need to be analyzed more in depth. WSrcMTGSet is very similar (though simplified) in functionality to the deck data wrapper stuff.
* Fix a bug where valid directories would be ignored and invalid ones
would be accepted for profiles and themes.
* Fix a compilation fault when compiling in debug mode.
- Fix issue 144 (Sound is either 0 or 100%), for PSP ONLY. The methods are now here for linux/windows, but only the music volume method will work currently, and it sets the volume globally. Patch by Yeshua with some cleanup by myself.
- removed some unused code. Please review!
- Added protection from() auto keyword. It is still possible to use protection from [color] in abilities, but when it is not possible, please use protection from([target]) in auto=
The thing complicating a "100% functionality" fix for this is that all translation changes are destructive- they overwrite the original text. We want this behavior, as it prevents having two sets of strings in memory (the untranslated and translated sets) when card text currently takes hundreds of kb... however, to switch languages we'd want the original text available, which means reloading all the set files. That's not optimal either...
As it is, the player must reset their PSP for translated card text. Again, not optimal. I'm submitting an issue on this topic.
* Metadata is currently only used in exactly one place: the set's "Pretty Name" is displayed when the set is first unlocked.
* WGuiImage now has a function to set scaling.
* Confirmable options used to only check for confirmation when selecting another option. They now autoconfirm when saving & exiting. This bug could possibly result in loading the wrong theme. Fixed.
* Changing profiles, canceling, then changing to the profile again could also result in improper theme loading. Fixed.
* The alternate theme is just useful for testing themes, but please comment on the changes to the main theme.
* Options reading and initializing separated from UI.
* Options UI abstracted a bit, made more easily extensible.
* With some extension, the options UI could be used as the basis for more complicated themes, if we load a file and use it to do stuff like GameStateOptions::Start()
-Issue 26: added an option for Mana Display. This needs a bit testing, but it should also lower the priority of "manapool slowness" as this mode probably fixes performance issue as well.
-I can't seem to be able to "save" some options (hand position, mana display) ion the windows version, does this change break something ?
-more JGE cleanup
-removed calls to BindTexture that were made out of the Render methods. Please let me know if this has side effects (and let's discuss it in that case)
-Still no luck with the purple screen bug :(
* Restored OptionItem saving fix.
* Booster duplicate replacement was picking random cards from the wrong set. My fault, but I've fixed it :)
* The problem with cache had nothing to do with memory fragmentation, but I've switched to an array rather than map<> just in case. The actual issue was that in GameStateDeckViewer, I'd given the cache unlimited space, thinking that AttemptNew would recover from any bad_allocs. Unfortunately, the image loading routines and similar stuff called by various implementations of WResource::Attempt() could fail halfway through, leaking memory.
The temporary solution is to set a proper limit (8000000 px, more or less) and (in case we still run out of memory) test to make certain CACHE_SPACE_RESERVED can be malloc'd/free'd. The proper solution would be to keep byte-perfect records of memory used (right now we're kinda fuzzy-- we track pixels per image and bytes per sound, but not the space for jquads or other incidentals) instead of testing a malloc, and potentially cleaning up all calls inside of Attempt() so they fail without leaks. That's what I'm working on now.
Still, it's nice to have identified the problem. This version of the cache should be fully functional, it's just a bit inelegant.
* Numerous cache fixes, reduced filesystem access.
* Cache fails very gracefully.
* Cache is now a templated class, with individual caches per data-type.
* Much easier to extend.
* Extensively debugged. Try compiling with -DDEBUG_CACHE.
* Caches limits can be set on a per-item basis.
* hgeParticleSystemInfo are now cached, mana particles now fall back to defaults.
* Samples are not cached, but track filesystem misses using the cache backbone.
* Avatars are cached. Default baka avatar is now baka.jpg, to prevent collision with player.
A note on the retrieval types:
RETRIEVE_MANAGE puts a resource into a seperate, managed resource list.
Managed resources are guarenteed valid for the lifetime of the program.
Retrieving a managed quad promotes the associated texture to managed. Don't do that by mistake.
Calls to Resources.Refresh() will attempt to reload managed resources in place.
RETRIVE_LOCK (and by extension, RETRIEVE_VRAM), returns a resource after locking it.
A resource may have many locks, and remains in cache until they are all released.
If the resource is managed, it returns it unmodified.
A note on quads:
Unlike all other RetrieveWhatever() functions, the default behavior for RetrieveQuad is RETRIEVE_LOCK. Worse, Release(JQuad*) is slow, and will /always/ release one lock from the associated texture.
There's a long and complicated explanation for this, involving support for live relinking of textures to existing quads, but basically what it means is that we only use RetrieveQuad for quads we intend to store and later Release(). If a temporary quad is needed, the preferred method is to use NEW JQuad* and SAFE_DELETE with RetrieveTexture(). RetrieveTempQuad is also provided, but is only guaranteed until the next call to the cache.
Note that RetrieveCard has none of these problems.
jean.chalard
wagic.jeck
wagic.the.homebrew@gmail.com