| --- Log opened Wed Nov 04 00:00:16 2009 |
| 00:03 | | * AbuDhabi waits patiently while the script executes. There are only for loops so it probably won't go on forever. |
| 00:04 | < AbuDhabi> | I think I'll have a bath meanwhile. |
| 00:04 | < MyCatVerbs> | "Only for-loops" is still Turing-complete, unless you aren't allowed to ever modify the loop index in the loop body. :) |
| 00:05 | < AbuDhabi> | That's why I said 'probably'. |
| 00:09 | < MyCatVerbs> | Well, do you modify loop indices or not? If you don't then there's no real "probably" about it. :) |
| 00:09 | < McMartin> | Augh, what the fuck, Apple |
| 00:09 | < McMartin> | Yes, let's totally have smart quotes in our default computer names |
| 00:09 | < AbuDhabi> | I do not. |
| 00:10 | < McMartin> | That won't confuse the fuck out of POSIX tools we're nominally compatible with, no sir |
| 00:10 | < Derakon[work]> | :( |
| 00:10 | | * McMartin ? Unicode |
| 00:11 | | * MyCatVerbs ? unicode. |
| 00:11 | < MyCatVerbs> | (That was u+fffd, REPLACEMENT CHARACTER, and yes I did it on purpose ^~) |
| 00:11 | < McMartin> | (That's the goal, yeah) |
| 00:12 | | * AbuDhabi finds errors, fixes. |
| 00:12 | < MyCatVerbs> | McMartin: what DE do you use, out of curiousity? |
| 00:12 | < McMartin> | "All of them"~ |
| 00:13 | < Derakon[work]> | McM is in the porting business. |
| 00:13 | < MyCatVerbs> | Ah, cool. |
| 00:13 | | * gnolam ? Unicode. |
| 00:13 | < AbuDhabi> | Three randomly generated masks gave me a correctness of 9%, 8% and 0.5%. |
| 00:13 | < McMartin> | More seriously, I've used at various points MSVC6, MSVS 2005, MSVS 2008, MinGW on the Windows side. |
| 00:13 | < McMartin> | On the Linux side, I stick with Emacs and whatever madness the project requires on the side |
| 00:13 | < McMartin> | On Mac depending on project either Emacs + tools or XCode outright. |
| 00:14 | < MyCatVerbs> | I was just going to point out that, with gnome+gtk, control-shift-u lets you type codepoints directly. |
| 00:14 | < McMartin> | Oh, I don't IRC from a DE. |
| 00:14 | < MyCatVerbs> | Oh. Your job must be fun. :) |
| 00:14 | | * McMartin is using PuTTY on Windows at the moment to IRC with. |
| 00:15 | < McMartin> | ? |
| 00:15 | < MyCatVerbs> | I was actually asking about your desktop environment rather than your development environment, but (to misquote Khellog Albran) that's okay, I wanted to know what your development chain was too. :) |
| 00:16 | < MyCatVerbs> | s/that's okay/That's all right/ |
| 00:16 | < McMartin> | Aha. In that case, XP, Win7, 10.5, 10.6, and Gnome+Gtk. |
| 00:16 | < MyCatVerbs> | And I'm supposed to be a priest in that context, too. :) |
| 00:16 | | * MyCatVerbs ? unicode. |
| 00:23 | < Namegduf> | I ? unicode, too. |
| 00:24 | < gnolam> | It's sad that my RLO override didn't actually get through, though. :( |
| 00:28 | | * Reiver ² unicode |
| 00:30 | | * AbuDhabi ? Unicode? |
| 00:34 | | Derakon[work] [Derakon@Nightstar-d44d635e.ucsf.edu] has quit [[NS] Quit: Leaving] |
| 00:38 | | * MyCatVerbs ? Unicode. |
| 00:42 | | AbuDhabi [farkoff@Nightstar-93dc0e2d.adsl.tpnet.pl] has quit [[NS] Quit: Death is always at your back, for all return to the earth from which they came, the fire of life is at the births to come, the tears of a mourning woman are at your left, and the wind blows forth from your right.] |
| 00:58 | < gnolam> | Rargh. |
| 00:58 | < gnolam> | I really wish Maxima had chosen a better name for itself. |
| 01:17 | < MyCatVerbs> | What's Maxima? |
| 01:17 | < MyCatVerbs> | I'd JFGI but it sounds like your principal complaint is the fact that they picked a difficult-to-google-for name. |
| 01:18 | < SmithKurosaki> | ;.; |
| 01:23 | | Derakon[AFK] is now known as Derakon |
| 01:35 | < gnolam> | MyCatVerbs: symbolic algebra program. |
| 01:35 | < gnolam> | And yes, that is my complaint. |
| 01:36 | < MyCatVerbs> | I do not understand why people don't google their program names before sticking to them. |
| 01:37 | < SmithKurosaki> | Beause they think they're original |
| 01:37 | < MyCatVerbs> | It's bad enough picking a nick that isn't google-able. Why the Hell would you stake the future of your company on a word that's hard for your customers to track down!? |
| 01:38 | < MyCatVerbs> | SmithKurosaki: but but but. |
| 01:38 | < gnolam> | In this case, it's especially bad since the name of the program is in the same domain as whatever you want to google for help for. :P |
| 01:38 | < SmithKurosaki> | I do see the point to your side, |
| 01:38 | < MyCatVerbs> | They -have- to check the damn trademark registry when they want to register the name of the product, don't they? |
| 01:38 | < MyCatVerbs> | How could they possibly fail to notice at that point!? |
| 01:38 | < SmithKurosaki> | idk |
| 01:39 | < MyCatVerbs> | gnolam: Oh wow, that's hilarious in how wrong and awful it is. |
| 01:39 | < gnolam> | Trying to find how to do a specific math operation with a program named after a piece of mathematical terminology is... tricky. :P |
| 01:40 | < SmithKurosaki> | ;. |
| 02:01 | | Attilla [The.Attilla@FBC920.398CA6.7725E7.C5C5C6] has quit [Ping timeout: 121 seconds] |
| 02:10 | | Attilla [The.Attilla@FBC920.81F6E9.8B40DC.05B20F] has joined #code |
| 02:24 | | * Rhamphoryncus ? Unicode, but ? UTF-16 (nevermind UCS-2 or CESU-8!) |
| 02:29 | | SmithKurosaki [Smith@Nightstar-683a7925.dsl.teksavvy.com] has quit [Ping timeout: 121 seconds] |
| 02:30 | | SmithKurosaki [Smith@Nightstar-683a7925.dsl.teksavvy.com] has joined #code |
| 02:33 | | Attilla [The.Attilla@FBC920.81F6E9.8B40DC.05B20F] has quit [Ping timeout: 121 seconds] |
| 02:34 | < gnolam> | As long as you don't ? anything. |
| 02:44 | < Rhamphoryncus> | bah, I'm full of ?? |
| 02:45 | | Attilla [The.Attilla@FBC920.81F6E9.4846F5.E10AE4] has joined #code |
| 03:16 | | Attilla [The.Attilla@FBC920.81F6E9.4846F5.E10AE4] has quit [Connection reset by peer] |
| 03:41 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has quit [[NS] Quit: Z?] |
| 04:16 | < ToxicFrog> | It's done! DONE! |
| 04:17 | < ToxicFrog> | And the next project uses Visual Studio too! |
| 04:17 | < ToxicFrog> | I am dead, and in hell! |
| 04:26 | | * SmithKurosaki hugs TF |
| 04:26 | < SmithKurosaki> | Your crack will heal you |
| 04:40 | < Vornicus> | Do you need multiple guns, floating around you, controlled by murder thoughts? |
| 04:41 | < ToxicFrog> | Quite possibly. |
| 04:41 | < SmithKurosaki> | That would be nice for me atm |
| 05:05 | | Syloqs-AFH [Syloq@NetworkAdministrator.Nightstar.Net] has quit [Connection reset by peer] |
| 07:01 | | AnnoDomini [farkoff@Nightstar-93dc0e2d.adsl.tpnet.pl] has joined #code |
| 07:10 | | Derakon is now known as Derakon[AFK] |
| 07:15 | | You're now known as TheWatcher |
| 09:03 | | Vornicus is now known as Vornicus-Latens |
| 09:06 | | crem_ [moo@Nightstar-8ca3eea7.adsl.mgts.by] has quit [[NS] Quit: Ep0xa 1.2v] |
| 09:15 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has joined #code |
| 09:58 | | Rhamphoryncus [rhamph@Nightstar-a62bd960.abhsia.telus.net] has quit [Client exited] |
| 10:27 | < Namegduf> | So, hmm, if I have an RPN calculator accepting input on the terminal with basic +-*/^% operations, no source, and I want to try to figure out any specific ways it implements things so I can duplicate it, what kind of gotchas should I look for? I've looked at division towards zero/negative infinity, modulo output sign, and how overflow in calculations is handled, as well as limits of the stack and how overflow/underflow in that is handled. |
| 10:28 | | * Namegduf is interested in basically how calculators can semi-sorta-reasonably-vary in implementation |
| 10:36 | < gnolam> | Test the width of its floating point representation? |
| 10:37 | < Namegduf> | It's purely integral, but nice idea. |
| 10:37 | < Namegduf> | (I did check and note the limits of that) |
| 10:43 | < simon`> | see how it handles bogus alphabetic input? |
| 10:46 | < Namegduf> | Okay, tried that, too. |
| 11:11 | | AnnoDomini [farkoff@Nightstar-93dc0e2d.adsl.tpnet.pl] has quit [Ping timeout: 121 seconds] |
| 11:14 | < Reiver> | BEDMAS implementation, specifically the B? |
| 11:17 | | AnnoDomini [farkoff@Nightstar-fa0045ff.adsl.tpnet.pl] has joined #code |
| 11:27 | | Attilla [The.Attilla@FBC920.81F6E9.4846F5.E10AE4] has joined #code |
| 11:42 | | Attilla [The.Attilla@FBC920.81F6E9.4846F5.E10AE4] has quit [[NS] Quit: ] |
| 12:07 | | Tarinaky [Tarinaky@Nightstar-f4275bb6.adsl.virginmedia.net] has quit [Ping timeout: 121 seconds] |
| 12:08 | | Attilla [The.Attilla@FBC920.0A905B.8E5905.BA3BAF] has joined #code |
| 12:17 | < Namegduf> | Reiver: Nice; I hadn't thought of that. Turns out it doesn't support them, but I hadn't checked. |
| 12:18 | < Reiver> | It tends to be one of the basics, to the point where people forget it needs checking. :) |
| 12:19 | < Namegduf> | Yeah. |
| 12:20 | < Reiver> | Does it have Pi? |
| 12:20 | < Reiver> | If so, what value does it use? (Hard to figure out, I know, but) |
| 12:21 | < Reiver> | (Unlikely, but) |
| 12:22 | | Tarinaky [Tarinaky@Nightstar-4ed4c883.adsl.virginmedia.net] has joined #code |
| 12:29 | < Namegduf> | Doesn't seem to have it. |
| 12:30 | < Namegduf> | Unless it responds to the unicode symbol or something, which would be remarkably clever. |
| 12:33 | < Namegduf> | (It's previously not had any fancy input methods displayed) |
| 12:34 | < Namegduf> | So far, it seems to mostly follow C semantics, including, I think, the implementation of it reading octals (stopping at the first non-octal character, but removing the whole set of digits from considered input) |
| 12:36 | < Namegduf> | Thanks for the ideas so far; I was mostly wondering if there were any traditionally "implementation defined" mechanics I was missing/not aware of. |
| 12:41 | < gnolam> | Yay for per-pixel (Phong) lighting. I can decrease my mesh resolution by half without it being noticeable at all. And it still looks good if I halve it again. |
| 12:42 | < Namegduf> | Neat. |
| 12:42 | < simon`> | Namegduf, how did you check the stack size? |
| 12:42 | < Namegduf> | simon`: I used it until it ran out. |
| 12:44 | < gnolam> | And since everything's procedural, the rendering resolution doesn't affect the physical model at all. |
| 12:47 | < gnolam> | Next up: adaptive mesh. And then I'll be able to get equal or better visual resolution for half the vertices. :) |
| 13:27 | | AnnoDomini [farkoff@Nightstar-fa0045ff.adsl.tpnet.pl] has quit [Ping timeout: 121 seconds] |
| 13:45 | | You're now known as TheWatcher[afk] |
| 15:33 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has quit [[NS] Quit: Reboot] |
| 15:36 | | gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has joined #code |
| 15:37 | | AnnoDomini [farkoff@Nightstar-fa0045ff.adsl.tpnet.pl] has joined #code |
| 16:05 | | Syloqs_AFH [Syloq@NetworkAdministrator.Nightstar.Net] has joined #code |
| 16:05 | | Reivles [reaverta@ServerAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds] |
| 16:06 | | Syloqs_AFH is now known as Syloqs-AFH |
| 16:15 | < AnnoDomini> | Does anyone know how to move rows and columns internally in a matrix, in Mathcad 14? |
| 16:16 | < AnnoDomini> | I mean, I have coordinates for a value, and want that value to be moved to position (0,0), without altering what the matrix represents. |
| 16:58 | | You're now known as TheWatcher |
| 17:03 | | * AnnoDomini wonders how you move a row, because columns are easy enough. |
| 17:36 | < simon`> | does grub expect that /boot is in hd0,0? |
| 17:37 | < SmithKurosaki> | If that is what it says |
| 17:37 | < SmithKurosaki> | What's going on simon` ? |
| 17:37 | < simon`> | SmithKurosaki, a horrible experience with Kubuntu is what's happening. |
| 17:37 | < SmithKurosaki> | :( |
| 17:38 | < SmithKurosaki> | I need more details |
| 17:38 | < simon`> | well |
| 17:39 | < simon`> | it started out as this: 80G windows, 319G kubuntu, 1G swap |
| 17:39 | < SmithKurosaki> | kk |
| 17:39 | < SmithKurosaki> | All on one hdd? |
| 17:40 | < SmithKurosaki> | Can you put /boot/grub/grub.conf or /boot/grub/menu.lst into a pastebin and post it |
| 17:40 | < SmithKurosaki> | You will most likely have the menu.lst |
| 17:40 | | * simon` needs to rant a little to get where he's currently at ;) |
| 17:41 | < SmithKurosaki> | kk |
| 17:41 | < simon`> | the kubuntu broke during an upgrade, so I removed its partitions and recreated the partitions and mkfs.ext4'ed the main one |
| 17:41 | < simon`> | then I had a hell of a time getting kubuntu on a usb key. eventually I worked it out. |
| 17:42 | < SmithKurosaki> | ok |
| 17:42 | < simon`> | during that time I fought with my department's proxy and its wifi load balancing several times. |
| 17:42 | < simon`> | (for some reason, the proxy wouldn't let me download stuff beyond 350MB.) |
| 17:42 | < simon`> | anyways. |
| 17:43 | < simon`> | so the kubuntu installer is completely whack. the radio buttons were broken and floated into each other, and I could hardly read what it said |
| 17:43 | < SmithKurosaki> | ok |
| 17:44 | < simon`> | after installing, the installer had managed to repartition the harddisk into the following: 80G windows, 120G nothing, 120G kubuntu, 5G swap, 1G nothing |
| 17:44 | < SmithKurosaki> | Wow |
| 17:44 | < simon`> | so what it essentially did was that it took my attempt at a preconfigured partition table, and broke it into two. |
| 17:44 | < simon`> | it installed fine, booted fine |
| 17:44 | < simon`> | but half of the space was useless |
| 17:45 | < SmithKurosaki> | right |
| 17:45 | < simon`> | I figured I'd move / to sda2 and simply live with the other 120G partition as /home, since that's not a bad strategy at all. |
| 17:45 | < simon`> | (and in the meantime, make sure it only has *one* 1G swap partition) |
| 17:45 | < simon`> | so that apparently worked out |
| 17:46 | < SmithKurosaki> | (So far sounds like your grub config is fucked |
| 17:46 | < simon`> | except that sda3 (the new /home) would only identify as a 1.1G disk. I suspect my cfdisk can't handle large sizes, but otherwise I have no idea why it did that. |
| 17:46 | < simon`> | it is, but I'm getting to that ;P |
| 17:46 | < simon`> | grub has nothing to do with the partition table itself |
| 17:46 | < simon`> | but the partition table and/or the filesystems I try to generate on the partitions have some issue calculating on sda3, which was a pain I couldn't resolve |
| 17:47 | < simon`> | so I decided, since I was running out of time for the evening, to just ignore this new /home and make it work later |
| 17:47 | < simon`> | I recreate a /home in sda2 (first 120G partition), update sda2/etc/fstab to function with this new layout, reboot |
| 17:48 | < simon`> | and that's when it occurs to me that the grub that kubuntu automatically installed thinks /boot is in the wrong place. |
| 17:48 | < simon`> | and the grub tools are an absolute, undocumented hell, I think |
| 17:49 | < SmithKurosaki> | I am not completely sure about that |
| 17:49 | < simon`> | well, I would run grub-install /dev/sda |
| 17:49 | < simon`> | and it segfaults. |
| 17:49 | < SmithKurosaki> | Shit |
| 17:50 | < simon`> | (I am in a usb live disk doing this.) |
| 17:50 | < SmithKurosaki> | Ahh |
| 17:50 | < simon`> | is that a problem? |
| 17:50 | < SmithKurosaki> | Normally I just edit the config files manually |
| 17:50 | < simon`> | well |
| 17:50 | < SmithKurosaki> | Have you done that before/ |
| 17:50 | < simon`> | currently the grub that resides in MBR doesn't know where /boot is supposed to be. |
| 17:50 | < SmithKurosaki> | Hmm |
| 17:51 | < simon`> | I need to install grub anew and at the same time tell it that /boot is found in /dev/sda2. |
| 17:51 | < SmithKurosaki> | I don't have an MBR partition on my computer, I just edit the one in the OS |
| 17:51 | < simon`> | at that point, /boot/grub/menu.lst can be reached and things will start making sence. |
| 17:51 | < simon`> | sense* |
| 17:51 | < simon`> | you don't have a bootloader in the MBR of your main harddisk? |
| 17:52 | < SmithKurosaki> | TF may be able to help, outside of editing the OS's grub config, I am not well versed enough |
| 17:52 | < simon`> | the problem is I do this every five years |
| 17:52 | < simon`> | so either I've forgotten how to do it, the standard bootloader has changed, or the favourite bootloader changes its file layout completely. |
| 17:53 | < SmithKurosaki> | No, I don't. For a long time I had different OSs on different drives, so one of the OSs was made primary and I used the grub menu from that |
| 17:53 | < SmithKurosaki> | Who knows, maybe you can ditch the MBR and do it directly |
| 17:54 | < SmithKurosaki> | Whenever the primary OS location changed, I would have to spend about an hour trying to figure out where the hdds were in relation to that drive |
| 17:58 | < simon`> | SmithKurosaki, if you ditch the MBR, you need to boot on another device in order to decide which system to start. |
| 17:58 | < SmithKurosaki> | o.09 |
| 17:59 | < simon`> | SmithKurosaki, what one can do is install a minimal bootloader that just boots a system at some partition without asking. |
| 17:59 | < SmithKurosaki> | Before I had U8.04, fedora 9, xp pro. xp and fedora were on one disk and 8.04 was on the other |
| 18:00 | < SmithKurosaki> | I wa using fedora's grub for the longest time until it went pop. Then I started using 8.04's after editing the fedora one to get the proper drive layout |
| 18:00 | < SmithKurosaki> | I just had to swap which disk was first in the boot sequence in bios |
| 18:00 | | Rhamphoryncus [rhamph@Nightstar-a62bd960.abhsia.telus.net] has joined #code |
| 18:02 | | SmithKurosaki [Smith@Nightstar-683a7925.dsl.teksavvy.com] has left #code ["Leaving"] |
| 18:02 | | SmithKurosaki [Smith@Nightstar-683a7925.dsl.teksavvy.com] has joined #code |
| 18:03 | < SmithKurosaki> | You can't do the same thing as me though because you a) have an MBR and b) have only one drive |
| 18:12 | | Derakon[work] [Derakon@Nightstar-d44d635e.ucsf.edu] has joined #code |
| 18:22 | < Derakon[work]> | Well, Visual Studio 2008 is installed on the cockpit computer. I just have to figure out how to get the program to build. |
| 18:22 | < Derakon[work]> | Maybe then we can get some debugging symbols. |
| 18:26 | | * TheWatcher hands Dera a chicken, and obsidian knife, and a portable sacrificial altar |
| 18:27 | | * Derakon[work] clothes the chicken in evil robes and gives it the knife and an egg. |
| 18:34 | < simon`> | are there any non-sucky terminal mail clients? gnus depends on xemacs and mutt depends on postfix. |
| 18:37 | < Derakon[work]> | Man, I haven't mucked with terminal mail since college, when we all used pine. |
| 18:37 | < Derakon[work]> | But that probably qualifies as sucky. |
| 18:37 | < simon`> | yes, I think it does. |
| 18:47 | | * TheWatcher actually quite likes pine |
| 18:47 | | * TheWatcher is also probably clinically insane, so this doesn't say much |
| 18:51 | < simon`> | I just wonder why terminal mail and news clients are so terribly complicated and oudated. |
| 18:51 | < simon`> | I guess it relates to the complexity of supporting old protocols. |
| 19:00 | | Syloqs-AFH [Syloq@NetworkAdministrator.Nightstar.Net] has quit [Connection reset by peer] |
| 19:03 | <@TheWatcher> | Not really |
| 19:04 | <@TheWatcher> | It's probably down to the fact that Most People will use graphical or web clients these days, and the people using the terminal clients Like How They Work, Thank You Very Much. |
| 19:04 | < Derakon[work]> | Why does the code have xyzPos() and posXYZ()? |
| 19:04 | | * Derakon[work] sighs. |
| 19:05 | < Derakon[work]> | The former is simply the latter with an offset added, I note. |
| 19:05 | < Derakon[work]> | They're both getters. |
| 19:05 | | Syloqs_AFH [Syloq@NetworkAdministrator.Nightstar.Net] has joined #code |
| 19:06 | | * TheWatcher patpats Dera |
| 19:06 | | Syloqs_AFH is now known as Syloqs-AFH |
| 20:27 | | crem [moo@Nightstar-8ca3eea7.adsl.mgts.by] has joined #code |
| 20:38 | | crem [moo@Nightstar-8ca3eea7.adsl.mgts.by] has quit [Ping timeout: 121 seconds] |
| 20:42 | | crem [moo@Nightstar-8ca3eea7.adsl.mgts.by] has joined #code |
| 20:54 | < Derakon[work]> | Mental note: doing "% > /tmp/t.txt" will not accomplish anything much. |
| 20:54 | < Derakon[work]> | It's generally helpful to provide a command that supplies output before trying to redirect said output. |
| 20:56 | < Derakon[work]> | try: pass # X.nanoTalker.disconnect() # why does this crash on windows ?? 20040722 |
| 20:56 | < Derakon[work]> | except: |
| 20:56 | < Derakon[work]> | pass |
| 20:56 | < Derakon[work]> | Hm. Whitespace went screwy there. |
| 20:56 | < Derakon[work]> | But it's basically "try: pass; except: pass" |
| 20:59 | | * Derakon[work] writes a Perl script to track down functions that are not being called, gets 143 results. |
| 20:59 | < Derakon[work]> | Out of 771 total. |
| 20:59 | < Derakon[work]> | That's...pretty terrible. |
| 21:03 | <@TheWatcher> | ... |
| 21:03 | <@TheWatcher> | That's either terrible, or someone who learnt all their programming from an Object Oriented Programming in Java book~ |
| 21:05 | | GeekSoldier [Rob@Nightstar-e86e3e0d.ip.cablemo.net] has quit [[NS] Quit: Praise "BOB"!] |
| 21:06 | < gnolam> | That guy never heard of YAGNI. |
| 21:06 | < Derakon[work]> | Gnolam has it. |
| 21:07 | < Derakon[work]> | Some of this stuff might actually be useful (assuming it works), but if it's not being used, then out it goes. |
| 21:07 | < Derakon[work]> | There's also commented-out code all over the place. |
| 21:07 | < Derakon[work]> | Presumably because he wasn't certain if someday he might need it again. |
| 21:07 | <@TheWatcher> | >.< |
| 21:08 | < Derakon[work]> | I, on the other hand, have the magic of Source Control at my beck and call. |
| 21:09 | < Derakon[work]> | Which is why, despite having added many new files for new functionality, the codebase has actually shrunk since I got my hands on it. |
| 21:09 | < Derakon[work]> | I've eliminated about 4% of the old codebase so far, I estimate. |
| 21:11 | < AnnoDomini> | I have eliminated all the juice. |
| 21:20 | < Rhamphoryncus> | except: pass is almost never a good idea x_x |
| 21:20 | < Derakon[work]> | Agreed. :\ |
| 21:21 | < Derakon[work]> | It's frequent in this codebase for "I know this may not be initialized yet; if it is initialized, then I want to do something with it, otherwise I don't care." |
| 21:21 | < Rhamphoryncus> | Python itself does it in a couple places, but they've been shrinking, and the remainder are probably bad ideas too |
| 21:21 | < Rhamphoryncus> | There's usually a right way to do it involving hasattr or the like |
| 21:21 | < Derakon[work]> | Hell, there's tons of places where it does "try: varname; except: create varname and attach it to the module" |
| 21:21 | < McMartin> | AIEE THE FUNCTION IN YELLOW |
| 21:22 | < Rhamphoryncus> | Or give a default of None |
| 21:22 | < Rhamphoryncus> | That's only mildly bad |
| 21:23 | < SmithKurosaki> | McM: Really now? |
| 21:23 | < Derakon[work]> | SK: hasattr is generally a bad idea. |
| 21:24 | < Derakon[work]> | Since it signifies that your program is groping at metadata to figure out what to do instead of having it be more explicit. |
| 21:24 | < gnolam> | And on the topic of The Function in Yellow, "hasattr" is dangerously close to "Hastur". |
| 21:24 | < SmithKurosaki> | o.0 |
| 21:24 | < Derakon[work]> | Heh. |
| 21:24 | < McMartin> | gnolam: That being why I shouted, yes~ |
| 21:24 | < Rhamphoryncus> | Derakon[work]: right, but hasattr is the best way to do a bad idea :) |
| 21:25 | < McMartin> | Just remember, if you invoke it three times in a block, you go mad forever |
| 21:25 | < Rhamphoryncus> | Or at least until you refactor |
| 21:25 | | * TheWatcher eyes |
| 21:26 | < Derakon[work]> | Rhamphoryncus: yeah, but it's still a bad sign. |
| 21:26 | < Derakon[work]> | It's like seeing duct tape instead of a gaping hole in the wall. |
| 21:26 | < Derakon[work]> | Technically the tape is better than the alternative... |
| 21:26 | | * Derakon[work] takes a file from exactly 1000 lines to 820. |
| 21:27 | <@TheWatcher> | Great, now I'm somehow going to have to work into one of my projects a function whose comments note it is the 'Function Not To Be Described, which ears a yellow fold marker over its definition" |
| 21:27 | <@TheWatcher> | *wears |
| 21:27 | < Rhamphoryncus> | Derakon[work]: naw. It's not that bad until you realize hasattr is one of the few remaining places that does swallow all exceptions ;) |
| 21:28 | < Derakon[work]> | :( |
| 21:28 | < AnnoDomini> | Anyone here grok sets of linear equations? |
| 21:28 | | * Derakon[work] eyes this module. Functions in it: Must_Restart, sleep, msleep, fixMRC_missingLastSection. |
| 21:28 | < AnnoDomini> | I need to know what, exactly, gets changed when I move a column or row in the coefficients matrix. |
| 21:30 | < Derakon[work]> | Ah hah. The "G" in "sebG.py" stands for "General". *sigh* |
| 21:32 | < McMartin> | :( |
| 21:33 | < AnnoDomini> | beer_pump.c:335: robust error in 'traffic_cone()' - 'traffic_light' is not a budgerigar |
| 21:33 | < McMartin> | What the shit |
| 21:33 | <@TheWatcher> | ... |
| 21:34 | < McMartin> | Re: moving a matrix around; transposing columns means that you're transposing variables |
| 21:34 | < McMartin> | Transposing rows means you're reordering the equations in the statement of the problem, which is more or less a no-op |
| 21:36 | < AnnoDomini> | Thanks! |
| 21:36 | < Derakon[work]> | I find it generally helps to think "What would this look like if I attached variable names to everything?" |
| 21:39 | < AnnoDomini> | Yes, I understand more or less how it works. Just in my current state of fatigue I am unable to brain. |
| 21:40 | | * Derakon[work] carefully extracts the dumb from AnnoDomini and disposes of it in the biohazards bin. |
| 21:41 | < Derakon[work]> | Hm. So far, on the files that have unused functions in them, I'm removing about 20% of the total file. |
| 21:46 | < Derakon[work]> | Oooh, took that 951-line file down to 361 lines. |
| 21:52 | < AnnoDomini> | Thanks. |
| 21:52 | < AnnoDomini> | Assignment completed. |
| 21:52 | < Derakon[work]> | Congrats! |
| 21:53 | < AnnoDomini> | Curiously, with picking the largest element to be in (0,0) leads to poorer results than with normal Gauss-Jordan solving. |
| 21:53 | < AnnoDomini> | This time, I mean. |
| 21:53 | | * AnnoDomini presses Ctrl+F9 a few times. |
| 21:54 | < AnnoDomini> | It's giving me better results, most of the time, by a few orders of magnitude. |
| 21:57 | < AnnoDomini> | Stupid pendrive. What do you mean, it cannot be stopped? |
| 22:05 | < Derakon[work]> | The plan set in motion so long ago cannot be stopped by any mortal agency. You, and all your kind, are doomed. |
| 22:06 | < AnnoDomini> | Unless I just, you know, pull it out of the socket. |
| 22:09 | < Derakon[work]> | And then smash it at Hellforge? |
| 22:10 | < AnnoDomini> | That seems excessive. |
| 22:15 | | Vornicus-Latens is now known as Vornicus |
| 22:26 | | AnnoDomini [farkoff@Nightstar-fa0045ff.adsl.tpnet.pl] has quit [[NS] Quit: You can't eat someone's pet hamburger!] |
| 22:32 | < McMartin> | AnnoDomini: Gaussian elimination and the Graham-Schmidt process are both extremely sensitive to floating point instability. |
| 23:00 | | * gnolam blarghs. |
| 23:04 | < McMartin> | Oops. |
| 23:06 | | You're now known as TheWatcher[T-2] |
| 23:10 | | mode/#code [+o Attilla] by ChanServ |
| 23:10 | | You're now known as TheWatcher[zZzZ] |
| 23:13 | < gnolam> | I can't seem to get my normals in order. |
| 23:16 | | * Derakon[work] tries to figure out an automated way to find commented-out code. |
| 23:25 | < McMartin> | This is why I prefer #if 0 ing it out. |
| 23:27 | < Derakon[work]> | To my knowledge Python has on such construct. |
| 23:27 | < Derakon[work]> | You either # each line or do a large ''' or """ block comment. |
| 23:32 | < Derakon[work]> | ...it occurs to me that I could just search for the '#' character. |
| 23:32 | < Derakon[work]> | The vast majority of its uses will be for commented-out code. |
| 23:35 | < McMartin> | ^# is probably safer still. |
| 23:35 | < McMartin> | In-code comments are typically indented. |
| 23:36 | < Derakon[work]> | Yeah, so are some of the commented-out blocks. But point taken. |
| 23:36 | < Derakon[work]> | % find . -name "*py" | xargs grep "^#" | wc -l |
| 23:36 | < Derakon[work]> | 479 |
| 23:37 | < Derakon[work]> | Ooh, the pattern "#.*(" should do a pretty good job of catching stuff. |
| --- Log closed Thu Nov 05 00:00:31 2009 |