| --- Log opened Sat May 08 00:00:55 2010 |
| 00:01 | | shade_of_cpux is now known as cpux |
| 00:03 | | You're now known as TheWatcher[T-2] |
| 00:06 | | You're now known as TheWatcher[zZzZ] |
| 00:19 | | Rhamphoryncus [rhamph@Nightstar-8931f88f.abhsia.telus.net] has joined #code |
| 01:13 | | * Vornicus fiddles with QT Designer. Tries to figure out what precisely he's doing. |
| 01:19 | | Derakon [Derakon@Nightstar-5213d778.ca.comcast.net] has joined #code |
| 01:19 | | mode/#code [+o Derakon] by Reiver |
| 01:30 | <@Vornicus> | hokay lessee. |
| 01:31 | <@Vornicus> | Over on the right i want two boxes, with titles, and radio buttons in there. Then on the left I want a treeview. |
| 01:32 | <@Vornicus> | The treeview is actually built from checkboxes. |
| 01:39 | <@Derakon> | The tickets are now diamonds. |
| 01:39 | <@Vornicus> | Are you on a horse? |
| 01:40 | <@Derakon> | Possibly~ |
| 01:44 | | * Derakon sighs at someone describing themselves as a "Windows fan" just because that's all they've ever used for any remotely significant amount of time. |
| 02:20 | | Derakon is now known as Derakon[AFK] |
| 02:22 | < cpux> | I say we take Steve Ballmer and Steve Jobs, stick them in the same room, and see which one can out-crazy the other one. |
| 02:22 | <@McMartin> | I think they should both be locked in a room with Steve the Avatar. |
| 02:23 | <@McMartin> | Three Steves Enter, One Steve Leaves. |
| 02:24 | < cpux> | Only on Pay-Per-View! $49.95! |
| 02:55 | | * Vornicus fiddles with the treeview thing, trying to figure out what he's aiming to do here. |
| 02:58 | <@McMartin> | I forget, does Qt merge Tree and List views? |
| 02:59 | <@McMartin> | They tend to be the Nastiest Widgets Ever, either way, in all systems |
| 02:59 | <@McMartin> | I need to learn how to do table views in it at some point |
| 02:59 | <@McMartin> | But! Priorities. |
| 02:59 | <@McMartin> | QBlorple is clearly, at best, #4 on my List. |
| 03:00 | <@McMartin> | After FotH, the remaining tasks I have to do for UQM 0.7, and the "do something with Love2D so you can have an opinion about it" project. |
| 03:00 | <@Vornicus> | QT does separate List and Tree views. |
| 03:01 | <@Vornicus> | Unfortunately the TreeView is the Obvious Right Answer for the task I have for it. |
| 03:01 | < celticminstrel> | UQM is only at 0.7? |
| 03:02 | <@McMartin> | No |
| 03:02 | <@McMartin> | UQM is at 0.6.2 on actual releases. |
| 03:02 | < celticminstrel> | Why is it not even at 1.x? |
| 03:02 | <@McMartin> | It was "basically done" from a "I want to play the game" at around 0.4, and everything else has been refactoring the unusuable code base. |
| 03:03 | <@McMartin> | 1.0 was "we can walk away from it and the people who want to make mods will be able to do so without forking and hacking the engine" |
| 03:03 | <@McMartin> | And yeah, that's at about 0.7 |
| 03:03 | < celticminstrel> | Oh, you have a roadmap. |
| 03:04 | <@McMartin> | Most of my work for 0.7 has been further work in the long, slow process of untangling the content from weird nonstandard index/resource specifier formats. |
| 03:04 | <@McMartin> | At least it's all text now. |
| 03:04 | <@McMartin> | But there's still like five nonoverlapping systems in 0.6.x |
| 03:04 | <@McMartin> | In 0.7 that drops to three. |
| 03:04 | <@McMartin> | Which may be low enough, really; there's no reason save formats should be unified with the rest per se |
| 03:06 | <@McMartin> | But yeah, after 1.0's maintenance releases, that's the point where coredev officially retires. |
| 03:23 | < Namegduf> | Hmm. |
| 03:24 | < Namegduf> | What would people advise for Python configuration files? I think it's come up before, someone suggested ConfigParser, which I found ugly but didn't have a justification for. |
| 03:24 | < Namegduf> | I've used ConfigParser, and it's worked okay, but it's not scaling too well in style to heavily commented, strictly broken into sections configuration files. |
| 03:24 | < Namegduf> | I'm wondering if anyone has other ideas. |
| 03:25 | < Namegduf> | Python's JSON parser doesn't permit comments, counting it out, and as it's part of an existing project which isn't too huge, adding a YAML parser isn't feasible (it'd add about 50% to the size of things) |
| 03:25 | < Namegduf> | Which were both my first two ideas. |
| 03:32 | < celticminstrel> | md5 output is always alphanumeric, right? |
| 03:34 | <@Vornicus> | md5 technically outputs a sequence of bytes -- it just generally gets displayed as hexadecimal. |
| 03:42 | < celticminstrel> | Seems like it outputs it formatted as hexadecimal rather than as raw bytes. |
| 03:42 | <@Vornicus> | Right. |
| 03:43 | <@Vornicus> | If you were to use an md5 summer in a program, it would probably receive it as bytes. |
| 03:46 | < celticminstrel> | ...is there some advantage to applying md5 to the output of md5? |
| 03:48 | <@Vornicus> | ...no? |
| 03:48 | < celticminstrel> | I didn't think so. |
| 03:51 | < cpux> | md5sum is a 32-byte checksum, and it's typically represented in hexadecimal, yes. |
| 03:51 | < cpux> | And I can't imagine md5ing an md5 being any useful. |
| 03:52 | <@Vornicus> | 16-byte |
| 03:52 | <@Vornicus> | 32 hexadecimal digits. |
| 03:52 | < cpux> | right. I always confuse that. |
| 03:53 | < cpux> | md5's 16 byte. sha-1's 20-byte. |
| 03:54 | < cpux> | There an accepted sha-3 yet? |
| 04:19 | | * Vornicus fiddles with qt. okay. adding a treeview is all well and good, but now I have to write the damn code. |
| 04:26 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has quit [[NS] Quit: Z?] |
| 04:39 | <@McMartin> | Vornicus: Qt Assistant is your only friend |
| 04:52 | <@Vornicus> | Apparently! |
| 05:34 | | * Rhamphoryncus ponders utf-8 vs utf-32 for his python implementation |
| 05:34 | < Rhamphoryncus> | (utf-32 is right out) |
| 05:34 | <@McMartin> | utf-32 is properly called UCS4, tbh |
| 05:35 | <@McMartin> | UCS4 is only useful if you're writing C. |
| 05:35 | < Rhamphoryncus> | No, UTF-32 and UCS4 are synonymous |
| 05:36 | < Rhamphoryncus> | well almost |
| 05:36 | < Rhamphoryncus> | ISO 10646 defines UCS-4. Unicode defines UTF-32 |
| 05:37 | <@McMartin> | Hm. I thought those werer defined to track each other? |
| 05:39 | < Rhamphoryncus> | They are now, yes |
| 05:40 | < Rhamphoryncus> | I believe UCS-4 defines a little more information not in unicode, but I don't think anybody cares |
| 05:40 | < Rhamphoryncus> | whereas UCS-2 is usually used (correctly or not, I'm not sure) to refer to the archaic pre-surrogate form of UTF-16 |
| 05:42 | <@McMartin> | Yeah, I have no idea how those relate |
| 05:42 | < Rhamphoryncus> | I've never heard anybody discuss BOM for UCS-4 either, which suggests they have some different mental context for it |
| 05:43 | < Rhamphoryncus> | huh. ISO 10646 only has the BMP. However, they will revise it in the future to add the rest |
| 05:44 | <@McMartin> | Oho |
| 05:44 | < Rhamphoryncus> | Which is odd.. they UCS-4 has a range of 0..2**31, and has the same character set, but doesn't allow it all? For UCS-2, sure, but UCS-4 too? I'm not sure that's right |
| 05:44 | <@McMartin> | I was unaware of that |
| 05:45 | <@McMartin> | UCS-4: It's UCS-2 with double the space requirements, all of which have to be zero |
| 05:45 | <@McMartin> | :bravo: |
| 05:45 | < Rhamphoryncus> | lol |
| 05:47 | < Rhamphoryncus> | heh, UCS-4 has private use codes not accessible through UTF-16 |
| 05:47 | < Rhamphoryncus> | IOW, UCS-4 is a superset of unicode |
| 05:48 | < Rhamphoryncus> | it does sound like UCS-4 supports outside the BMP though |
| 05:51 | < celticminstrel> | They all do. |
| 05:51 | < Rhamphoryncus> | and ISO 10646 explicitly allows you to implement only certain subsets. Thus, UCS-2 is perfectly legal ISO 10646, but not an implementation of unicode |
| 05:52 | < Rhamphoryncus> | celticminstrel: eh? |
| 05:53 | < celticminstrel> | UTF-8, UTF-16, UTF-32... they should all support characters outside the Basic Multilingual Plane. |
| 05:54 | < Rhamphoryncus> | UTF-8, UTF-16, and UTF-32 are implementations of Unicode and support the entire unicode character set. UCS-2 and UCS-4 are implementations of ISO 10646 and do not necessarily support the entire unicode character set |
| 05:55 | < celticminstrel> | ...why wouldn't UCS-2 support outside the Basic Multilingual Plane? |
| 05:55 | < Rhamphoryncus> | the UCS-2 spec does not include surrogates. It is specifically only the BMP |
| 05:55 | < Rhamphoryncus> | ISO 10646 matched up the character sets, NOT the encodings |
| 05:56 | < celticminstrel> | UCS-4 allows for values above 0x10FFFF? |
| 05:56 | < Rhamphoryncus> | Even for UCS-4 vs UTF-32 there's substantial differences in conformance requirements. Under UCS-4 it's perfectly legal to have values about 0x10FFFF, but not in UTF-32 |
| 05:58 | < celticminstrel> | So why would anyone use UCS-2? |
| 05:58 | < Rhamphoryncus> | that's a good question |
| 05:59 | < Rhamphoryncus> | The biggest reason I can figure out is they're confused about what the standards %breally%b say |
| 05:59 | < Rhamphoryncus> | oh that came out well |
| 05:59 | < Rhamphoryncus> | The biggest reason I can figure out is they're confused about what the standards really say |
| 05:59 | < celticminstrel> | ...%b? |
| 05:59 | < Rhamphoryncus> | xchat stupidity :P |
| 05:59 | < Rhamphoryncus> | why have something normal like ctrl-b for bold when you can have them type in % B instead |
| 06:00 | < celticminstrel> | I do the former. |
| 06:00 | < celticminstrel> | It's really no harder. |
| 06:00 | < celticminstrel> | Anyway, bed. |
| 06:00 | | celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has quit [[NS] Quit: *hums* Can't stay now!] |
| 06:01 | < Rhamphoryncus> | I suspect that ISO 10646 defines no special handling of the byte order mark. I'm still trying to determine that though |
| 06:02 | < Rhamphoryncus> | It doesn't help that the BOM is a perfectly legal scalar value anyway, and thus SHOULD show up in the resulting text :P |
| 06:05 | < Rhamphoryncus> | and for UTF-8 it's a defect in the standard. An encoder that chooses to emit it removes the feature of being an ascii superset |
| 06:08 | < Rhamphoryncus> | direct quote from ISO 10646, if there was any doubt: "NOTE 1 - Since UCS-2 only contains the repertoire of the BMP it is not fully interoperable with UCS-4, UTF-8 and UTF-16." |
| 06:12 | < Rhamphoryncus> | "NOTE 2 - When confined to the code positions in Planes 00 to 10 (U+0000 to U+10FFFF), UCS-4 is also referred to as UCS Transformation Format 32 (UTF-32). The Unicode Standard, Version 4.0, defines the following forms of UTF- 32:" |
| 06:20 | | Serah [Z@2C3C9C.B2A300.F245DE.859909] has quit [Ping timeout: 121 seconds] |
| 06:20 | < Rhamphoryncus> | huh. So if I use UTF-32 internally I'm actually switching between UTF-32BE and UTF-32LE depending on platform |
| 06:21 | < Rhamphoryncus> | (and hypothetically, a nonstandard variant of UTF-32 on platforms with weird byte order) |
| 06:53 | | Serah [Z@5E691D.FC7C16.7074BE.9ADB29] has joined #code |
| 07:12 | | Serah [Z@5E691D.FC7C16.7074BE.9ADB29] has quit [Ping timeout: 121 seconds] |
| 07:33 | | Serah [Z@3A600C.A966FF.5BF32D.8E7ABA] has joined #code |
| 07:51 | | Derakon[AFK] is now known as Derakon |
| 08:15 | | Derakon is now known as Derakon[AFK] |
| 08:28 | | Serah [Z@3A600C.A966FF.5BF32D.8E7ABA] has quit [Ping timeout: 121 seconds] |
| 08:38 | | Rhamphoryncus [rhamph@Nightstar-8931f88f.abhsia.telus.net] has quit [Client exited] |
| 08:44 | | You're now known as TheWatcher |
| 09:17 | | Vornicus is now known as Vornicus-Latens |
| 09:19 | | * McMartin gets his beta report from Vorn, now has 60 action items |
| 09:19 | <@McMartin> | Thanks~ |
| 09:19 | <@TheWatcher> | OW >.> |
| 09:21 | <@McMartin> | This is why we have testers |
| 09:21 | <@McMartin> | This is also why we *start* testing four and a half months in advance |
| 09:22 | <@TheWatcher> | What're you doing, anyway? |
| 09:22 | <@McMartin> | Interactive fiction game |
| 09:22 | <@TheWatcher> | Aah, okay |
| 09:23 | <@McMartin> | So yeah, a good number of these are going to be a sentence or two to fix. |
| 09:23 | <@McMartin> | Example: Does the player mean entering the belt: it is completely impossible.[* Seriously, what the fuck, it's like the first choice ALL THE TIME even when the room is full of chairs] |
| 09:24 | <@McMartin> | Hm, actually, I don't recall if completely impossible is supported in this version yet. |
| 09:24 | <@McMartin> | It might just have to be "very unlikely" |
| 09:34 | <@McMartin> | Among the other bug reports: "Actually, most bromine compounds are solids" |
| 09:44 | <@jerith> | :-) |
| 09:45 | <@jerith> | McMartin: Is this for ifcomp? |
| 10:43 | <@McMartin> | With luck! |
| 10:43 | <@McMartin> | It was supposed to be for *last* year's IFComp, that didn't really happen |
| 10:43 | | * McMartin fixes a dozen issues, calls it a night |
| 10:57 | | AnnoDomini [annodomini@Nightstar-c069db15.adsl.tpnet.pl] has joined #code |
| 10:57 | | mode/#code [+o AnnoDomini] by Reiver |
| 12:02 | | Reiv[Graduate] [orthianz@Nightstar-9b1f5915.xnet.co.nz] has quit [Ping timeout: 121 seconds] |
| 12:06 | | Reiver [reaverta@ServerAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds] |
| 12:19 | | Reiver [reaverta@ServerAdministrator.Nightstar.Net] has joined #code |
| 12:19 | | mode/#code [+qo Reiver Reiver] by ChanServ |
| 12:28 | <@AnnoDomini> | Is there a way to make folders in Gnome under Debian behave like Windows folders, in that opening one will bring it up in the same window, rather than summoning a new one? |
| 12:28 | <@jerith> | Knowing Gnome, I wouldn't be surprised if they took that option away because it's "too complicated" or something. :-/ |
| 12:47 | < Tarinaky> | Finished Effective C++. |
| 12:47 | < Tarinaky> | AnnoDomini: You could try using a different File Manager. |
| 12:58 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has joined #code |
| 14:35 | | Reiv[Graduate] [orthianz@Nightstar-7aed6a6b.xnet.co.nz] has joined #code |
| 14:47 | <@ToxicFrog> | AnnoDomini: edit, preferences, behaviour, enable "browser windows" |
| 15:01 | | celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has joined #code |
| 15:36 | < Tarinaky> | I have a question. How easy would it be to set up, in SDL, a system where portraits are generated by recolouring layers? |
| 15:37 | < Tarinaky> | So having a stock shape and storing noses and eyes and lips and whatever else as seperate layers and blitting them ontop. |
| 15:37 | < Tarinaky> | But how would you go about applying colour filters? |
| 15:40 | < gnolam> | Answer to question 1: really easy. |
| 15:40 | < Tarinaky> | Including the recolouring? |
| 15:41 | < gnolam> | What kind of recoloring did you have in mind? |
| 15:41 | < Tarinaky> | Well. Largely depends on what I can get away with while keeping it looking 'okay'. |
| 15:42 | < Tarinaky> | As a way of increasing the number of permutations without increasing workload. |
| 15:43 | < Tarinaky> | But it'd also be cool if the colour of the entire portrait... in a sort of... Dorian Grey/Gothic way for Vampires. |
| 15:43 | < Tarinaky> | *entire portrait could be muted |
| 15:43 | < gnolam> | I think SDL_gfx can do some basic tinting, but otherwise you can always do what you want manually. |
| 15:46 | < Tarinaky> | My knowledge of image manipulation is limited though which rather makes me paranoid about what I can/can't achieve :/ |
| 16:14 | | Zed [Zed@Nightstar-e4835f03.or.comcast.net] has joined #code |
| 16:28 | | cpux [Moo@Nightstar-20a84089.dyn.optonline.net] has quit [[NS] Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]] |
| 16:32 | < Tarinaky> | Any suggestions on maybe some reading material on how to do colour filters and stuff... bitwise so to speak? |
| 16:55 | | Derakon[AFK] is now known as Derakon |
| 17:19 | < Alek> | aw crap, it's still here. |
| 17:19 | < Alek> | the justin.tv popup. |
| 17:19 | < Alek> | 5 pages overnight. |
| 17:19 | < Alek> | halo3forum, specifically, Halo 3 Mythic Brawl with Kyuure |
| 17:19 | | Tarinaky [Tarinaky@Nightstar-d9b9d891.adsl.virginmedia.net] has quit [Connection closed] |
| 17:30 | < gnolam> | Tarinaky: well, that depends on what you want to do. Image processing in general is not trivial (but it might be, depending on what you want to do). |
| 17:30 | < gnolam> | Aw crap, he quit. |
| 18:10 | | Vornicus-Latens is now known as Vornicus |
| 18:34 | | Serah [Z@3A600C.A966FF.5BF32D.8E7ABA] has joined #code |
| 18:42 | | Tarinaky [Tarinaky@Nightstar-d9b9d891.adsl.virginmedia.net] has joined #code |
| 19:19 | <@AnnoDomini> | ToxicFrog: Okay, thanks. |
| 19:23 | | RichardBarrell [mycatverbs@Nightstar-58acb782.cable.virginmedia.com] has quit [Ping timeout: 121 seconds] |
| 20:31 | | aoanla [AndChat@Nightstar-8901103f.range217-42.btcentralplus.com] has joined #code |
| 20:34 | | aoanla [AndChat@Nightstar-8901103f.range217-42.btcentralplus.com] has quit [[NS] Quit: ] |
| 20:55 | | Serah [Z@3A600C.A966FF.5BF32D.8E7ABA] has quit [Ping timeout: 121 seconds] |
| 21:01 | | AnnoDomini [annodomini@Nightstar-c069db15.adsl.tpnet.pl] has quit [Ping timeout: 121 seconds] |
| 21:03 | | AnnoDomini [annodomini@Nightstar-c19d06d6.adsl.tpnet.pl] has joined #code |
| 21:03 | | mode/#code [+o AnnoDomini] by Reiver |
| 21:04 | < celticminstrel> | 4:04 - this time does not exist! :D |
| 21:18 | | cpux [Moo@Nightstar-20a84089.dyn.optonline.net] has joined #code |
| 21:20 | <@McMartin> | Ah, I7 wackiness |
| 21:20 | <@McMartin> | http://pastebin.com/cWzQYUm9 |
| 21:21 | <@McMartin> | As for SDL: if you want to use paletting, it's trivial, but you probably don't want to use paletting. |
| 21:21 | < celticminstrel> | Why not? |
| 21:21 | <@McMartin> | Because it restricts you to 256 colors |
| 21:21 | < celticminstrel> | Sounds reasonable. |
| 21:21 | <@McMartin> | Hasn't been for a decade. |
| 21:22 | < celticminstrel> | Hasn't been what? Reasonable? |
| 21:22 | <@McMartin> | Indeed. |
| 21:22 | <@McMartin> | Makes effects like alpha blending impossible, for one. |
| 21:22 | <@TheWatcher> | 256 colours is, frankly, laughable unless you're writing dwarf fortress~ |
| 21:23 | < celticminstrel> | Does that impossibility imply if you're copying a paletted image onto a non-paletted surface? |
| 21:23 | <@McMartin> | You can do tricks like UQM used to and have a different... yeah, you can do that |
| 21:23 | < celticminstrel> | Have a different what> |
| 21:23 | < celticminstrel> | ? |
| 21:23 | <@McMartin> | A different palette per blittable, onto a surface that's true-color |
| 21:23 | <@McMartin> | However, then you're doing a graphic conversion on every blit and are thus spending 2005-level computer power to render 1993-era graphics. |
| 21:24 | <@McMartin> | As for pixel-by-pixel changes. |
| 21:25 | <@McMartin> | The way you do that is by locking the surface with SDL_LockSurface, editing the SDL_Surface->data member (whose format is specified by its "format" specifier), and then unlocking it with SDL_UnlockSurface. |
| 21:25 | <@McMartin> | Unless you're using a carefully designed format that's easy to edit, this can be something of a nightmare |
| 21:25 | <@McMartin> | The easiest first thing to do is to instead of applying color filters, just increase the number of sprites you have. Memory is often cheaper than CPU these days. |
| 21:27 | <@Vornicus> | unless you /actually/ need to dynamically generate the colors, just make more sprites. |
| 21:27 | < celticminstrel> | By applying colour filters you're referring to things like swapping the red and green channels? |
| 21:28 | <@Vornicus> | And if you /do/ need to do the dynamic generation, do it /once/, with /all/ the sprites you need to do it on, and use them for the entire level. |
| 21:28 | <@McMartin> | it sounded like you meant things like applying sepia tones. |
| 21:28 | <@McMartin> | Or turning skin color from light brown to dark brown |
| 21:28 | <@Vornicus> | (in any case, you can do this using a color convolution matrix and two images, one that's just the palette-zone colors. |
| 21:28 | < celticminstrel> | Yeah, pre-generating any colour-filtered sprites sounds like a good idea. |
| 21:59 | | Serah [Z@2C3C9C.B2A300.F245DE.859909] has joined #code |
| 23:31 | | You're now known as TheWatcher[T-2] |
| 23:33 | | You're now known as TheWatcher[zZzZ] |
| 23:39 | | * Derakon dives back into the Jetblade source code. |
| 23:39 | <@Derakon> | I'd forgotten that the map module is 1400 lines long... |
| 23:39 | <@Derakon> | Might be time to split that up a bit. |
| 23:52 | | AnnoDomini [annodomini@Nightstar-c19d06d6.adsl.tpnet.pl] has quit [[NS] Quit: That's gneiss.] |
| 23:52 | | Tarinaky [Tarinaky@Nightstar-d9b9d891.adsl.virginmedia.net] has quit [Ping timeout: 121 seconds] |
| 23:57 | < Alek> | wooo jetblade. |
| --- Log closed Sun May 09 00:00:56 2010 |