#gemrb@irc.freenode.net logs for 4 Oct 2011 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:06:35] <gembot> build #20 of nmake-msvc++6 is complete: Exception [exception minimal test upload] Build details are at http://buildbot.gemrb.org/builders/nmake-msvc%2B%2B6/builds/20
[00:13:15] <-- Maighstir has left IRC (Ping timeout: 260 seconds)
[00:14:31] <CIA-26> GemRB: 03bradallred * r02892aaa71ba 10gemrb/gemrb/core/GUI/ (TextArea.cpp TextArea.h): remove ivar smooth from TextArea. its function is replaced by the simpler per-pixel scrolling mechanic.
[00:14:31] <CIA-26> GemRB: 03bradallred * r1c62b2a3cb36 10gemrb/ (8 files in 7 dirs): Merge branch 'master' of ssh://gemrb.git.sourceforge.net/gitroot/gemrb/gemrb
[00:19:38] --> joneirik has joined #gemrb
[01:10:39] <-- barra__away has left IRC (*.net *.split)
[01:10:44] <-- xrogaan has left IRC (*.net *.split)
[01:10:45] <-- gembot has left IRC (*.net *.split)
[01:10:45] <-- jschall has left IRC (*.net *.split)
[01:10:46] <-- DrMcCoy has left IRC (*.net *.split)
[01:10:47] <-- fuzzie has left IRC (*.net *.split)
[01:10:48] <-- Drakkar has left IRC (*.net *.split)
[01:10:50] <-- tomprince has left IRC (*.net *.split)
[01:10:50] <-- CIA-26 has left IRC (*.net *.split)
[01:10:52] <-- wjp has left IRC (*.net *.split)
[01:13:17] --> gembot has joined #gemrb
[01:13:17] --> jschall has joined #gemrb
[01:13:17] --> barra__away has joined #gemrb
[01:13:17] --> fuzzie has joined #gemrb
[01:13:17] --> DrMcCoy has joined #gemrb
[01:13:17] --> xrogaan has joined #gemrb
[01:13:17] --> Drakkar has joined #gemrb
[01:13:17] --> tomprince has joined #gemrb
[01:13:17] --> CIA-26 has joined #gemrb
[01:13:17] --> wjp has joined #gemrb
[01:14:22] --> brad_a has joined #gemrb
[01:58:09] <gembot> build #24 of nmake-msvc++6 is complete: Exception [exception minimal test] Build details are at http://buildbot.gemrb.org/builders/nmake-msvc%2B%2B6/builds/24 blamelist: Tom Prince <tom.prince@ualberta.net>
[02:21:31] <tomprince> does the options display actually work for people, or does it always display diabled?
[03:43:01] <-- joneirik has left IRC (Remote host closed the connection)
[03:46:44] <gembot> build #26 of nmake-msvc++6 is complete: Success [build successful] Build details are at http://buildbot.gemrb.org/builders/nmake-msvc%2B%2B6/builds/26
[03:53:23] <tomprince> So a setting of './Cache/' didndn't work, but 'Cache' did, it seems.
[03:56:46] <gembot> build #246 of msvc++6 is complete: Success [build successful] Build details are at http://buildbot.gemrb.org/builders/msvc%2B%2B6/builds/246
[04:45:14] <gembot> build #222 of mingw32 is complete: Success [build successful] Build details are at http://buildbot.gemrb.org/builders/mingw32/builds/222
[05:01:31] <CIA-26> GemRB: 03tom.prince * rd26e8a20a56f 10gemrb/ (3 files in 2 dirs):
[05:01:31] <CIA-26> GemRB: win32: Add option to use stdio rather than conio for output.
[05:01:31] <CIA-26> GemRB: The buildbot can't easily capture the console output, so add an option to
[05:01:31] <CIA-26> GemRB: disable it.
[05:16:26] <CIA-26> GemRB: 03bradallred * r281c74615bf4 10gemrb/ (3 files in 2 dirs): Merge branch 'master' of ssh://gemrb.git.sourceforge.net/gitroot/gemrb/gemrb
[05:16:34] <CIA-26> GemRB: 03bradallred * r16240e681190 10gemrb/gemrb/core/Font.cpp:
[05:16:34] <CIA-26> GemRB: Fix for drop-caps shifting up and disappearing in scrolling TextAreas.
[05:16:34] <CIA-26> GemRB: Note that the SetupString calculation is still broken for auto scrolling text.
[05:16:34] <CIA-26> GemRB: 03bradallred * rf52c7d9c375d 10gemrb/gemrb/core/Font.cpp: skip whitespace when finding the first character to be a drop-cap.
[05:16:35] <CIA-26> GemRB: 03bradallred * r69ad51ed0e15 10gemrb/gemrb/plugins/OpenALAudio/OpenALAudio.cpp:
[05:16:35] <CIA-26> GemRB: Work around yet another stagnant error issue on Mac.
[05:16:36] <CIA-26> GemRB: Wish I knew the source.
[05:19:00] --> edheldil_ has joined #gemrb
[05:25:00] <-- edheldil_ has left IRC (Ping timeout: 255 seconds)
[05:42:23] <tomprince> lynxlynxlynx: There are binaries at http://buildbot.gemrb.org/binaries/gemrb-{nmake,win32}-v0.6.6.zip but I don't have a way to test them.
[06:39:48] <-- brad_a has left IRC (Quit: brad_a)
[06:48:48] --> brad_a has joined #gemrb
[06:53:26] <-- brad_a has left IRC (Client Quit)
[07:02:19] --> lynxlynxlynx has joined #gemrb
[07:02:20] <-- lynxlynxlynx has left IRC (Changing host)
[07:02:20] --> lynxlynxlynx has joined #gemrb
[07:02:20] --- ChanServ gives channel operator status to lynxlynxlynx
[07:15:54] <lynxlynxlynx> tomprince: thanks
[07:16:13] <lynxlynxlynx> any windows users awake yet?
[07:50:46] <tomprince> btw, they get generated automatically, for every (group of) commit(s).
[08:05:26] <-- barra__away has left IRC (Read error: Connection reset by peer)
[08:08:44] <lynxlynxlynx> yeah, i saw (needed to change the filenames)
[08:09:23] <lynxlynxlynx> i uploaded them to sf, let's see what happens :)
[08:09:33] <lynxlynxlynx> btw, why is the nmake build so much smaller?
[08:11:08] <fuzzie> if the -win32 one is mingw, maybe it's unstripped?
[08:14:51] <lynxlynxlynx> 4 vs 14M
[08:15:47] <fuzzie> yes, it's unstripped.
[08:16:53] <fuzzie> and the nmake one is indeed from vc++ so it just has a .pdb reference.
[08:43:18] <-- lynxlynxlynx has left IRC (Ping timeout: 245 seconds)
[09:04:27] --> lynxlynxlynx has joined #gemrb
[09:04:27] <-- lynxlynxlynx has left IRC (Changing host)
[09:04:27] --> lynxlynxlynx has joined #gemrb
[09:04:27] --- ChanServ gives channel operator status to lynxlynxlynx
[09:39:34] --> pugvader_ has joined #gemrb
[09:42:43] <-- pugvader has left IRC (Ping timeout: 244 seconds)
[12:05:18] --> Beh0lder has joined #gemrb
[12:05:29] <Beh0lder> hi all
[12:08:56] <Beh0lder> I can't get to work official SDL for Android(((
[12:15:15] <edheldil> Hi
[13:22:03] <-- Gekz has left IRC (Read error: Operation timed out)
[13:25:05] --> Gekz has joined #gemrb
[13:25:08] <-- Gekz has left IRC (Changing host)
[13:25:08] --> Gekz has joined #gemrb
[13:43:30] <gembot> build #225 of mingw32 is complete: Exception [exception interrupted] Build details are at http://buildbot.gemrb.org/builders/mingw32/builds/225
[13:51:19] <gembot> build #226 of mingw32 is complete: Exception [exception interrupted] Build details are at http://buildbot.gemrb.org/builders/mingw32/builds/226
[13:53:12] <tomprince> watch mingw32
[13:53:27] <tomprince> gembot: watch mingw32
[13:53:28] <gembot> watching build mingw32 #227 until it finishes..
[13:54:17] <tomprince> lynxlynxlynx: ^--- the build is a release mode build of v0.6.6
[13:55:05] <lynxlynxlynx> a debug build is fine, do you think that's needed?
[14:03:28] <gembot> Hey! build mingw32 #227 is complete: Success [build successful]
[14:03:28] <gembot> Build details are at http://buildbot.gemrb.org/builders/mingw32/builds/227
[14:04:54] <CIA-26> GemRB: 03tom.prince * r125dda82d30d 10gemrb/CMakeLists.txt: (log message trimmed)
[14:04:54] <CIA-26> GemRB: cmake: Fix behaviour wrt cmake policy CMP0012.
[14:04:54] <CIA-26> GemRB: Old versions of cmake interpret
[14:04:54] <CIA-26> GemRB: IF(Yes)
[14:04:54] <CIA-26> GemRB: by dereferencing a variables 'Yes' (which is unlikely to be a true value). In
[14:04:54] <CIA-26> GemRB: particular, if variable 'option' has the value 'Yes', then
[14:04:54] <CIA-26> GemRB: IF(${option}"
[14:05:13] <tomprince> Doesn't matter to me.
[14:06:50] <tomprince> It does make that builder run about ~10 min faster...
[14:15:12] <-- gembot has left IRC (Ping timeout: 260 seconds)
[14:31:25] --> SiENcE has joined #gemrb
[14:47:25] --> brad_a has joined #gemrb
[15:03:12] <-- lynxlynxlynx has left IRC (Ping timeout: 248 seconds)
[15:06:16] <-- SiENcE has left IRC (Quit: @all: cya)
[15:15:51] --> lynxlynxlynx has joined #gemrb
[15:15:51] <-- lynxlynxlynx has left IRC (Changing host)
[15:15:51] --> lynxlynxlynx has joined #gemrb
[15:15:51] --- ChanServ gives channel operator status to lynxlynxlynx
[15:24:43] <-- lubos has left IRC (Quit: Leaving.)
[15:35:27] --> Maighstir has joined #gemrb
[16:39:32] --> Yoshimo has joined #gemrb
[17:22:49] --> jschall_ has joined #gemrb
[17:23:19] <-- jschall_ has left IRC (Remote host closed the connection)
[17:27:06] <-- jschall has left IRC (Ping timeout: 260 seconds)
[17:43:18] <-- pugvader_ has left IRC (Ping timeout: 245 seconds)
[17:45:26] --> pugvader has joined #gemrb
[18:07:50] <brad_a> ttf patch http://dl.dropbox.com/u/13866402/ttf.patch
[18:10:33] <fuzzie> "if (reInit){" is silly style
[18:10:44] <fuzzie> also commits which don't build are maybe not very nice since they prohibit bisecting..
[18:13:21] <fuzzie> (and they make the patch pretty difficult to read since I can't tell how it works)
[18:14:41] <fuzzie> also "#define PARAGRAPH_START_X 5;" is horrible but nothing to do with you :)
[18:15:08] <fuzzie> reassuringly that dates back to 2003, when it was committed by balrog
[18:15:35] <fuzzie> brad_a: got screenshots? :)
[18:15:53] <fuzzie> (and you have commit access, right?)
[18:15:58] <brad_a> yes i do
[18:16:13] <brad_a> to both
[18:16:31] <brad_a> http://dl.dropbox.com/u/13866402/teaser.png
[18:17:26] <wjp> brad_a: by the way, in reply to yesterday: those bg2/tob saves from me are from the original BG2, so while they're old, they shouldn't be outdated
[18:17:39] <brad_a> well i can squash some commits together if we would rather have larger buildable ones
[18:18:23] <fuzzie> oh, yes, also in reply to yesterday: that cutscene mouse bug is known, the textscreen code isn't restoring mouse state, a trivial fix at this point i think since avenger routed it via EF_TEXTSCREEN as requested
[18:18:40] <brad_a> so is anyboy else able to replicate being able to control the party etc during the battle sequence at the end of BG2 prologue?
[18:18:49] <brad_a> ah
[18:19:38] <fuzzie> ok not quite trivial but still
[18:19:58] <fuzzie> Interface.cpp:625 is the line which enables the mouse for the textscreen
[18:20:18] <brad_a> i do have a problem in that i have no idea how to add that TTF plugin to cmake. im not quite sure how all that include(FindXXX) and such works
[18:20:18] <fuzzie> so i think ideally you would remove that
[18:20:22] <fuzzie> and then do it from TextScreen.py
[18:20:40] <fuzzie> so then you can save/restore the mouse state in python land.
[18:20:58] <fuzzie> and i'm not sure either and can't look right at this moment buti 'm sure we can work it out
[18:21:15] <lynxlynxlynx> each commit should definitely be a selfsufficient individual
[18:21:30] <brad_a> ok ill rebase and sqaush some
[18:21:36] <lynxlynxlynx> i can look at the cmake stuff
[18:22:42] <brad_a> thank you
[18:25:47] <brad_a> oh one thing not in that patch that i need to ask about
[18:27:02] <brad_a> on line 776 on TextArea.cpp Height is being divided by size[1].h which in my ttf plugin is 0
[18:27:12] <brad_a> i changed that to ftext->maxHeight
[18:27:36] <brad_a> seems to work, but i have seen that size[1] thing a couple places and im wondering what the deal is
[18:39:08] <lynxlynxlynx> probably ok
[18:39:28] <lynxlynxlynx> why would you want to limit the display within the range from the fonts table?
[18:41:30] <brad_a> im not sure what you are asking
[18:42:19] <lynxlynxlynx> there's a fixme at the end of the plugin about that
[18:42:30] <lynxlynxlynx> last char limit
[18:45:40] <brad_a> oh that. couple of reasons
[18:46:14] <brad_a> 1. if you dont need fancy characters then you can save memory by limiting the characterset
[18:47:06] <brad_a> 2 im thinking for unicode languages ther may be a need to change the range to get the correct glyph set
[18:50:10] <brad_a> right now that ttf font plugin is limited to ascii range and there is not a need for that limit other than to prevent massive sprites with useless glyphs
[18:58:04] <lynxlynxlynx> true, unicode fonts can be big
[19:05:35] --> edheldil_ has joined #gemrb
[19:11:33] <brad_a> once i implement using a provided range in fonts.2da then as far as i can tell this should work with any localization of game files. languages that dont have a BAM should be possible right?
[19:11:53] <brad_a> as long as somebody localizes the strings
[19:22:41] <Yoshimo> brad, unicode as in weird chinese or russian letters?
[19:23:32] <brad_a> yes
[19:23:43] <brad_a> an double byte language
[19:26:35] <lynxlynxlynx> the tlk format probably can't handle it without extension
[19:28:10] <brad_a> so how is Chinese currently implemented?
[19:28:38] <Yoshimo> wasnt there a patch on the forum for chinese? did it get merged?
[19:29:53] <lynxlynxlynx> all nonascii languages just used extended bams
[19:30:14] <lynxlynxlynx> chinese of course doesn't fit into the 255
[19:31:02] <lynxlynxlynx> i wonder what they used in the tlk, a transliteration?
[19:31:17] <brad_a> that was what i wanted to know
[19:32:19] <brad_a> however they did it the same thing will be able to work with TTF without having to make a BAM
[19:35:10] <brad_a> oh ill bet they just cut each glyph in half
[19:35:18] <brad_a> so that wouldnt work
[19:35:45] <brad_a> maybe im not sure if im talking sense
[19:37:10] <lynxlynxlynx> nothing important yet
[20:11:24] <fuzzie> the chinese tlk has double-byte chars
[20:12:00] <fuzzie> as i noted to brad_a a week or so back, the font code in the patch converts all strings to uint16* and then renders that
[20:12:30] <fuzzie> but it's not unicode
[20:14:15] <brad_a> how is that not unicode?
[20:14:30] <fuzzie> because they're not unicode chars
[20:14:36] <brad_a> what are they then?
[20:14:41] <fuzzie> language-specific
[20:15:56] <fuzzie> shift-jis for japanese, i guess probably GB2312 for chinese but not sure, maybe big5, didn't look
[20:17:39] <brad_a> :\ i'm probably displaying my ignorance of the subject but i dont understand why they dont use unicode? do they not even map to unicode like ascii does?
[20:18:02] <fuzzie> unicode only became widespread relatively recently
[20:18:19] <brad_a> i see
[20:19:01] <fuzzie> i mean it is possible i'm wrong about the encodings here, it has been a while
[20:19:54] <fuzzie> but traditionally windows used 'code pages' which basically were system-wide character encodings, whether 8-bit (with stuff like Latin-1) or 16-bit (shift-jis, gbk, big5, etc)
[20:20:47] <brad_a> well the ttf plugin does use uint16 but it is a unicode value and the sdl ttf library only supports latin1 utf8 or unicode anyway
[20:21:02] <fuzzie> yes, so if you wanted to support them for ttf you'd need a lookup table
[20:21:35] <brad_a> yeah. well ill leave that as an exercise to somebody that knows Chinese :-P
[20:22:22] <brad_a> is there a reason why they couldnt just export the tlk as unicode?
[20:24:15] <fuzzie> who?
[20:26:16] <Yoshimo> i think having it running on latin based languages is a higher priority
[20:26:38] <brad_a> well im discussing for educational purposes im not actually going to do anything :)
[20:26:48] <fuzzie> well, i would like to merge the patch in question
[20:27:00] <fuzzie> not going to make it work with ttf though..
[20:27:08] <brad_a> sure
[20:27:17] <fuzzie> i don't think SDL_ttf does unicode sufficiently anyway
[20:27:18] <Yoshimo> educational purposes, wasnt that why they offer hacks for iphones and "trial software"? ;)
[20:27:21] <brad_a> if it requires a table like you say then dont bother
[20:27:59] <edheldil_> it's just 2^16 chars at most ...
[20:28:01] <fuzzie> i'm looking at the source code and it just tries rendering every uint16 as an individual character as far as i can tell
[20:31:16] <Yoshimo> doesnt sound much edheldil ;)
[20:41:43] <fuzzie> can always autogenerate tables
[20:41:58] <fuzzie> original engine just uses standard win32 api funcs
[20:42:45] <brad_a> still not gonna do it :-P
[20:46:22] <brad_a> hope no Chinese people want to play on a mobile device or hope they have better eyesight than me.
[20:47:01] <fuzzie> well the Android tablets tend to not be quite as crazily dense as e.g. the iPad :)
[20:53:32] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[20:55:17] --> lynxlynxlynx has joined #gemrb
[20:55:17] <-- lynxlynxlynx has left IRC (Changing host)
[20:55:17] --> lynxlynxlynx has joined #gemrb
[20:55:17] --- ChanServ gives channel operator status to lynxlynxlynx
[20:56:01] <brad_a> speaking of android. i dont suppose anybody in here knows if android can even use the TTF plugin? i want to assume that you can build libfreetype for android
[21:00:48] <brad_a> as far as i can tell android is good to go with TTF plugin.
[21:21:15] <-- Yoshimo has left IRC (Quit: Yoshimo)
[21:31:59] <-- brad_a has left IRC (Ping timeout: 260 seconds)
[22:04:07] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[22:29:37] <tomprince> I wonder if the custom fonts should be first in the search path. The source should probably be called Fonts or Custom Fonts, anyway.
[22:32:30] <tomprince> I'd seperate BAMFont to a differnt file.
[22:33:03] <tomprince> You made a couple of things virtual, that you didn't then override.
[22:35:38] <tomprince> I'd be tempted yo say that you shouln't actaully be sublassing Font, but have a FontResource type that has a GetFont, which does the work that Initialize is currently doing.
[22:59:08] <-- edheldil_ has left IRC (Ping timeout: 245 seconds)
[23:07:15] --> brad_a has joined #gemrb
[23:07:21] <brad_a> tomprince: I tried having BAMFont in its own file and it wouldnt build tho i cant remember why.
[23:07:51] <-- pugvader has left IRC (Quit: Lost terminal)
[23:08:53] <brad_a> sorry i made those methods virtual for testing purposes. I guess i should un virtualize them now
[23:20:04] <brad_a> to discard that virtual print method commit do i just rebase -i and delete it from the list?
[23:37:59] <-- Maighstir has left IRC (Quit: .)
[23:46:39] <tomprince> brad_a: that should work.
[23:49:20] <tomprince> (I didn't realise when I first looked at tit that you posted an mbox of patches rather than just a patch)
[23:52:20] <tomprince> Also, BAMFont shouldn't inherit from BAMImporter, it should just have a BAMImporter member.
[23:52:47] <brad_a> I guess. I actually went down that path originally and didnt like it