#gemrb@irc.freenode.net logs for 26 Apr 2010 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:23:37] <-- tomprince_loki has left IRC (Ping timeout: 265 seconds)
[02:53:53] <CIA-74> GemRB: 03tom.prince * rb03e5f4c5a2a 10gemrb/gemrb/ (10 files in 7 dirs):
[02:53:54] <CIA-74> GemRB: Plugins: Move registrations of cleanup and initializers to PluginMgr.
[02:53:54] <CIA-74> GemRB: This is a preliminary step to enabling static linking of plugins.
[02:53:54] <CIA-74> GemRB: core won't necessarily be initialized when plugin registration is
[02:53:54] <CIA-74> GemRB: happening.
[02:53:55] <CIA-74> GemRB: 03tom.prince * rf46a81c407d1 10gemrb/gemrb/core/ (Interface.cpp PluginMgr.cpp PluginMgr.h):
[02:53:55] <CIA-74> GemRB: PluginMgr: Separate loading of plugins from construction.
[02:53:55] <CIA-74> GemRB: This is in preparation for supporting static linking.
[03:18:01] --> Gekz_ has joined #GemRb
[03:18:04] <-- Gekz has left IRC (Read error: Connection reset by peer)
[03:58:47] --> raevol has joined #GemRb
[04:20:40] <-- Nomad010 has left IRC (Ping timeout: 258 seconds)
[05:49:20] <-- raevol has left IRC (Quit: Leaving.)
[05:50:30] <CIA-74> GemRB: 03tom.prince * r3327c2009fd7 10gemrb/gemrb/includes/plugindef.h: plugindef.h: Document plugin declaration macros.
[05:51:09] <tomprince> Is the difference between bcs and bs files *exactly* the same as the difference between scripts in $ROOT/scripts and scripts in override/chitin.key?
[05:55:47] --> Gekz has joined #GemRb
[05:58:42] <-- Gekz_ has left IRC (Ping timeout: 276 seconds)
[06:51:58] --> barraAway has joined #GemRb
[07:18:45] --> Maighstir_laptop has joined #GemRb
[07:57:18] <-- kahen has left IRC (Ping timeout: 276 seconds)
[08:07:31] <-- Gekz has left IRC (Quit: This computer has gone to sleep)
[08:13:45] <-- Maighstir_laptop has left #GemRb
[08:14:17] --> kahen has joined #GemRb
[08:19:59] <fuzzie> tomprince: i think it is more complicated, but i can't find where it was discussed.
[08:25:43] <fuzzie> also wouldn't be surprised if it's different in eg. iwd2 and we didn't notice.
[08:36:33] --> Gekz has joined #GemRb
[08:47:33] <-- |Cable| has left IRC (Remote host closed the connection)
[10:15:27] --> Nomad010 has joined #GemRb
[10:50:39] <-- Nomad010 has left IRC (Remote host closed the connection)
[10:50:52] --> Nomad010 has joined #GemRb
[10:59:14] <-- Gekz has left IRC (Ping timeout: 240 seconds)
[11:13:44] --> Gekz has joined #GemRb
[11:53:39] --> nomad__ has joined #GemRb
[11:57:36] <-- Nomad010 has left IRC (Ping timeout: 258 seconds)
[12:14:16] <-- nomad__ has left IRC (Ping timeout: 245 seconds)
[12:16:14] <-- Gekz has left IRC (Ping timeout: 246 seconds)
[12:49:46] --> lynxlynxlynx has joined #GemRb
[12:49:46] --- ChanServ gives channel operator status to lynxlynxlynx
[12:54:01] <lynxlynxlynx> tomprince: + * entry points to the a plugin,
[12:54:59] <lynxlynxlynx> + * Registers a resource through ResourceManger.
[12:55:32] <lynxlynxlynx> + * @param[in] ext Extention of resource files.
[14:44:52] <tomprince> lnyxlynxlynx: Feel free to fix it.
[14:56:16] <lynxlynxlynx> ok
[15:02:24] <tomprince> fuzzie: My thought was to change ai script lookup to use a Directory Importer in ../scripts, and remove scripts from the override path.
[15:02:48] <tomprince> And then at some later point, change scripts to be resource, with both exts.
[15:06:13] <fuzzie> i think that would be game-specific
[15:06:50] <fuzzie> the scripts/ directory being only important to the UI?
[15:09:39] <fuzzie> oh, i see, our SClassID.h has this stupid hack for BS files.
[15:09:52] <fuzzie> that should be 0x3F9.
[15:10:05] <CIA-74> GemRB: 03lynxlupodian * r15d3bf00ee79 10gemrb/gemrb/includes/plugindef.h: plugindef.h: fixed syntax errors in the macro descriptions
[15:15:06] <tomprince> bg2/iwd/pst seem to have only bcs files in override/bifs
[15:16:52] --> raevol has joined #GemRb
[15:17:16] <fuzzie> they get biffed by mods/etc
[15:17:54] <fuzzie> we had some discussion about this .. somewhere
[15:18:47] <fuzzie> but i couldn't find it with some application of google
[15:19:31] <fuzzie> and i've completely forgotten how it works at all
[15:25:59] <fuzzie> yes, i still can't find it. grr.
[15:28:36] <fuzzie> weidu and NI and etc do have the id, though
[15:28:59] <fuzzie> so i am not just going mad
[15:31:12] --> barra_away has joined #GemRb
[15:31:13] <edheldil> id for baf files?
[15:32:19] <fuzzie> no id for baf files :)
[15:32:59] <fuzzie> but they have 0x3F9 for .bs, as opposed to 0x3EF for .bcs
[15:33:48] <-- barraAway has left IRC (Ping timeout: 258 seconds)
[15:35:28] <tomprince> On epoint of data on mods: my current installs of bwp/ca don't have any bs files in override.
[15:38:15] <-- raevol has left IRC (Read error: Connection reset by peer)
[15:40:18] <fuzzie> well, it is easy enough to test in the original engines
[15:40:40] <fuzzie> i imagine iwd2, pst and bg2 would be sufficient
[15:41:07] <fuzzie> although it seems to involve working out exactly how the slots work, and the notes out there are a bit inconsistent on that point
[15:42:53] <fuzzie> but there are plenty of mods out there doing things which just don't work, and obviously a lack of a mod doing something isn't any information at all, so trying to debug by inspection of mod files is not really something you can conclude anything from, in my experience
[15:43:26] <fuzzie> in particular i spent a while trying to fix up iwd2 things which turned out to never work in the original, people just copying things from bg2 and assuming they'd work
[15:53:10] --> Gekz has joined #GemRb
[16:16:45] <tomprince> I think the only build system changes needed for static linking is making PluginMgr a singleton, and adding and ifdef in plugindef that activates switches the macros around.
[16:17:33] <tomprince> I was wondering if I should just push those changes to trunk. It'd make it fairly easy to test with msvc6, for example/
[16:20:59] --> tomprince_loki has joined #GemRb
[16:23:12] <fuzzie> well, it doesn't seem likely to break anything?
[16:25:41] <tomprince_loki> No, it doesn't seem likely.
[16:55:08] <tomprince_loki> We seem to have a few variables that are only used in asserts.
[17:07:37] <tomprince_loki> or.cz/static-link: Code changes, proof-of-concept-static-link: hack-n-slash cmake build system (needs GNU ld).
[17:14:03] <fuzzie> is that PluginMgr guaranteed to be allocated at the point where you use that singleton?
[17:15:12] <tomprince_loki> It should be, function local statics are intialized when code first touches the declaration.
[17:18:18] --> |Cable| has joined #GemRb
[17:19:13] <fuzzie> and nothing problematic tries using it at destruction time, i assume..
[17:20:34] <fuzzie> static-link seems fine, then
[17:20:42] <tomprince_loki> No, we don't have anything running at destruction time. Except now ~PluginMgr. If ever we do, then we would need something more complicated.
[17:22:20] <fuzzie> quite neat, really :)
[17:23:45] <CIA-74> GemRB: 03tom.prince * rb6c112227e38 10gemrb/gemrb/plugins/CREImporter/CREImporter.cpp:
[17:23:45] <CIA-74> GemRB: CREImporter: Move code from static initialization to plugin initialization.
[17:23:45] <CIA-74> GemRB: core won't necessarily be around at static initialization time.
[17:23:47] <CIA-74> GemRB: 03tom.prince * r08944f092231 10gemrb/gemrb/core/ (5 files):
[17:23:47] <CIA-74> GemRB: PluginMgr: Make this a singleton.
[17:23:47] <CIA-74> GemRB: This is in preparation for static linking.
[17:23:49] <CIA-74> GemRB: 03tom.prince * r18b7d30a132b 10gemrb/gemrb/includes/plugindef.h:
[17:23:49] <CIA-74> GemRB: plugindef: Add code path for use when static linking.
[17:23:49] <CIA-74> GemRB: None of the build systems use this yet. If a plugin is compiled
[17:23:49] <CIA-74> GemRB: with STATIC_LINK defined, and included in the main executable,
[17:23:49] <CIA-74> GemRB: then things should just work.
[17:24:30] <tomprince_loki> Getting the build system to handle isn't quite straight forward, but it does just work. I haven't tried mixed static/dynamic, but that should just work as well.
[17:25:23] <tomprince_loki> I don't know why you would ever want to, but it wouldn't be too hard to support loading plugins at any point in time now, either.
[17:59:28] <lynxlynxlynx> [ResourceManager]: Searching for Tried CIMF4INV.plt [ERROR] <-- looks odd since there is no filename
[18:07:49] <CIA-74> GemRB: 03tom.prince * rd74803e98ff6 10gemrb/gemrb/core/ResourceManager.cpp: ResourceManager: Add missing print of filename in Resource case of GetResource.
[18:07:57] <tomprince_loki> That compiles anyway. :)
[18:11:44] <lynxlynxlynx> works too
[18:12:24] <lynxlynxlynx> not sure what's with these inventory palletes
[18:12:54] <lynxlynxlynx> at least 1pp has them
[18:12:54] <tomprince_loki> plt aren't palettes, they are images that get a palette applied to them.
[18:14:05] <tomprince_loki> They are a weird image format that each pixel has a byte for palette, and a byte for intensity.
[18:14:19] <lynxlynxlynx> oh, it's from bg2
[18:15:47] <tomprince_loki> And I think that 1pp either zeros them out, or stores a bam in there, since the thought is the original graphics are nicer. Plus, I don't think there exists tools to edit plts. :)
[19:22:53] --> Edheldil_ has joined #GemRb
[20:28:03] <-- tomprince_loki has left IRC (Ping timeout: 265 seconds)
[20:28:32] <-- tomprince has left IRC (Ping timeout: 265 seconds)
[20:36:21] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[20:37:07] --> lynxlynxlynx has joined #GemRb
[20:37:08] --- ChanServ gives channel operator status to lynxlynxlynx
[20:45:43] --> tomprince has joined #GemRb
[20:54:53] --> tomprince_loki has joined #GemRb
[21:31:07] --> cfchris6_ has joined #GemRb
[21:33:38] <-- cfchris6 has left IRC (Ping timeout: 240 seconds)
[21:56:41] <-- Edheldil_ has left IRC (Remote host closed the connection)
[22:03:17] --> Nomad010 has joined #GemRb
[22:05:16] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[22:34:53] <-- cfchris6_ has left IRC (Quit: leaving)
[23:10:27] <CIA-74> GemRB: 03tom.prince * r657ccd80926a 10gemrb/gemrb/ (core/VFS.h plugins/GUIScript/GUIScript.cpp):
[23:10:27] <CIA-74> GemRB: PathAppend: Get rid of only use outside of VFS.
[23:10:27] <CIA-74> GemRB: Signed-off-by: Tom Prince <tom.prince@ualberta.net>
[23:10:28] <CIA-74> GemRB: 03tom.prince * r896991b2db9e 10gemrb/gemrb/ (4 files in 4 dirs):
[23:10:28] <CIA-74> GemRB: PathJoin: Be more consistent about using this.
[23:10:28] <CIA-74> GemRB: This is in preparation for moving the CaseSensitive logic
[23:10:29] <CIA-74> GemRB: from ResolveFilePath to PathJoin.
[23:10:29] <CIA-74> GemRB: Signed-off-by: Tom Prince <tom.prince@ualberta.net>
[23:10:31] <CIA-74> GemRB: 03tom.prince * r7700e87e55bb 10gemrb/gemrb/core/ (SaveGameIterator.cpp VFS.cpp VFS.h):
[23:10:31] <CIA-74> GemRB: PathJoinExt: Add function to call PathJoin with a file and extension.
[23:10:31] <CIA-74> GemRB: Signed-off-by: Tom Prince <tom.prince@ualberta.net>
[23:10:33] <CIA-74> GemRB: 03tom.prince * ra7c1d826c87d 10gemrb/gemrb/ (12 files in 6 dirs):
[23:10:33] <CIA-74> GemRB: PathJoin: Make PathJoin automatically resolve case, as it builds up the path.
[23:10:33] <CIA-74> GemRB: This make ResolveFilePath obsolete, except for getting paths from the config files.
[23:10:33] <CIA-74> GemRB: Signed-off-by: Tom Prince <tom.prince@ualberta.net>
[23:10:42] <CIA-74> GemRB: 03tom.prince * r80979b41a593 10gemrb/gemrb/core/ResourceManager.cpp:
[23:10:42] <CIA-74> GemRB: ResourceManager: Don't assume that resource names are limited to 8 characters.
[23:10:42] <CIA-74> GemRB: Using resource manger for sounds, we pass in pathnames, rather than filenames,
[23:10:42] <CIA-74> GemRB: so they can be longer than 8 characters.
[23:10:43] <CIA-74> GemRB: 03tom.prince * r1669ac8b2431 10gemrb/gemrb/ (7 files in 4 dirs):
[23:10:43] <CIA-74> GemRB: MUSImporter: Switch to using DirectoryImporter, rather than opening files directly.
[23:10:43] <CIA-74> GemRB: This is in preparation for turning SoundMgr into a Resource.
[23:10:43] <CIA-74> GemRB: Signed-off-by: Tom Prince <tom.prince@ualberta.net>
[23:10:44] <CIA-74> GemRB: 03tom.prince * rd2b6b668f032 10gemrb/ (37 files in 10 dirs):
[23:10:44] <CIA-74> GemRB: Resource support for sound.
[23:10:46] <CIA-74> GemRB: Signed-off-by: Tom Prince <tom.prince@ualberta.net>
[23:10:55] <CIA-74> GemRB: 03tom.prince * rc57ac09f34b0 10gemrb/ (53 files in 15 dirs): Merge branch 'sound'
[23:24:11] --> Gekz_ has joined #GemRb
[23:24:17] <-- Gekz has left IRC (Read error: Connection reset by peer)
[23:32:18] <-- barra_away has left IRC (Quit: Verlassend)
[23:43:32] <CIA-74> GemRB: 03tom.prince * rf6cfee258dd4 10gemrb/gemrb/core/ActorBlock.cpp:
[23:43:32] <CIA-74> GemRB: Prevent an unused variable warning when using cmake in Release mode.
[23:43:32] <CIA-74> GemRB: cmake defines NDEBUG, which disables asserts, making some variables unused.
[23:43:32] <CIA-74> GemRB: Signed-off-by: Tom Prince <tom.prince@ualberta.net>