| --- Log opened Wed Mar 28 00:00:58 2007 |
| 00:01 | | Thaqui [~Thaqui@Nightstar-12000.jetstream.xtra.co.nz] has joined #code |
| 00:01 | | mode/#code [+o Thaqui] by ChanServ |
| 00:17 | | Jo}{n is now known as EvilDarkLord |
| --- Log closed Wed Mar 28 00:45:44 2007 |
| --- Log opened Wed Mar 28 00:45:49 2007 |
| 00:45 | | TheWatcher[afk] [~chris@Nightstar-29731.dsl.in-addr.zen.co.uk] has joined #code |
| 00:45 | | Irssi: #code: Total of 20 nicks [8 ops, 0 halfops, 0 voices, 12 normal] |
| 00:46 | | Irssi: Join to #code was synced in 28 secs |
| 00:46 | | Sy [Syloq@Nightstar-8008.hsd1.or.comcast.net] has left #code [] |
| 00:58 | | ReivSLEP [~reaverta@IRCop.Nightstar.Net] has joined #Code |
| 00:58 | | Syloq [Syloq@Admin.Nightstar.Net] has joined #code |
| 01:00 | | ReivSLEP is now known as Reiv |
| 01:54 | | * Raif eats it. |
| 02:18 | | McMartin[TheCity] is now known as McMartin |
| 02:22 | | Thaqui [~Thaqui@Nightstar-12000.jetstream.xtra.co.nz] has left #code [Leaving] |
| 02:25 | | gnolam [Lenin@Nightstar-13557.8.5.253.se.wasadata.net] has quit [Quit: Z?] |
| 02:46 | | Chalcedon [Chalceon@Nightstar-2422.dialup.ihug.co.nz] has joined #code |
| 02:46 | | mode/#code [+o Chalcedon] by ChanServ |
| 03:02 | | Chalcedon [Chalceon@Nightstar-2422.dialup.ihug.co.nz] has quit [Quit: gone] |
| 03:10 | | Janus [~Cerulean@Nightstar-10302.columbus.res.rr.com] has quit [Quit: Jouets de Dieu, jouets de jouets, les jouets de me, na?tre Clair enfant voire.] |
| 04:23 | | Chalcedon [Chalceon@Nightstar-6532.dialup.ihug.co.nz] has joined #code |
| 04:23 | | mode/#code [+o Chalcedon] by ChanServ |
| 04:27 | | Vornicus-Latens [~vorn@Admin.Nightstar.Net] has quit [Ping Timeout] |
| 04:32 | | Vornicus-Latens [~vorn@67.50.40.ns-3674] has joined #code |
| 04:33 | | Chalcy [Chalceon@Nightstar-6532.dialup.ihug.co.nz] has joined #code |
| 04:33 | | mode/#code [+o Chalcy] by ChanServ |
| 04:34 | | ChalcyGone [Chalceon@Nightstar-6532.dialup.ihug.co.nz] has joined #code |
| 04:34 | | Chalcedon [Chalceon@Nightstar-6532.dialup.ihug.co.nz] has quit [Killed (NickServ (GHOST command used by ChalcyGone))] |
| 04:35 | | Chalcy [Chalceon@Nightstar-6532.dialup.ihug.co.nz] has quit [Killed (NickServ (GHOST command used by ChalcyGone))] |
| 04:35 | | ChalcyGone is now known as Chalcedon |
| 04:57 | | Chalcedon [Chalceon@Nightstar-6532.dialup.ihug.co.nz] has quit [Quit: gone] |
| 05:00 | | Reiv is now known as ReivClass |
| 07:42 | | MahalWork is now known as Mahal |
| 07:42 | | Mahal is now known as MahalWork |
| 07:42 | | MahalWork is now known as Mahal |
| 07:44 | <@Raif> | Vorn: Halflife 2 had dynamic skeletal... and some very cool interpolation techniques. |
| 07:45 | | ReivClass is now known as Reiv |
| 10:10 | | Vornicus-Latens is now known as Vornicus |
| 11:00 | | Thaqui [~Thaqui@Nightstar-12000.jetstream.xtra.co.nz] has joined #code |
| 11:00 | | mode/#code [+o Thaqui] by ChanServ |
| 12:22 | | AnnoDomini [~farkoff@Nightstar-29600.neoplus.adsl.tpnet.pl] has quit [Ping Timeout] |
| 12:22 | | KBot [~karma.bot@Nightstar-29708.neoplus.adsl.tpnet.pl] has joined #Code |
| 12:23 | < KBot> | KarmaBot v1.19. online and ready. Type "!help commands" for command list. |
| 12:23 | | KarmaBot [~karma.bot@Nightstar-29600.neoplus.adsl.tpnet.pl] has quit [Ping Timeout] |
| 12:23 | | KBot is now known as KarmaBot |
| 12:27 | | AnnoDomini [~farkoff@Nightstar-29708.neoplus.adsl.tpnet.pl] has joined #Code |
| 12:27 | | mode/#code [+o AnnoDomini] by ChanServ |
| 13:30 | | AnnoDomini [~farkoff@Nightstar-29708.neoplus.adsl.tpnet.pl] has quit [Killed (NickServ (GHOST command used by AbuDhabi))] |
| 13:30 | | AnnoDomini [~farkoff@Nightstar-29490.neoplus.adsl.tpnet.pl] has joined #Code |
| 13:31 | | KarmaBot [~karma.bot@Nightstar-29708.neoplus.adsl.tpnet.pl] has quit [Ping Timeout] |
| 13:39 | | KarmaBot [~karma.bot@Nightstar-29490.neoplus.adsl.tpnet.pl] has joined #Code |
| 13:40 | < KarmaBot> | KarmaBot v1.19. online and ready. Type "!help commands" for command list. |
| 13:56 | < AnnoDomini> | "How many C programmers does it take to change a lightbulb?" "None. They forgot to declare it." |
| 13:56 | < AnnoDomini> | :P |
| 14:09 | | AnnoDomini [~farkoff@Nightstar-29490.neoplus.adsl.tpnet.pl] has quit [Ping Timeout] |
| 14:10 | | KBot [~karma.bot@Nightstar-29165.neoplus.adsl.tpnet.pl] has joined #Code |
| 14:10 | | KarmaBot [~karma.bot@Nightstar-29490.neoplus.adsl.tpnet.pl] has quit [Ping Timeout] |
| 14:12 | | KarmaBot [~karma.bot@Nightstar-29165.neoplus.adsl.tpnet.pl] has joined #Code |
| 14:12 | < KarmaBot> | KarmaBot v1.19. online and ready. Type "!help commands" for command list. |
| 14:13 | | KBot [~karma.bot@Nightstar-29165.neoplus.adsl.tpnet.pl] has quit [Ping Timeout] |
| 14:14 | | AnnoDomini [~farkoff@Nightstar-29165.neoplus.adsl.tpnet.pl] has joined #Code |
| 14:14 | | mode/#code [+o AnnoDomini] by ChanServ |
| 16:30 | | Reiv is now known as ReivZzz |
| 16:40 | | * Vornicus fiddles with database stuff. |
| 16:55 | | Reiv [~reaverta@IRCop.Nightstar.Net] has joined #Code |
| 16:57 | | * ToxicFrog fiddles with a set of Lua functions for packing and unpacking binary data |
| 16:57 | | * Reiv ponders if there'd somehow be a way to tie 'middle click' to a keyboard combo. |
| 16:57 | | ReivZzz [~reaverta@IRCop.Nightstar.Net] has quit [Ping Timeout] |
| 16:58 | < ToxicFrog> | Hmm. I have bcdfimsuxz, what else do I need? |
| 17:01 | < Vornicus> | TF: expand those, would you. |
| 17:01 | < ToxicFrog> | (that's boolean, character (synonym for s1), double, float, signed int, bitmask, string/arbitrary data, unsigned int, skip, null-terminated/padded string |
| 17:01 | < Vornicus> | http://docs.python.org/lib/module-struct.html |
| 17:02 | < ToxicFrog> | Yeah, I'm avoiding the "let's have eighteen different characters for 'int'" thing. |
| 17:03 | < Vornicus> | http://www.ruby-doc.org/core/classes/Array.src/M002245.html |
| 17:03 | < ToxicFrog> | Instead you just append the width you want. |
| 17:03 | < ToxicFrog> | Ie, |
| 17:03 | < ToxicFrog> | Ie, 's124' is a 124-character string, 'u4' is a uint32_t, 'i2' is a signed short, etc. |
| 17:03 | < Vornicus> | AH |
| 17:03 | < ToxicFrog> | This also means you can read 24-bit ints with u3 and i3. |
| 17:04 | < Vornicus> | That is more sensical by half. |
| 17:04 | < ToxicFrog> | Which python's struct() does not permit, tsk tsk. |
| 17:04 | < Reiv> | Apparently "Middle Mouse Button Keyboard Shortcut Binding XP" is not a very useful string to search for in Google. |
| 17:04 | < Reiv> | Then again, maybe 4AM is a bad time to become obsessed with fixing a problem like this. |
| 17:05 | < ToxicFrog> | 'sN' reads exactly N bytes; 'zN' reads exactly N bytes, but returns only the data up to the first nul; 'z' reads up to the first nul. |
| 17:05 | < ToxicFrog> | And so forth. |
| 17:06 | < Vornicus> | But by the same token, there are no less than four floating point types. |
| 17:06 | < ToxicFrog> | There are? |
| 17:06 | < Vornicus> | float, double, long double, and quad. |
| 17:06 | < ToxicFrog> | ...bugger. |
| 17:06 | < Vornicus> | The latter two are rare, but I have seen at least one program that does long doubles. |
| 17:06 | < ToxicFrog> | And with those, you can't just specify an arbitrary width. |
| 17:07 | < ToxicFrog> | (I only need floats for this, but someone else will almost certainly want doubles) |
| 17:07 | < Vornicus> | Eh. Not a big deal; they're rare enough that most languages do not handle them. |
| 17:08 | < ToxicFrog> | I'm actually writing this for libss1edit, but it's generally useful enough that I'll release it as a seperate lib. |
| 17:08 | < ToxicFrog> | And small enough that I probably will actually document and release it. |
| 17:08 | < ToxicFrog> | Bah, the Python one doesn't support booleans or bitmasks either. |
| 17:09 | < Vornicus> | Python doesn't really know what a bitmask is. I don't know why it doesn't support booleans though |
| 17:09 | | * Reiv arg. |
| 17:10 | < Reiv> | Is it really so impossible to imagine that someone out there hasn't written a program to get a keyboard shortcut that acts as a middle-mouse-button-click? |
| 17:10 | < ToxicFrog> | ('bN' is basically 'uN', then it returns (val ~= 0)) |
| 17:11 | < Vornicus> | ~=? |
| 17:11 | < ToxicFrog> | Lua's inequality operator. |
| 17:11 | < Vornicus> | ah |
| 17:11 | < ToxicFrog> | Equivalently (not (val == 0)) |
| 17:12 | < ToxicFrog> | Reiv: I don't even know if binding keystrokes to mouse clicks is generally possible in windows, although it should be. |
| 17:12 | < ToxicFrog> | You may be looking in the wrong places, though. Look for macro software, then create a macro of middle-click and bind it. |
| 17:13 | < ToxicFrog> | Vornicus: Lua doesn't know what a bitmask is, but I still want to be able to extract one from binary files. |
| 17:13 | < ToxicFrog> | It returns it as a list of booleans. |
| 17:14 | < Vornicus> | ah |
| 17:16 | < ToxicFrog> | For that matter, python's struct() will read pointers, but not bitmasks? |
| 17:16 | < ToxicFrog> | Weirdness. |
| 17:16 | < Vornicus> | I don't quite get that. |
| 17:17 | | * ToxicFrog ponders the Ruby one |
| 17:17 | < ToxicFrog> | Ok, I'm totally ganking @ |
| 17:17 | < ToxicFrog> | And this has made me aware I need some way to specify endianness. |
| 17:19 | < Vornicus> | Python's does pretty good on that one. |
| 17:19 | < ToxicFrog> | How does it handle it? |
| 17:19 | < Vornicus> | First character in a pack string defines the endianness. |
| 17:20 | < Vornicus> | If it is in @ = < > ! |
| 17:20 | < Vornicus> | In the doc I sent you, it's about 2/3 of the way down the page. |
| 17:22 | < ToxicFrog> | Hmm. I'll probably steal that, too. |
| 17:22 | < ToxicFrog> | Mine distinguishes between width and count, too. |
| 17:22 | < ToxicFrog> | u3,4 is four 24bit unsigned ints. |
| 17:22 | < Vornicus> | That's pretty cool. |
| 17:23 | < ToxicFrog> | Putting it in front looks better - 4u3 - but then we have to mandate whitespace between format specifiers. |
| 17:23 | < Vornicus> | I prefer your way, I think |
| 17:23 | < ToxicFrog> | To me, u3,4 looks like a range of 3 to 4. |
| 17:23 | < ToxicFrog> | Whereas 4u3 looks like 4*u*3 |
| 17:24 | < Vornicus> | u3*4 |
| 17:24 | < ToxicFrog> | That works. |
| 17:24 | | * ToxicFrog updates the formal grammar |
| 17:26 | < ToxicFrog> | FORMAT = ALIGN FIELD FORMAT | e |
| 17:26 | < ToxicFrog> | ALIGN = '=' | '<' | '>' | e |
| 17:26 | < ToxicFrog> | FIELD = typechar WIDTH COUNT |
| 17:26 | < ToxicFrog> | WIDTH = integer | e |
| 17:26 | < ToxicFrog> | COUNT = '*' integer | e |
| 17:26 | < ToxicFrog> | ...whoops |
| 17:26 | < ToxicFrog> | FORMAT = ALIGN FIELDS |
| 17:26 | < ToxicFrog> | FIELDS = FIELD FIELDS | e |
| 17:27 | < ToxicFrog> | Although that still looks weird to me. u(3*4) instead of 4(u3). Maybe I'll just have it accept both. |
| 17:29 | | * ToxicFrog also picks a for align-to, which will be necessary for the resource file handler |
| 17:30 | < ToxicFrog> | So, now we have: data types boolean, float, double, signed int, bitmask, string/data, unsigned int, cstring |
| 17:31 | < ToxicFrog> | And control types endianness, seek-offset, align-width, and skip/pad |
| 17:31 | < ToxicFrog> | ...hmm. |
| 17:31 | | * ToxicFrog fiddles |
| 17:32 | < ToxicFrog> | Ok, @ is absolute seek, + is relative seek forward, - is relative seek backward |
| 17:32 | < Vornicus> | + is a synonym for skip? |
| 17:32 | < ToxicFrog> | x is equivalent to + when reading, except that it actually reads (and discards) data, and will thus work on non-seekable streams |
| 17:33 | < ToxicFrog> | When writing, + is 'call :seek()', x is 'write nulls' |
| 17:33 | < Vornicus> | aha |
| 17:35 | < ToxicFrog> | ('xN' thus becomes equivalent to 'zN',"") |
| 17:35 | < ToxicFrog> | (but is more convenient) |
| 17:36 | < ToxicFrog> | The idea here is that when calling destruct(), you can pass it either a string (which it treats as a buffer) or something else, which it will assume is a file-like object and use :read() and :seek() on and so forth. |
| 17:37 | < ToxicFrog> | And thus it will work on any file- (or, as long as you don't try to seek, stream-) -like object, as long as it uses the same API as Lua's io library. |
| 17:38 | < ToxicFrog> | For construct(), if the first argument is the format string, it constructs and returns a buffer; otherwise it once again assumes it's a file-object and tries to talk to it directly. |
| 17:39 | < ToxicFrog> | So you can go, say, construct(fout, "z124u4", "LG RES file v3", toc_offs) and it will drop the data directly into the file. |
| 17:50 | < Vornicus> | Arg! I also want to make an X-Com-alike! I should stop gaming, or something. |
| 17:54 | < Reiv> | Vorn: Tell you what. |
| 17:54 | < Reiv> | Finish Settlers of Catan. |
| 17:54 | < Reiv> | Then, look into that X-Com remake folks are working on, see if there's Any Hope For It. |
| 17:54 | < Reiv> | And then |
| 17:54 | < Reiv> | Set up a project to remake Master of Orion. |
| 17:54 | < Vornicus> | Heh |
| 17:55 | < Reiv> | (I'm assuming it's source code is Horribly Arcane and Unable To Be Cracked?) |
| 17:55 | < Vornicus> | heh |
| 17:55 | < Vornicus> | I'm not going to break open the MoO source code, that would not be kosher. |
| 17:57 | < Reiv> | 1) No? |
| 17:58 | | Reiver [~reiver@Nightstar-28072.ubs-dsl.xnet.co.nz] has quit [Ping Timeout] |
| 17:58 | < Reiv> | 2) I was more thinking along the lines of if you wanted to get a carbon-copy-esque game, it would at least let you see what things did which without exhaustive statistical testing. |
| 17:58 | < Vornicus> | Actually the "physics" of MoO are pretty clear. |
| 18:00 | < Vornicus> | The map generation - what rarity various planets and stars are - and the exact population growth function are the only things I don't know yet. |
| 18:04 | < Vornicus> | oh, and the exact mechanics of how a race's tech tree is chosen. |
| 18:05 | | Reiver [~reiver@Nightstar-28072.ubs-dsl.xnet.co.nz] has joined #Code |
| 18:07 | | * Vornicus fiddles with his database design. |
| 18:22 | | gnolam [Lenin@Nightstar-13557.8.5.253.se.wasadata.net] has joined #Code |
| 18:22 | | mode/#code [+o gnolam] by ChanServ |
| 18:37 | | * Vornicus fiddlefiddles |
| 18:38 | <@AnnoDomini> | * Rome burnsburns. |
| 19:50 | | gnolam [Lenin@Nightstar-13557.8.5.253.se.wasadata.net] has quit [Quit: Reboot] |
| 20:09 | | gnolam [Lenin@Nightstar-13557.8.5.253.se.wasadata.net] has joined #Code |
| 20:10 | | mode/#code [+o gnolam] by ChanServ |
| 20:28 | | Mahal is now known as MahalShower |
| 20:45 | | MahalShower is now known as Mahal |
| 20:55 | | Mahal is now known as MahalWork |
| 21:05 | < Vornicus> | gnarg. ORMs that essentially demand you follow their rules suck. |
| 21:23 | <@gnolam> | ORMs? |
| 21:24 | < Vornicus> | Object-Relational Models |
| 21:25 | < Vornicus> | The thing you use to talk to a database and turn the stuff you get from there into language objects. |
| 21:25 | <@gnolam> | Ah. Never encountered the abbreviation before. |
| 21:49 | < MyCatVerbs> | Mappings! |
| 21:49 | < MyCatVerbs> | gnolam: I am led to believe that it is the most insanely difficult PITA going. Or something. |
| 21:49 | < Vornicus> | Mappings, yeah |
| 21:50 | < Vornicus> | And, it is a very hard problem. So far the easiest one I've dealt with is Ruby on Rails' ActiveRecord, which still demands you put in a lot of stuff, it can't just read your schema files and make all the relationships go. |
| 21:50 | < MyCatVerbs> | AFAIK, the problem is that OO often leads people to build structures which resemble interestingly-shaped graphs, while RDMSes are amenable to rather flatter storage systems? |
| 21:51 | < MyCatVerbs> | The wierd thing is that this problem is probably easier in a shit language like C++ than a useful one like Ruby. Huh. |
| 21:53 | < Vornicus> | The current batch of ORMs have two problems. |
| 21:53 | < MyCatVerbs> | Jah? |
| 21:54 | < Vornicus> | 1. they assume too much about your schema |
| 21:54 | < Vornicus> | 2. they don't ask the schema enough. |
| 21:54 | < Vornicus> | So ActiveRecord, for instance, ignores foreign key relationships. |
| 21:54 | < MyCatVerbs> | ...what? Those both sound like implementation details, rather than essential complexity. Are you *sure* that's not just the tip of an iceberg? |
| 21:55 | < Vornicus> | These are the tip of the iceberg, but they're the problems I see in front of me. |
| 21:55 | < MyCatVerbs> | Ook. |
| 21:55 | < Vornicus> | I can't evaluate any other problems of the relationships without getting this crap out of the way. |
| 21:56 | < MyCatVerbs> | Ook. :/ |
| 21:57 | <@gnolam> | MyCatVerbs: oh, I know what it is. I just hadn't come across the abbreviation of it. :) |
| 21:57 | < MyCatVerbs> | gnolam: What others are there? |
| 21:58 | < Vornicus> | AR /is/ kinda nice, it has stuff that lets a database table act as a tree... but I still have to tell it everything about the relationships in my tables, and if I don't use /its specific table scheme/ I have to tell it everything about everything in my tables. |
| 22:04 | < Vornicus> | And Ruby does have the right kinds of toys in it - openable classes means that it, technically, should be able to figure out the tables at startup and add the appropriate methods to the classes in question. |
| 22:05 | < Vornicus> | It's just... too bad that more power is not being driven into making the orm understand more about the underlying database so I don't have to say things twice. |
| 22:33 | | MyCatVerbs [~mycatownz@Nightstar-13709.lurkingfox.co.uk] has quit [Client exited] |
| 22:37 | | gnolam [Lenin@Nightstar-13557.8.5.253.se.wasadata.net] has quit [Quit: Reinstalling OS.] |
| 23:08 | | MyCatVerbs [~mycatownz@Nightstar-13709.lurkingfox.co.uk] has joined #code |
| 23:57 | | Chalcedon [Chalceon@Nightstar-1800.dialup.ihug.co.nz] has joined #code |
| 23:57 | | mode/#code [+o Chalcedon] by ChanServ |
| --- Log closed Thu Mar 29 00:00:59 2007 |