#gemrb@irc.freenode.net logs for 5 Nov 2011 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:02:22] --> brad_a_ has joined #gemrb
[00:03:59] <-- brad_a has left IRC (Ping timeout: 248 seconds)
[00:04:00] --- brad_a_ is now known as brad_a
[00:32:45] <-- brad_a has left IRC (Ping timeout: 259 seconds)
[00:52:13] --> brad_a has joined #gemrb
[00:52:47] <-- Maighstir has left IRC (Quit: .)
[01:02:51] <-- brad_a has left IRC (Quit: brad_a)
[01:56:48] <chiv> small patch to fix use item paging when item list exceeds 12 http://pastebin.com/XW95JbLP
[01:57:48] <chiv> its quite hard to achieve that feat but hey
[02:01:38] --> Drakkar has joined #gemrb
[02:04:03] <-- PixelScum has left IRC (Ping timeout: 258 seconds)
[02:16:05] <tomprince> chiv: It seems like it always displays both arrows, if there are more items than will fit, rather than when there are objects scrolled off in the given direction.
[02:16:29] <tomprince> It seems like in the majority of cases, there will be less items on the screen at once.
[02:16:54] <chiv> it shouldnt - it should only be showing when there are more than 12
[02:19:34] <chiv> http://imageshack.us/photo/my-images/42/screenshotiho.png/
[02:19:41] <tomprince> Yes, but when there are, it seeems (I haven't tested this) that the current code will show 11, but your code shows 10.
[02:19:49] <chiv> http://imageshack.us/photo/my-images/42/screenshot1msn.png/
[02:20:40] <chiv> I think the left button is always there for a reason
[02:20:51] <chiv> so you dont accidentally page left and use an item
[02:21:03] <chiv> that seems to be the way spells work
[02:32:15] <chiv> so yeah, it does indeed show 10 items if paging is on
[02:34:57] <tomprince> Well, I'll leave others to comment on it. I haven't played much since I stated developing. :P
[02:35:29] <chiv> hah, very true
[02:36:23] <chiv> maybe I should carry on playing
[02:36:30] <tomprince> There shuold at the very least be a comment about the use of more as bit field. Or perhaps better, use two variables.
[02:37:12] <chiv> well, I just copied the way the spell list did it - not my method :)
[02:38:45] <chiv> i am a keep it simple type
[03:45:06] --> brad_a has joined #gemrb
[03:59:02] <-- brad_a has left IRC (Quit: brad_a)
[05:33:26] <chiv> I really dont get what bioware were thinking... if the amount of inventory items can be infinite, why did they choose to arbitrarily limit how many slots you have when the inventory is already limited by weight
[08:34:19] --> lynxlynxlynx has joined #gemrb
[08:34:19] <-- lynxlynxlynx has left IRC (Changing host)
[08:34:19] --> lynxlynxlynx has joined #gemrb
[08:34:19] --- ChanServ gives channel operator status to lynxlynxlynx
[08:54:49] <chiv> lynxlynxlynx: been working on a little quick load screen for the tob start screen
[08:55:38] <lynxlynxlynx> screen? a button is enough
[08:55:57] <chiv> well, i was going to have a button, then i decided up and down buttons would be nice
[08:55:59] <lynxlynxlynx> or two if you mean the game choice one
[08:56:22] <lynxlynxlynx> that's not much different than the load window, why not just add quickload buttons?
[08:56:29] <chiv> when im finished you can load any game from the start game
[08:56:37] <chiv> ^ screen
[08:56:45] <chiv> default to the last one
[08:56:51] <chiv> so thats the function of both
[08:57:40] <chiv> anyway i was just trying to get to grips with python
[08:58:22] <lynxlynxlynx> :)
[09:12:09] --> Yoshimo has joined #gemrb
[09:21:38] <-- chiv has left IRC (Ping timeout: 265 seconds)
[09:55:29] --> chiv has joined #gemrb
[09:55:52] * chiv was killed by power cut
[09:57:02] <chiv> i've got the basic functionality of the quick load screens running properly, but i cant get the game to assign the right list of save games for soa/tob
[09:59:39] <lynxlynxlynx> it probably relies on oldgame
[10:00:10] <Yoshimo> chiv what youre working on this time?
[10:00:29] <chiv> changing the start screen to be more useful
[10:00:53] <chiv> i set oldgame = 0 before trying to get the tob saves
[10:01:19] <chiv> but i get soa saves
[10:01:42] <chiv> then i go into tob mode, go back, i have tob saves in both areas
[10:01:46] <chiv> odd
[10:04:47] <lynxlynxlynx> check what other variables we save there
[10:05:49] <lynxlynxlynx> i've got the sorting /sorted/ out :)
[10:05:59] <lynxlynxlynx> now the tedious control stuff
[10:11:58] <chiv> nice
[10:12:12] <chiv> hah, now I have a saev game in the list, which I cant find on my hd
[10:20:00] <chiv> hm, gemrb head seems to think tob games are in mpsave
[10:24:24] <lynxlynxlynx> they are
[10:24:49] <chiv> ah
[10:49:16] <chiv> this is all sort of strange o_O
[10:50:01] <chiv> I can get it to work, if I make a button that only refreshed the tob saves list
[10:50:19] <chiv> but if I try to assign them at the same time, they invariably end up identical
[10:52:00] <chiv> maybe python is trying to be clever and save time by passing the same result for the same function
[10:52:49] <chiv> or its just pixies have urinated in my tea and I am experiencing hallucinations as a result
[10:54:23] <fuzzie> hm, what're you doing exactly?
[10:55:54] <fuzzie> as long as you always set SaveDir before you call GetSaveGames, it should be ok
[10:56:01] <chiv> setting play mode, savedir, oldgame and then SoAGames = GemRB.GetSaveGames()
[11:02:14] <chiv> patch (just for start.py) http://pastebin.com/QBk1b0mY
[11:03:16] <Yoshimo> could it be possible that the original engine loads savegames a lot faster than gemrb does?
[11:03:31] <chiv> actually gemrb seems to be faster
[11:03:34] <fuzzie> Yoshimo: as in, once you click 'load'?
[11:03:46] <Yoshimo> yea
[11:03:57] <fuzzie> the original engine is faster at loading original savegames.
[11:04:21] <Yoshimo> there are diffrences between a "vanilla" savegame and a gemrb one?
[11:04:25] <fuzzie> yes
[11:05:48] <Yoshimo> what kind of diffrences?
[11:05:51] <fuzzie> we don't delay area changes
[11:06:09] <fuzzie> so if you load an original game, gemrb goes and immediately handles all pending area changes
[11:09:11] <fuzzie> chiv: that's weird
[11:10:10] <chiv> what happens?
[11:10:45] <fuzzie> i can't run it here but all the code seems fine
[11:18:44] <chiv> by the way, the original save behaviour is save in /save for single player games, /mpsave for multi, with the -TOB suffix meaning tob games
[11:19:35] <chiv> comments in the gemrb code refer to 'mission pack save' which is a misunderstanding
[11:19:43] <fuzzie> it's not that simple.
[11:20:09] <fuzzie> original bg2 identifies ToB saves by means of checking which world map it's using.
[11:21:24] <chiv> still, the way it is it wont list tob games saved with bgmain.exe
[11:21:36] <fuzzie> yeah, it's broken
[11:21:51] <fuzzie> just not quite so simple to do as checking names
[11:22:49] <chiv> are you sure? it seems to positively flag tob games in the /save directory
[11:23:02] <chiv> if shoved
[11:23:52] <fuzzie> none of my ToB saves have that in them
[11:23:53] --> Maighstir has joined #gemrb
[11:24:12] <fuzzie> except for the auto/quick saves
[11:24:20] <fuzzie> i mean, the name
[11:24:51] <chiv> oh yeah
[11:25:29] <fuzzie> but the ToB world map (worldm25.wmp) should identify ToB saves afaik.
[11:25:41] <fuzzie> i think Maighstir worked out all the other details in the log somewhere.
[11:26:15] <fuzzie> from memory, Maighstir said mpsave is always multiplayer. but i forget.
[11:27:26] <chiv> oh well anyway, pic: http://imageshack.us/photo/my-images/220/startu.png/
[11:28:02] <fuzzie> that could be nice
[11:29:05] <chiv> yeah, just need some button gfx
[11:29:51] <chiv> other than it listing the wrong saves, it all works as intended
[11:36:11] <chiv> just checked again, in the load screens gemrb does properly list the games from the right mode, excluding games from the other mode
[11:37:08] <Maighstir> mpsave is always multiplayer except for iwd which always saves there (I assume it treats both singe and multi the same, unlike baldur's gate which has some other differences beween the modes), and the missionpack save install location from totsc (which I'm quite sure is a mistake, since the save isn't visible when choosing "Mission Pack" from the single-player menu).
[12:09:09] <chiv> if it works though, why would start2.py change the save dir for throne of bhaal games? leftover from a hack?
[12:11:23] <Maighstir> I assume from when someone thought "mp" in "mpsave" meant "missionpack" (as in "expansion pack")
[12:12:16] <chiv> so, I should just leave the save dir alone then
[12:12:24] <chiv> here is the culprit http://tinyurl.com/6aerma6
[12:13:04] <Maighstir> and that, I assume, comes from the fact that the missionpack save in totsc is only installed in the mpsave folder (the "mission pack" button, however, look in the singleplayer folder)
[12:14:34] <chiv> quite an epic history then :)
[12:15:12] <Maighstir> either they thought that people would only play the missionpack save in multiplsayer, or someone at bioware messed up the installer
[12:17:43] <Maighstir> going by the actual bahaviour of the games however: iwd and iwd2 - always mpsave; bg and bg2 - depends on mode; pst - of course, doesn't even have multi, so always "save"
[12:18:14] <chiv> well since that is cleared up, here is a proper working quickload start screen (although it expects expansion games to be in /save) http://pastebin.com/GUsyQG9b
[12:19:29] <chiv> now i need food
[12:32:38] <chiv> I think my next trick will be to turn that stupid bag screen into an extension of the inventory page
[12:46:48] --> SiENcE has joined #gemrb
[12:59:41] <lynxlynxlynx> GUISTORE
[13:00:03] <lynxlynxlynx> it will also affect stores, inns, temples
[13:02:41] <chiv> Well, I would leave GUISTORE alone and just extend inventory page
[13:03:42] <chiv> since you can add new windows and buttons the potential is quite large :)
[13:06:08] <chiv> and if the bags are visible from the inventory page, then automatically bagging items would be a logical progression
[13:36:23] <lynxlynxlynx> you'll have to decouple it then
[13:51:38] <CIA-44> GemRB: 03lynxlupodian * re20ff1b2f0f7 10gemrb/gemrb/GUIScripts/ie_stats.py: ie_stats.py: fixed IE_CASTING's name to match the core
[13:51:39] <CIA-44> GemRB: 03lynxlupodian * r4ae9ef351e5f 10gemrb/gemrb/ (3 files in 2 dirs): GUIScripts: made [SG]etPlayerStat optionally accept global ids
[13:51:39] <CIA-44> GemRB: 03lynxlupodian * red5cfb15d0d3 10gemrb/gemrb/plugins/GUIScript/GUIScript.cpp: GemRB_GetSpell: also expose the hostile bit and the spell resref
[13:51:43] <CIA-44> GemRB: 03lynxlupodian * r414711ae9639 10gemrb/gemrb/GUIScripts/ (bg1/GUIMG.py bg2/GUIMG.py iwd/GUIMG.py): GUIMG: removed redundant loop
[13:58:39] <-- gembot has left IRC (Ping timeout: 248 seconds)
[14:00:18] <lynxlynxlynx> almost there
[14:00:53] <lynxlynxlynx> the bar is displayed, but scrolling doesn't work, there are no numbers and wrong spells get cast :)
[14:01:23] <lynxlynxlynx> seems i'll need another guiscript binding for the latter
[14:02:50] <chiv> hah, or you could rename it a patch for wild magic
[14:03:07] <chiv> interactive version
[14:05:28] <chiv> i dont see a provision in the code for selling from one store to another, am i missing something?
[14:06:34] <lynxlynxlynx> nope
[14:07:03] <lynxlynxlynx> i'm not sure you can enter a store while in a store in gemrb
[14:07:27] <chiv> the guiscript plugin only accounts for the current store
[14:07:43] <lynxlynxlynx> everything does
[14:07:58] <chiv> so buying/selling to a bag like in bgmain.exe not possible :(
[14:08:02] <lynxlynxlynx> needs fixing, so you can go yodawg on us with that :)
[14:08:24] <chiv> i will need it, so will try :)
[14:08:33] <fuzzie> someone, maybe tomprince, already fixed most of what's needed for such a thing
[14:17:22] --> gembot has joined #gemrb
[16:08:28] --> brad_a has joined #gemrb
[16:16:43] <-- brad_a has left IRC (Quit: brad_a)
[16:41:22] --> brad_a has joined #gemrb
[16:49:06] <chiv> I've not actually written anything but pseudocode yet, but I believe I can simplify all inventory/store/container transactions quite a bit
[17:04:12] <Yoshimo> sounds good
[17:05:16] <chiv> cant promise timeframe though, you know what happens when you pull a thread from a shirt...
[17:07:49] <Yoshimo> you can end up naked
[17:07:58] <brad_a> :-P
[17:13:36] <chiv> well, or you end up needing a new shirt :)
[17:16:27] <chiv> signing off
[17:16:33] <-- chiv has left IRC (Quit: Page closed)
[17:29:53] <-- SiENcE has left IRC (Quit: cya)
[17:44:52] <lynxlynxlynx> doh
[17:45:11] <lynxlynxlynx> the numbers didn't show, since SetText can also take strrefs
[18:10:08] <-- Maighstir has left #gemrb
[18:11:03] --> Maighstir has joined #gemrb
[18:17:53] <brad_a> tomprince: change the CLANG in that ifdef to __clang__ and go ahead and commit that.
[18:51:28] <-- brad_a has left IRC (Remote host closed the connection)
[18:51:37] --> brad_a has joined #gemrb
[18:52:45] <CIA-44> GemRB: 03lynxlupodian * r822c795b78a7 10gemrb/gemrb/ (3 files in 2 dirs):
[18:52:45] <CIA-44> GemRB: GemRB_CheckSpecialSpell: don't garble the input
[18:52:45] <CIA-44> GemRB: plus two related constifications
[18:52:46] <CIA-44> GemRB: 03lynxlupodian * r2fc32326cb4e 10gemrb/gemrb/ (3 files in 2 dirs):
[18:52:46] <CIA-44> GemRB: changed FindSpellInfo to return the index+1 instead of just 0/1 and
[18:52:46] <CIA-44> GemRB: made use of it in the new GemRB_CheckSpecialSpell binding
[19:05:46] --> brad_a_ has joined #gemrb
[19:07:41] <-- brad_a has left IRC (Ping timeout: 240 seconds)
[19:07:41] --- brad_a_ is now known as brad_a
[19:11:15] <brad_a> any idea if/when G3 will be available again?
[19:14:22] <lynxlynxlynx> nope
[19:14:40] <lynxlynxlynx> hopefully they're fixing the admin area stuff
[19:15:04] <lynxlynxlynx> that and the resulting spam situation has been silly for over a year now
[19:16:28] <brad_a> just as long as its not gone for good :)
[19:19:07] <lynxlynxlynx> nah, it always comes back
[19:21:08] <lynxlynxlynx> ah, found the culprit
[19:22:32] <Yoshimo> what admin area stuff?
[19:25:19] <lynxlynxlynx> i don't know, but it sounded pretty silly
[19:28:23] <Yoshimo> well shs is not save from spam either
[22:15:58] --> chiv has joined #gemrb
[22:18:01] <-- chiv has left IRC (Client Quit)
[22:21:46] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[22:51:55] <brad_a> finally got valgrind working :) its pretty slick.
[22:58:03] <Yoshimo> and what does it tell you about gemrb?
[23:02:32] <brad_a> well i havent released the ttf coe yet because there was a random crash
[23:02:48] <brad_a> valgrind was able to tell me exactly the line where it was occuring
[23:03:57] <brad_a> simple mistake of setting up whitespace characters for fonts that end before the space character
[23:04:28] <Yoshimo> nice tool
[23:04:32] <brad_a> sure is
[23:07:36] <brad_a> it also tells my python is a mess
[23:07:48] <brad_a> but gemrb seems to be clean
[23:07:51] <fuzzie> python is always a mess
[23:08:06] <brad_a> how do i have it ignore pythin?
[23:08:10] <fuzzie> it does memory management by sort of randomly poking at possibly-invalid addresses
[23:08:27] <fuzzie> there's a valgrind 'suppressions' file in the python source
[23:08:39] <brad_a> nice
[23:09:17] <fuzzie> i guess pointing you at /usr/lib/valgrind/python.supp doesn't help because you will not be running debian/ubuntu/etc :P
[23:09:31] <brad_a> nope
[23:09:39] <brad_a> im using apples python build anyway
[23:15:03] <brad_a> looks like i have a few memory leaks… not sure how exactly
[23:16:13] <brad_a> nevermind
[23:16:25] <brad_a> those were from a previous run when i terminated with ctrl+c
[23:16:38] <brad_a> quitting normally has no leaks
[23:28:30] <tomprince> brad_a: testing/python.supp
[23:33:25] <fuzzie> ah. even more convenient, nice :)
[23:39:42] <-- Yoshimo has left IRC (Quit: Yoshimo)
[23:53:44] <brad_a> oh neat. thanks.