| --- Log opened Wed Nov 03 00:00:26 2010 |
| 00:33 | | cpux [chatzilla@Nightstar-c978de34.dyn.optonline.net] has quit [Connection closed] |
| 00:38 | | cpux [chatzilla@Nightstar-c978de34.dyn.optonline.net] has joined #code |
| 00:43 | | Attilla [Some.Dude@Nightstar-31ae674d.threembb.co.uk] has quit [[NS] Quit: ] |
| 00:47 | | Orthia [orthianz@Nightstar-83bdff96.xnet.co.nz] has quit [Connection reset by peer] |
| 00:52 | | Orthia [orthianz@Nightstar-a885237f.xnet.co.nz] has joined #code |
| 00:54 | | Zed [Zed@Nightstar-556ea8b5.or.comcast.net] has joined #code |
| 01:02 | | Derakon[AFK] is now known as Derakon |
| 01:39 | < celticminstrel> | Okay, why does making my prompt bold break linewrapping? |
| 01:40 | < celticminstrel> | And history navigation, too. |
| 01:40 | < celticminstrel> | This is Mac Terminal, if anyone is wondering. |
| 01:48 | <@ToxicFrog> | Probably you've made a mistake while making it bold. |
| 01:48 | < celticminstrel> | I used $(tput bold) and $(tput sgr0). |
| 02:00 | < celticminstrel> | No suggestions? |
| 02:07 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has quit [[NS] Quit: Z?] |
| 02:10 | < celticminstrel> | This is entirely unrelated, but how do I get a link error in a .o file whose name corresponds to none of the input source files? :| |
| 02:11 | <@ToxicFrog> | I have no experience with tput, sorry |
| 02:12 | <@ToxicFrog> | What's the name? |
| 02:13 | < celticminstrel> | It's an insanel-- oh wait, you mean the object file name. |
| 02:14 | <@ToxicFrog> | Depending on what you're building and on one system, there may be one or more .o files that are linked in automatically, for example, the pre-main function |
| 02:14 | < celticminstrel> | It seems to vary each time the compiler is run. |
| 02:14 | < celticminstrel> | Most recent was cch5BGuT.o |
| 02:14 | <@ToxicFrog> | Oh, that looks like a randomly generated temporary filename! |
| 02:15 | <@ToxicFrog> | Are you using seperate compile and link steps? |
| 02:15 | < celticminstrel> | The missing symbols are insanely long template names. (I'm compiling Boost.Wave manually here, for info.) |
| 02:15 | <@ToxicFrog> | Or just going straight from source to binary? |
| 02:15 | < celticminstrel> | Um, I'm going from source to .o? |
| 02:15 | < celticminstrel> | And this happens on that step. |
| 02:15 | <@ToxicFrog> | That's not possible. |
| 02:15 | < celticminstrel> | Huh? |
| 02:15 | <@ToxicFrog> | Source -> .o is compilation, not linking; it is not possible for link errors to occur during this step. |
| 02:15 | < celticminstrel> | I know! |
| 02:16 | <@ToxicFrog> | Either it's not a link error, or you aren't doing what you think you are. |
| 02:16 | < celticminstrel> | That's partly why I'm confused. |
| 02:16 | <@ToxicFrog> | Paste the command you're using. |
| 02:16 | < celticminstrel> | Well, it says "undefined symbols". |
| 02:16 | < celticminstrel> | g++ $(WARNINGS) $(INCLUDES) cpp.cpp -o cpp.o 2>> error.log |
| 02:17 | <@ToxicFrog> | That's not compile, that's compile+link |
| 02:17 | < celticminstrel> | I thought you just said it's compiling. 9_9 |
| 02:17 | <@ToxicFrog> | You said that you were going from source to .o |
| 02:17 | < celticminstrel> | Which is what I'm doing. |
| 02:17 | <@ToxicFrog> | I am telling you that that's not what you're doing; you're going from source to program which you have a .o extension on. |
| 02:17 | <@ToxicFrog> | You're telling gcc to compile the file cpp.cpp, then link it into a program called "cpp.o" |
| 02:17 | < celticminstrel> | ...oh. |
| 02:18 | <@ToxicFrog> | As opposed to compile cpp.cpp to object code stored in cpp.o, which will be used for linking later. |
| 02:18 | <@ToxicFrog> | Try using the "-c" option. |
| 02:18 | < celticminstrel> | In any case, the problem still occurs when I specify the libraries to link to. |
| 02:18 | <@ToxicFrog> | I'm heading out now, but I can go into detail in an hour or so |
| 02:19 | < celticminstrel> | Still have the errors on the link stage... but now they are in cpp.o as I would expect. |
| 02:20 | < celticminstrel> | I would assume this means I just need to figure out which library I'm missing. |
| 02:20 | < celticminstrel> | ...or more likely it's that I haven't compiled the Wave source files. |
| 02:21 | | Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has quit [Connection closed] |
| 02:24 | | Orthia [orthianz@Nightstar-a885237f.xnet.co.nz] has quit [Client closed the connection] |
| 02:25 | | Orthia [orthianz@Nightstar-a885237f.xnet.co.nz] has joined #code |
| 02:25 | | Orthia [orthianz@Nightstar-a885237f.xnet.co.nz] has quit [Client closed the connection] |
| 02:33 | | Orthia [orthianz@Nightstar-a885237f.xnet.co.nz] has joined #code |
| 02:41 | < celticminstrel> | Well, I got undefined symbols down to just one. That's always good. |
| 02:41 | < celticminstrel> | I'm not sure what I did differently than last night, but it seems to be working a lot better now. |
| 02:42 | | Orthia [orthianz@Nightstar-a885237f.xnet.co.nz] has quit [Client closed the connection] |
| 02:44 | | Orthia [orthianz@Nightstar-a885237f.xnet.co.nz] has joined #code |
| 03:08 | <@ToxicFrog> | celticminstrel: is there a reason you're not using whatever build system (make, etc) Wave comes with? |
| 03:08 | <@ToxicFrog> | But yes, generally this means either you haven't compiled all the parts of the program you need to link, or you have but you're forgetting one or more libraries. |
| 03:09 | < celticminstrel> | Apparently, the final link error is because one one of the source files only includes the crucial part if a constant is 0, and the constant is defined as 1 in a header. |
| 03:10 | <@ToxicFrog> | This really does sound like "I'm trying to compile it by hand, but there is additional stuff that needs to be done that the build system normally handles" |
| 03:10 | < celticminstrel> | At least, it looks that way. |
| 03:12 | < celticminstrel> | Unless I read it wrong. |
| 03:13 | <@ToxicFrog> | -that- sounds like an optional feature that some files were compiled with and some without. |
| 03:13 | <@ToxicFrog> | <ToxicFrog> celticminstrel: is there a reason you're not using whatever build system (make, etc) Wave comes with? |
| 03:13 | < celticminstrel> | Yeah. |
| 03:14 | | Kaura [kaura@A2BA3E.5613D5.A7C357.372C7B] has joined #code |
| 03:14 | < Kaura> | Teh Vorn |
| 03:14 | <@Vornicus> | Teh Kaura |
| 03:14 | <@Vornicus> | I have actually forgotten where we were. |
| 03:15 | | Roscoe [Roscoe@5CB9D8.0096B4.EDCEE2.CEC675] has joined #code |
| 03:15 | < Kaura> | I have too, but you have really bad timing on this. I just did another round of Pills, like, five minutes ago. =/ |
| 03:16 | <@Vornicus> | Whups |
| 03:16 | <@Vornicus> | Sowwy |
| 03:16 | < Kaura> | Eh, nobody's fault but the damn drugs~ |
| 03:20 | <@ToxicFrog> | celticminstrel: And that would be? |
| 03:20 | < celticminstrel> | I want to fiddle with it. |
| 03:23 | <@Vornicus> | With the build setup? |
| 03:23 | <@Vornicus> | Because if you're just changing code, you don't need to change the build setup, unless you're doing drastic changes. |
| 03:24 | < celticminstrel> | I'm not entirely sure what changes I'm doing yet. |
| 03:26 | <@Vornicus> | Then use the already-built build system. There's no point trying to fuck with it until you know the system inside and out. |
| 03:26 | <@Vornicus> | Or, you know, it's obviously busted. |
| 03:26 | < celticminstrel> | I guess I can try that... |
| 03:27 | < celticminstrel> | Hm, duplicating the code from that one file results in a duplicate symbol error, so why do I get undefined symbol otherwise? |
| 03:28 | <@ToxicFrog> | ... |
| 03:28 | <@ToxicFrog> | Start with a clean source tree, commit it to version control, use the included build system to verify it builds, then start fiddling with it. |
| 03:29 | < celticminstrel> | Working on this. |
| 03:57 | < celticminstrel> | Well, I got it to link without using the Boost build system. |
| 04:00 | <@ToxicFrog> | Why are you so insistent on not using the build system it comes with? |
| 04:01 | < celticminstrel> | Because there's no easy way to isolate just the one library from the build system; if using the Boost build system it insists on also compiling the Boost libraries upon which it depends rather than just dynamically linking to them. |
| 04:02 | < celticminstrel> | I see no reason to rebuild Boost.Filesystem (for example) when it's already build on my system and I'm only fiddling with Boost.Wave. |
| 04:04 | <@ToxicFrog> | Aah. |
| 04:06 | < celticminstrel> | I did try, but I gave up after a few attempts. |
| 04:06 | < celticminstrel> | The only remaining issue (as far as I know) is that dyld can't find them. |
| 04:11 | < celticminstrel> | Which is probably because they're in an unusual location, which is easily fixed. |
| 04:13 | <@ToxicFrog> | ...what OS are you building on? |
| 04:14 | < celticminstrel> | Mac |
| 04:14 | < celticminstrel> | I installed them manually though. |
| 04:17 | <@ToxicFrog> | celticminstrel: also, you didn't try, say: |
| 04:17 | <@ToxicFrog> | ./bootstrap.sh --with-libraries=wave |
| 04:17 | <@ToxicFrog> | ./bjam |
| 04:17 | < celticminstrel> | ...true. |
| 04:18 | < celticminstrel> | I only tried bjam, having recently installed the latest version. |
| 04:19 | <@ToxicFrog> | I believe bjam itself also supports --with-libraries, going by its --help |
| 04:19 | <@ToxicFrog> | (presumably if you, say, ran bootstrap to prepare everything but don't want to build everything just yet) |
| 04:21 | < celticminstrel> | Yeah, I tried that, but it tried to build filesystem anyway. |
| 04:21 | < celticminstrel> | (Because Wave depends on Filesystem.) |
| 04:22 | <@ToxicFrog> | Aah. |
| 04:24 | <@Vornicus> | what's Wave? |
| 04:58 | | Stalker [Z@3A600C.A966FF.5BF32D.8E7ABA] has quit [Ping timeout: 121 seconds] |
| 05:14 | | celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!] |
| 05:28 | | Ortiha [orthianz@Nightstar-3e40c65b.xnet.co.nz] has joined #code |
| 05:31 | | Orthia [orthianz@Nightstar-a885237f.xnet.co.nz] has quit [Ping timeout: 121 seconds] |
| 05:36 | | Derakon is now known as Derakon[AFK] |
| 06:11 | | Ortiha [orthianz@Nightstar-3e40c65b.xnet.co.nz] has quit [[NS] Quit: Going dooooown...] |
| 06:15 | | cpux is now known as shade_of_cpux |
| 06:20 | | Orthia [orthianz@Nightstar-3e40c65b.xnet.co.nz] has joined #code |
| 06:33 | | Zed [Zed@Nightstar-556ea8b5.or.comcast.net] has quit [Client closed the connection] |
| 06:34 | | Zed [Zed@Nightstar-556ea8b5.or.comcast.net] has joined #code |
| 08:27 | | Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has joined #code |
| 08:35 | | Rhamphoryncus [rhamph@Nightstar-473f8685.abhsia.telus.net] has quit [Client exited] |
| 08:52 | | Vornicus is now known as Vornicus-Latens |
| 09:10 | | Kaura is now known as Kaura|zzz |
| 09:17 | | You're now known as TheWatcher |
| 09:23 | | Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has quit [Ping timeout: 121 seconds] |
| 09:53 | | Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has joined #code |
| 10:02 | | Roscoe [Roscoe@5CB9D8.0096B4.EDCEE2.CEC675] has quit [[NS] Quit: ] |
| 11:33 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has joined #code |
| 11:41 | | celticminstrel [celticminst@Nightstar-f8b608eb.cable.rogers.com] has joined #code |
| 11:46 | | celticminstrel [celticminst@Nightstar-f8b608eb.cable.rogers.com] has quit [Ping timeout: 121 seconds] |
| 13:43 | | Stalker [Z@26ECB6.A4B64C.298B52.D80DA0] has joined #code |
| 14:19 | | * ToxicFrog ponders event handling |
| 14:19 | <@ToxicFrog> | So, it works like this: when an event E is received by an object, it does the following: |
| 14:19 | <@ToxicFrog> | - call self:E_before |
| 14:20 | <@ToxicFrog> | - foreach child, repeat this process on the child |
| 14:20 | <@ToxicFrog> | - call self:E |
| 14:20 | <@ToxicFrog> | - if any of the above return true, immediately return true (without making further calls) |
| 14:20 | <@ToxicFrog> | - return false |
| 14:21 | <@ToxicFrog> | So, event handlers return true to mean "I've handled this" or false for the converse. |
| 14:21 | <@ToxicFrog> | Now, however, I find myself needing a third option: "my children shouldn't see this event, but I haven't handled it myself" |
| 14:22 | <@ToxicFrog> | At present I've just rigged it so that a return of "skip" means this, but that feels ugly and hackish. |
| 15:48 | | Anno[Laptop] [annodomini@Nightstar-09acd4e7.adsl.tpnet.pl] has joined #code |
| 15:49 | | Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has quit [Connection closed] |
| 15:49 | | celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has joined #code |
| 16:25 | | Derakon [Derakon@Nightstar-1ffd02e6.ucsf.edu] has joined #code |
| 16:26 | | mode/#code [+o Derakon] by Reiver |
| 16:28 | | * Derakon mutters at printf, tries to get right-justification on floats. |
| 16:28 | <@Derakon> | "%4d" % 12 prints " 12", but "%4.4f" % 12.34 prints "12.3400" |
| 16:28 | <@Derakon> | Oh wait, right. |
| 16:28 | <@Derakon> | The first number is the width allocated to the entire number, not to the bits to the left of the decimal sign. |
| 16:29 | <@Derakon> | Thanks~! |
| 16:29 | < celticminstrel> | First number is width. Second number is precision. |
| 16:29 | < celticminstrel> | ...who're you thanking? |
| 16:29 | <@Derakon> | The channel in general. |
| 16:30 | < celticminstrel> | I don't remember whether right-justification is default or '-'. |
| 16:30 | <@Derakon> | Never underestimate the value of an audience when you have a problem to solve. |
| 16:30 | <@Derakon> | Right-justified is default. |
| 16:30 | <@Derakon> | Putting a '-' after the % makes it left-justified. |
| 16:31 | | kwsn [kwsn@Nightstar-1def9d24.dyn.centurytel.net] has quit [[NS] Quit: BEEP BEEP IMMA JEEP] |
| 16:31 | < celticminstrel> | See, I just look it up on cplusplus.com or man when I need to use it. |
| 16:32 | <@Derakon> | It's important to know the basics off the top of your head, though, so you at least know what is possible. |
| 16:32 | < celticminstrel> | I do know the basics. |
| 16:32 | <@TheWatcher> | Ah, #code: problem solving via Genius Loci >.> |
| 16:32 | < celticminstrel> | Hehe. |
| 16:41 | < celticminstrel> | Does 'port list' always return an up-to-date list or is there a command to update the list? |
| 16:48 | < Anno[Laptop]> | I'm trying to use a third-party application to connect to the internet using a cellular USB modem. It *seems* to work, since the application reports no errors when I select the configuration I believe to be correct, but I don't get a ppp interface from ifconfig when I check, and the last time I disconnected from my LAN to check, I couldn't browse the internet. Could it be that wicd or something is interfering with the cellular ... |
| 16:48 | < Anno[Laptop]> | ... access? |
| 17:03 | <@ToxicFrog> | Hrm. |
| 17:03 | <@ToxicFrog> | The chess module is written so that it creates a bunch of objects, then assembles them into a heirarchy with add(). |
| 17:03 | <@ToxicFrog> | This created problems because add() wasn't propagating, so if chess was loaded, no-one else saw the board or pieces - they were created but unattached. |
| 17:05 | <@ToxicFrog> | I made add() propagate, but that broke drop, because drop uses moveto which uses add/remove. |
| 17:06 | <@ToxicFrog> | I could resolve this by making add() not propagate again, and having the chess module assemble the tree leaves-first, but this can get ugly, especially for larger modules. |
| 17:07 | <@ToxicFrog> | I could make add/remove propagate and rewrite drop instead to call those from the server, which may be a better approach - or rather, the general principle that "things that alter the gamestate always propagate, this is not transitive" |
| 17:07 | <@ToxicFrog> | (although this may greatly increase network traffic) |
| 17:08 | <@ToxicFrog> | Or I could add new API functions for propagated add or similar. |
| 17:24 | | Attilla [Some.Dude@Nightstar-869518dc.threembb.co.uk] has joined #code |
| 17:24 | | mode/#code [+o Attilla] by Reiver |
| 17:26 | <@Derakon> | Anyone care to examine the slides I've made for my presentation? http://derakon.dyndns.org/~chriswei/temp2/slides/ |
| 17:26 | <@Derakon> | (Click on the image to go to the next) |
| 17:29 | <@Derakon> | The overall goal is to provide examples of some basic operations in Python, to people who may know how to program but have never been formally trained in the art. |
| 17:31 | | * TheWatcher notes a problem on the first slide |
| 17:31 | <@TheWatcher> | it should be 'SCIENCE!' ¬¬ |
| 17:31 | <@Derakon> | Right. |
| 17:31 | <@TheWatcher> | Ahem. |
| 17:31 | <@Derakon> | Maybe I should write "Wissenschaft" in there? |
| 17:31 | | * TheWatcher continues |
| 17:32 | < celticminstrel> | The slide entitled "Array Slices" is in fact about string formatting. |
| 17:32 | <@Derakon> | Oh, I forgot to change the title when I copied the slide boilerplate. Thanks. |
| 17:33 | < celticminstrel> | Or the slide about string formatting is entitled "Array Slices". Either way works. |
| 17:34 | <@TheWatcher> | I assume the slide on MRC files will actually involve a definition of what MRC files are? |
| 17:35 | < celticminstrel> | The output on the actual slide on Array Slices includes the full array, but there is no corresponding print statement. |
| 17:35 | < celticminstrel> | That is, there is no "print foo" before the "print foo[0:5]". |
| 17:35 | <@Derakon> | TW: it's something my audience will know about. |
| 17:36 | <@TheWatcher> | fairynuff |
| 17:36 | <@Derakon> | They're the file format used to store images taken by our microscope. |
| 17:36 | <@Derakon> | Celtic: whups. |
| 17:36 | <@Derakon> | Thanks again. |
| 17:37 | <@Derakon> | ...unfortunately, I don't have room to insert a "print foo" line. ?.? |
| 17:40 | | * Derakon just removes the first line in the output section. |
| 17:44 | <@Derakon> | Hm, don't think I'll include any detailed commandline argument handling. |
| 17:44 | <@Derakon> | Also, Python's new string formatting system is nifty but only in 2.7 and later. |
| 17:45 | < celticminstrel> | What? |
| 17:45 | < celticminstrel> | The {} stuff? |
| 17:45 | <@Derakon> | Yeah. |
| 17:45 | <@Derakon> | As opposed to the printf-style formatting. |
| 17:45 | < celticminstrel> | I thought it was available in 2.6 |
| 17:46 | <@Derakon> | Oh, you're right. |
| 17:46 | <@Derakon> | Still, most of the code around here runs on 2.5. :\ |
| 17:46 | < celticminstrel> | Yeah, looks like it's not in 2.5. |
| 17:46 | <@ToxicFrog> | Hmm. |
| 17:47 | <@ToxicFrog> | Building the tree leaves-first result in it not rendering. |
| 17:47 | < celticminstrel> | Huh, I have five separate versions of Python installed, apparently. |
| 17:47 | <@ToxicFrog> | I think have a bug somewhere else now. |
| 18:00 | | * Derakon tries to figure out why his slides are 808x616 pixels instead of 800x600 pixels. |
| 18:03 | <@Derakon> | Ehh, nothing a little imagemagick cropping can't fix. |
| 18:08 | <@Derakon> | Anyway, any topics you think I should be covering that I missed? |
| 18:36 | | Orthia [orthianz@Nightstar-3e40c65b.xnet.co.nz] has quit [Connection reset by peer] |
| 18:40 | | Rhamphoryncus [rhamph@Nightstar-473f8685.abhsia.telus.net] has joined #code |
| 18:44 | <@Derakon> | Hey, Rhamph, care to take a look at the slideshow for the presentation I'm giving this Friday? It's introducing biologists to Python. |
| 18:44 | <@Derakon> | http://derakon.dyndns.org/~chriswei/temp2/slides/ |
| 18:44 | <@Derakon> | Click on each slide to advance to the next. |
| 18:44 | < Rhamphoryncus> | alright |
| 18:45 | < Rhamphoryncus> | python's REPL is usually called the interactive interpreter |
| 18:47 | < Rhamphoryncus> | if n < 0: return None? That's bizarre |
| 18:47 | <@Derakon> | What should I return? |
| 18:48 | < Rhamphoryncus> | It's a different type. Either it should be a well-defined numeric value or it should be an exception, ValueError |
| 18:48 | <@Derakon> | Hm. |
| 18:48 | <@Derakon> | You've a point, but I wanted to introduce the None type... |
| 18:48 | < Rhamphoryncus> | introducing None should involve "is" |
| 18:48 | <@Derakon> | Fair point. Hm. |
| 18:50 | < Rhamphoryncus> | I'd use test2.py, not test-2, as the latter can't be imported |
| 18:50 | <@Derakon> | Fair. |
| 18:50 | < Rhamphoryncus> | And unlike integer division (which also made me twitch) that's not something useful to know about :) |
| 18:51 | <@Derakon> | Heh. |
| 18:51 | <@ToxicFrog> | Hrm. |
| 18:51 | < Rhamphoryncus> | Any back button in the slideshow? |
| 18:51 | <@ToxicFrog> | Ok, I now have a working leaves-first approach, but it's so ugly and so fragile that I don't feel at all comfortable committing to this design. |
| 18:51 | <@Derakon> | No, sorry. |
| 18:51 | <@Derakon> | But it'll loop around, or you can hit refresh to start over. |
| 18:53 | < Rhamphoryncus> | Files and strings shadows id and type |
| 18:53 | <@Derakon> | I don't follow... |
| 18:54 | <@Derakon> | Oh, the id and type variables are reserved names? |
| 18:54 | < Rhamphoryncus> | id and type are both builtin functions. You use them for local variables |
| 18:54 | <@Derakon> | Will change, danke. |
| 18:55 | < Rhamphoryncus> | String formatting makes me wonder about str() and repr(). For that matter, .format() |
| 18:55 | <@Derakon> | I'm sticking to features available in 2.5. |
| 18:55 | <@Derakon> | So .format() isn't available. |
| 18:55 | < Rhamphoryncus> | alright |
| 18:56 | <@Derakon> | str() and repr() (and type conversion in general) would probably be useful though. |
| 18:56 | | Tarinaky [Tarinaky@Nightstar-f349ca6d.plus.com] has joined #code |
| 18:56 | < Rhamphoryncus> | yeah. Might need an extra slide though |
| 18:56 | <@Derakon> | Very likely. |
| 18:58 | < Rhamphoryncus> | For instance, list doesn't provide __str__ so str() falls back to __repr__, which is why the contents of the list also use repr() |
| 18:58 | <@Derakon> | That's pretty detailed, though. |
| 18:58 | <@Derakon> | Very likely I'll only mention str(), int() and float(). |
| 18:58 | < Rhamphoryncus> | and there's the x == eval(repr(x)) contention, but that's explicitly something people should NOT use personally |
| 18:58 | | * Rhamphoryncus nods |
| 18:58 | <@Derakon> | Similarly I'm not mentioning dicts. |
| 18:58 | <@Derakon> | And tuples are getting finessed. |
| 18:59 | | * Derakon changes id => num, type => flavor. |
| 19:00 | < Rhamphoryncus> | oi, took me a minute to parse the numpy array example. Probably okay if you're explaining it though |
| 19:00 | | * Derakon nods. |
| 19:01 | < Rhamphoryncus> | Blank lines in the output would help |
| 19:01 | <@Derakon> | I'm fitting each slide into 800x600. |
| 19:01 | <@Derakon> | I suppose I could split across slides... |
| 19:01 | < Rhamphoryncus> | eh |
| 19:03 | < Rhamphoryncus> | Array slices numpy style has an out of place ) |
| 19:04 | < Rhamphoryncus> | line 3 of output |
| 19:04 | <@Derakon> | Whoops, where'd that come from? Good eye. |
| 19:06 | < Rhamphoryncus> | I've no clue about mrc or priithon stuff |
| 19:07 | <@Derakon> | Yeah, that's specific to our labs. |
| 19:07 | <@TheWatcher> | I was amused at the note about documentation there, though >.> |
| 19:07 | < Rhamphoryncus> | The outline is kinda deceptive |
| 19:08 | <@Derakon> | Yeah, some of the items there only get one slide. |
| 19:10 | < Rhamphoryncus> | aww man, 3 days downloading a video and it uses a codec I don't have x_x |
| 19:14 | <@Derakon> | Okay, so: work "is" into None example, add type conversion. Already got the id/type, test-2.py, and parentheses fails. |
| 19:15 | <@Derakon> | Thanks, Rhamph! |
| 19:15 | < Rhamphoryncus> | np |
| 19:16 | <@Derakon> | And now, lunch. |
| 19:24 | < gnolam> | http://www.ymacs.org/demo/ |
| 19:44 | <@Derakon> | Oh, Gnolam! |
| 19:44 | <@Derakon> | I have an OpenGL problem I've been wrestling with. |
| 19:44 | <@Derakon> | I don't suppose you'd be interested in taking a look? |
| 19:44 | <@Derakon> | There's a description here with screenshots: http://www.gamedev.net/community/forums/topic.asp?topic_id=586323 |
| 19:44 | < gnolam> | Will look as soon as I get back from the grocery store. |
| 19:45 | | * Derakon nods. |
| 19:57 | <@Derakon> | There, new type-conversion slide: http://derakon.dyndns.org/~chriswei/temp2/slides/06b.png |
| 19:59 | | * Tarinaky blinks. |
| 20:09 | <@Derakon> | Yes? |
| 20:12 | < Tarinaky> | I don't get the slide. |
| 20:12 | < Tarinaky> | I -assume- this is because I'm missing the rest of the presentation plus the lecture notes :p |
| 20:13 | <@Derakon> | Heh. |
| 20:13 | <@Derakon> | Here's the entire slideshow: http://derakon.dyndns.org/~chriswei/temp2/slides/ |
| 20:13 | <@Derakon> | It's basically "Okay, you know the basics of how to program. Here's how to program in Python." |
| 20:14 | <@Derakon> | With a side-dish of "Here's some domain-specific stuff for the data you're manipulating." |
| 20:16 | < Tarinaky> | Ahhh. Python. |
| 20:16 | < Tarinaky> | That'll be why it all looks horribly wrong :p |
| 20:16 | <@Derakon> | Ahh, I'm guessing you're one of those people that thinks that semantically-void curly braces and semicolons are the sign of a proper language. :p |
| 20:17 | < celticminstrel> | Semantically-void? |
| 20:17 | <@ToxicFrog> | celticminstrel: "without semantics", ie, having no effect on the behaviour of the program and existing only as hints for the parser |
| 20:17 | <@Derakon> | As a matter of practice, curly braces are redundant when combined with sensible code styling. |
| 20:18 | < celticminstrel> | TF: Oh, when you put it that way, it makes sense. |
| 20:18 | <@jerith> | Syntactic sugar causes cancer of the semicolon. |
| 20:18 | < Tarinaky> | Derakon: I prefer not to assume sensible code styling :p |
| 20:18 | < celticminstrel> | I was going to say "how are the semantically void?" |
| 20:18 | <@Derakon> | Tarinaky: I prefer not to work with people who can't manage sensible code styling. :p |
| 20:18 | <@ToxicFrog> | That said, you need something to delimit blocks; python uses indentation, C uses curlybraces, Lua uses do..end |
| 20:18 | <@jerith> | Tarinaky: That's why it's nice to have it syntactically enforced. |
| 20:19 | < Tarinaky> | Anyway. Nothing to do with 'proper languages' It's just not the languages that I'm most comfortable-in/first learned. |
| 20:19 | <@Derakon> | Yeah, curly braces in C have meaning for the parser, but no meaning for the user, was my point. |
| 20:19 | < Tarinaky> | Anything that's not C++ doesn't look like C++ and therefor I have to think a little bit more about it. |
| 20:19 | <@Derakon> | Heh, fair enough. |
| 20:19 | < Tarinaky> | No more, no less. |
| 20:19 | < celticminstrel> | What about JavaScript? |
| 20:19 | <@jerith> | They have meaning for the user because they have meaning for the parser. |
| 20:19 | <@Derakon> | My first language was...BASIC on the C64, but I never made it further than very small programs on that. |
| 20:20 | <@Derakon> | After that, C, with some Angband hacking. |
| 20:20 | <@jerith> | I used to refuse to mark assignments submitted without readable indentation. |
| 20:20 | <@Derakon> | All of my courses marked you down for not commenting your code. |
| 20:20 | <@ToxicFrog> | Derakon: they do, in fact, have meaning for the user, even if they are -in some cases- redundant. |
| 20:21 | <@Derakon> | TF: permit me my hyperbole. |
| 20:21 | < celticminstrel> | Dashes are not parentheses! |
| 20:21 | <@ToxicFrog> | Alright. :P |
| 20:21 | <@ToxicFrog> | celticminstrel: and I wasn't using them as such. |
| 20:21 | <@Derakon> | He was using them to infix a parenthetical. |
| 20:22 | <@ToxicFrog> | I was actually using them as a cross between emdash and boldface, which is not uncommon usage on IRC. |
| 20:22 | < celticminstrel> | Well, he spaced them like parentheses, anyway. |
| 20:22 | <@jerith> | I never marked people down for not commenting, but I marked down liberally for lack of sensible identifier names. |
| 20:22 | < Tarinaky> | The indentation thing gets in the way of using Python as an interactive prompt though. |
| 20:22 | <@ToxicFrog> | That is to say, the enclosed text should be read as though separated out from the main text - like this - but also as emphasized, like this but not as dramatically. |
| 20:22 | < celticminstrel> | I see. |
| 20:22 | <@jerith> | Tarinaky: I manage in ipython quite well. |
| 20:23 | < Tarinaky> | jerith: As a full bash replacement? |
| 20:23 | <@Derakon> | Indendations in the REPL are a bit annoying solely because the REPL doesn't auto-indent for you. |
| 20:23 | <@jerith> | No. I use bash as a full bash replacement. |
| 20:23 | < celticminstrel> | XD |
| 20:23 | <@Derakon> | ...oh, you meant as a shell replacement. Yeah, no. |
| 20:23 | <@ToxicFrog> | Tarinaky: that's not what people usually mean by "an interactive prompt" |
| 20:23 | < celticminstrel> | Interactive prompt ? shell |
| 20:23 | <@ToxicFrog> | They mean a REPL, which python does quite nicely at. |
| 20:23 | <@Derakon> | If I want to do heavy programming on the commandline, I break out Perl. |
| 20:23 | < Tarinaky> | Yeah. My bad. |
| 20:23 | <@jerith> | I haven't yet found anything that does bash's job better than bash. |
| 20:24 | <@Derakon> | Jerith: not even emacs~? |
| 20:24 | <@jerith> | Even emacs. |
| 20:24 | < celticminstrel> | Emacs is too obtuse. |
| 20:24 | <@jerith> | Although I have been known to use bash from inside emacs. |
| 20:24 | < Tarinaky> | I meant for quickly hacking out something that only needs to work once and therefor doesn't need to be legible beyond being written. |
| 20:24 | <@jerith> | Tarinaky: I find ipython very handy for that. |
| 20:24 | <@Derakon> | Tarinaky: ls *png | perl -ne 'chomp; $i = `identify $_`; $i =~ /PNG (\d+)x(\d+)/; $x = ($1-800)/2; $y = ($2-600)/2; `convert -crop 800x600+$x+$y $_ cropped/$_`' |
| 20:24 | <@jerith> | I used it that way three times today, actually. |
| 20:25 | < Tarinaky> | Derakon: Exactly. |
| 20:25 | < celticminstrel> | That's pretty obtuse too. |
| 20:25 | < Tarinaky> | The strict white space gets in the way of things like that imo :/ |
| 20:25 | <@Derakon> | So don't use it for that purpose then. |
| 20:25 | < celticminstrel> | It kind of does, yes. |
| 20:26 | <@Derakon> | Not every language needs to be suited to every task. |
| 20:26 | < celticminstrel> | Though, you could probably use \n and \t to "fake" indentation. |
| 20:26 | <@jerith> | I occasionally write Python one-liners as part of a shell pipeline. |
| 20:26 | <@Derakon> | I'd hate to think what our cockpit program (my main responsibility here) would be like if written in Perl or Bash. |
| 20:42 | <@ToxicFrog> | So. I've decided that the only way to do this properly is to make everything that updates the gamestate synchronized (or only callable from synchronized methods). |
| 20:43 | <@ToxicFrog> | But now I'm running into situations where it would be really handy to be able to call the non-synchronized versions, as it avoids a message roundtrip and splitting something into client and server halves. |
| 20:44 | | Orthia [orthianz@Nightstar-0f7d3eab.xnet.co.nz] has joined #code |
| 20:56 | | Kaura|zzz is now known as Kaura |
| 21:42 | | * Derakon digs through his lab's massive collection of miscellaneous computer parts, is able to find three analog DVI to VGA adapters but no digital DVI to VGA adapters. |
| 21:43 | <@Derakon> | Which means that in order to run my presentation, I'll first have to go shopping. >.< |
| 21:49 | | Derakon [Derakon@Nightstar-1ffd02e6.ucsf.edu] has quit [[NS] Quit: Leaving] |
| 22:02 | | Stalker is now known as Someone_Else |
| 22:02 | | Someone_Else [Z@26ECB6.A4B64C.298B52.D80DA0] has quit [[NS] Quit: If the world didn't suck, we'd all fall off.] |
| 22:28 | < gnolam> | Derakon: sorry, there's just too much code in there. |
| 22:33 | | celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!] |
| 22:49 | | Stalker [Z@26ECB6.A4B64C.298B52.D80DA0] has joined #code |
| 22:52 | | McMartin_ [mcmartin@Nightstar-87302fbb.pltn13.sbcglobal.net] has joined #code |
| 22:52 | | shade_of_cpux is now known as cpux |
| 22:53 | | McMartin [mcmartin@Nightstar-0daf5087.pltn13.sbcglobal.net] has quit [Operation timed out] |
| 22:56 | | McMartin_ [mcmartin@Nightstar-87302fbb.pltn13.sbcglobal.net] has quit [Ping timeout: 121 seconds] |
| 23:03 | | Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has joined #code |
| 23:15 | | McMartin [mcmartin@Nightstar-92a860e2.pltn13.sbcglobal.net] has joined #code |
| 23:15 | | mode/#code [+o McMartin] by Reiver |
| 23:21 | < Rhamphoryncus> | Derakon is going to be mightily disappointed when he finds out DVI-D to VGA adaptors don't exist as a "miscellaneous computer part". I've got one open on google for $230 |
| 23:25 | < gnolam> | Also: what kind of laptop doesn't have VGA out? |
| 23:26 | | McMartin [mcmartin@Nightstar-92a860e2.pltn13.sbcglobal.net] has quit [Ping timeout: 121 seconds] |
| 23:29 | | McMartin [mcmartin@Nightstar-4268be89.pltn13.sbcglobal.net] has joined #code |
| 23:29 | | mode/#code [+o McMartin] by Reiver |
| 23:32 | | Anno[Laptop] [annodomini@Nightstar-09acd4e7.adsl.tpnet.pl] has quit [[NS] Quit: leaving] |
| 23:51 | | Derakon[AFK] is now known as Derakon |
| 23:51 | <@Derakon> | Rhamph: yeah, ended up having to settle for mini-DVI to VGA. |
| 23:52 | < Rhamphoryncus> | How much? |
| 23:52 | <@Derakon> | Which works fine for me but is problematic for my coworker, whose laptop has mini-DisplayPort instead. |
| 23:52 | <@Derakon> | $18. |
| 23:52 | <@Derakon> | What I don't get is why, if there's mini-DVI to DVI-D and mini-DVI to VGA, there can't be DVI-D to VGA. |
| 23:53 | < Rhamphoryncus> | oh that's quite decent |
| 23:53 | < Rhamphoryncus> | is mini-DVI DVI-I or DVI-D? |
| 23:54 | <@Derakon> | Um, no idea. |
| 23:54 | <@Derakon> | My laptop's a 2005 Macbook, if that helps any. |
| 23:55 | < Rhamphoryncus> | looks like DVI-D |
| 23:55 | | Tarinaky [Tarinaky@Nightstar-f349ca6d.plus.com] has quit [Connection closed] |
| 23:55 | < Rhamphoryncus> | I suspect it's an economy of scale thing. Apple created a decent market for mini-DVI to VGA |
| 23:56 | <@Derakon> | Thing is, there's tons of DVI-A to VGA adapters. |
| 23:56 | <@Derakon> | Why no DVI-D to VGA? |
| 23:56 | <@Derakon> | Rep at Best Buy said it was impossible because one was digital and the other analog, but then again, he's a rep at Best Buy. |
| 23:57 | < Rhamphoryncus> | DVI-A to V GA is just a rewiring. The signal is compatible, by design |
| 23:58 | < Rhamphoryncus> | but digital to VGA requires you to process and emit an entirely new VGA signal |
| 23:58 | <@ToxicFrog> | Derakon: because DVI-A and VGA are the same protocol with different pinouts, whereas DVI-D to VGA requires an actual device to translate the digital protocol into an analog one. |
| 23:59 | <@Derakon> | TF: again, then why does mini-DVI => VGA work just fine? |
| 23:59 | <@Derakon> | Along with mini-DVI => DVI-D. |
| 23:59 | < Rhamphoryncus> | mini-DVI -> DVI-D is just as trivial. Same protocol again |
| 23:59 | <@ToxicFrog> | Because the video card interrogates the monitor to see whether it's expecting a digital or analog signal, and switches accordingly. |
| --- Log closed Thu Nov 04 00:00:15 2010 |