--- Log opened Tue Jun 26 00:00:06 2007 |
00:05 | <@Chalcedon> | random question |
00:05 | <@Chalcedon> | does anyone know whether or not visual basic is OO? |
00:05 | <@Chalcedon> | (no I'm not trying to write anything in it) |
00:08 | <@ToxicFrog> | Yes, it is. |
00:24 | < MyCatVerbs> | Kinda, ish. VB .NET, definately yes. As far as I understand it, anything before that (including VBA as embedded in Office products) occasionally uses OO-ish classes and objects, but doesn't actually allow you to define your own. |
00:24 | < MyCatVerbs> | I may be horribly wrong on that, punch me if that's the case. |
00:35 | <@Chalcedon> | hrm |
00:35 | <@Chalcedon> | Forj has got stuck with trying to build an access database |
00:35 | <@Chalcedon> | so I guess that's VBA |
00:35 | <@Chalcedon> | my bad. |
00:35 | <@ToxicFrog> | Eewweeweweeewww |
00:36 | <@Chalcedon> | yeah |
00:37 | <@Chalcedon> | he's not doing to badly, but he's having some issues getting it to filter stuff |
00:39 | <@Chalcedon> | ah! he sorted it |
00:39 | | * Chalcedon cookies TF |
00:53 | <@ToxicFrog> | yaycookies |
00:53 | | * ToxicFrog works on a patch to Lunar |
00:56 | | Reiver [~reiver@Nightstar-12195.worldnet.co.nz] has joined #Code |
00:56 | | mode/#code [+o Reiver] by ChanServ |
02:00 | | Chalcedon is now known as ChalcyAFK |
02:07 | | Forjadon [~Forjadon@Nightstar-15747.ue.woosh.co.nz] has joined #code |
02:07 | | mode/#code [+o Forjadon] by ChanServ |
02:08 | | Forjadon is now known as ChalcyLaptop |
02:26 | | ChalcyLaptop [~Forjadon@Nightstar-15747.ue.woosh.co.nz] has quit [Ping Timeout] |
02:29 | | gnolam [lenin@Nightstar-13557.8.5.253.se.wasadata.net] has quit [Quit: Connection reset by Adam Weishaupt.] |
02:54 | | TuxRug [~lm_lappy@Nightstar-23958.clsp.qwest.net] has joined #code |
03:01 | <@ToxicFrog> | Vornicus! |
03:01 | <@ToxicFrog> | What does a shear pitch matrix look like? |
03:02 | <@Vornicus> | a shear matrix in general looks like |
03:02 | <@Vornicus> | 1 1 0 |
03:02 | <@Vornicus> | 0 1 0 |
03:02 | <@Vornicus> | 0 0 1 |
03:02 | <@ToxicFrog> | And a rotational matrix? |
03:02 | <@Vornicus> | Which shears in the x direction away from the y axis |
03:03 | <@Vornicus> | a rotation matrix has each row adding up to 1, but it's very complicated. |
03:03 | <@ToxicFrog> | Aah. |
03:03 | <@ToxicFrog> | Hmm. |
03:03 | <@ToxicFrog> | I'm rummaging through the A1 source looking for matrices now, and so far I've only found one, MaraEye_2_OGLEye |
03:03 | <@Vornicus> | I don't know what the matrices for Marathon look like at all. |
03:03 | <@ToxicFrog> | Which is: |
03:04 | <@ToxicFrog> | 0 0 -1 0 |
03:04 | <@ToxicFrog> | 1 0 0 0 |
03:04 | <@ToxicFrog> | 0 1 0 0 |
03:04 | <@ToxicFrog> | 0 0 0 1 |
03:04 | <@ToxicFrog> | And I think is just a simple transform between different coordinate systems. |
03:04 | <@Vornicus> | You'd be right. |
03:05 | <@Vornicus> | This wouldn't be a constant matrix, it might be futzing around in viewports and shit. |
03:05 | <@ToxicFrog> | Oh, right. |
03:05 | | * ToxicFrog eyeballs render_view(), which looks like a good starting point |
03:05 | <@Vornicus> | look for glu code in particular, it does a lot of simple camera shit. |
03:05 | <@ToxicFrog> | ... |
03:06 | <@Vornicus> | er |
03:06 | <@ToxicFrog> | if (view->overhead_map_active) /* if the automap is active, render it */ |
03:06 | <@ToxicFrog> | { |
03:06 | <@Vornicus> | glu.h is GL Utilities |
03:06 | <@ToxicFrog> | render_overhead_map(view); |
03:06 | <@ToxicFrog> | } |
03:06 | <@ToxicFrog> | else /* do something complicated and difficult to explain */ |
03:06 | <@Vornicus> | This is the actual comment? |
03:06 | <@ToxicFrog> | Yes. |
03:06 | <@Vornicus> | Bitchin. Can you give me a file pointer? |
03:07 | <@ToxicFrog> | Source_Files/RenderMain/render.cpp @420 |
03:09 | <@ToxicFrog> | The actual heavy lifting appears to be in RenderMain/RenderRasterize.cpp:64:RenderRasterizerClass::render_tree() |
03:11 | <@Vornicus> | All right, let's see this madness. |
03:12 | <@ToxicFrog> | This in turn appears to call out to render_node_floor_or_ceiling and render_node_side |
03:12 | <@ToxicFrog> | Of course, it's entirely possible that the shear effects are applied before the vistree is generated and handed off to the renderer |
03:12 | <@Vornicus> | That would be scary. |
03:12 | <@ToxicFrog> | By all accounts, this is very scary code. |
03:13 | <@ToxicFrog> | But I don't see it as that scary. |
03:13 | <@ToxicFrog> | It's a matter of whether it goes camera->shear->vistree->render, or camera->vistree->shear->render. |
03:14 | <@ToxicFrog> | ...er. |
03:14 | <@ToxicFrog> | Unless the vistree is the in-memory map representation, not an IR for the renderer. |
03:14 | <@ToxicFrog> | Which appears to be the case. |
03:15 | <@ToxicFrog> | In which case, yes, that would be very scary and it can't possible be doing that, or it would break physics. |
03:15 | <@Vornicus> | We need to find the viewport calls, let me locate the right command for that. |
03:17 | <@Vornicus> | hm, glViewport... |
03:18 | <@ToxicFrog> | ./RenderOther/screen_sdl.cpp: glViewport(0, 0, width, height); |
03:18 | <@ToxicFrog> | ./Misc/sdl_dialogs.cpp: glViewport(0, 0, ScreenRect.w, ScreenRect.h); |
03:18 | <@ToxicFrog> | ./RenderMain/OGL_Render.cpp: glViewport(RectBounds[0], RectBounds[1], RectBounds[2], RectBounds[3]); |
03:18 | <@ToxicFrog> | glOrtho(0, ViewWidth, ViewHeight, 0, 1, -1); // OpenGL-style z |
03:18 | <@ToxicFrog> | glGetDoublev(GL_PROJECTION_MATRIX,Screen_2_Clip); |
03:18 | <@ToxicFrog> | What do those do? |
03:19 | <@Vornicus> | glviewport actually appears to clip the viewing area. |
03:20 | <@ToxicFrog> | Yes. |
03:20 | <@Vornicus> | glOrtho... creates an orthographic projection. |
03:21 | <@Vornicus> | glFrustrum creates a viewing frustrum for perspective projection. |
03:22 | <@Vornicus> | It's not used though. |
03:23 | <@Vornicus> | unless it's doing all the crazygonuts matrix transformations by hand... |
03:24 | | Thaqui [~Thaqui@Nightstar-26828.jetstream.xtra.co.nz] has joined #code |
03:24 | | mode/#code [+o Thaqui] by ChanServ |
03:24 | <@ToxicFrog> | If you look at RenderRasterizerClass::render_node_side, it appears to get two "posts" from the map structure (or rather, a vertical_surface_data), and from those, generate a poly. |
03:28 | <@Vornicus> | Which is sensical. The question really is, does it generate a poly in a way that plays nice with OpenGL camera stuff, or does it short circuit the process? |
03:29 | <@Vornicus> | Specifically, does it just render the poly appropriately and let OpenGL's fancypants camera do the actual positioning, or does it do its fancypants camera work itself and just say "have a pile of corners"? |
03:30 | <@Vornicus> | Personally I don't see any reason for it to /not/ take advantage of OpenGL's camera implementation. |
03:30 | <@ToxicFrog> | Well, it has to support software rendering too. |
03:31 | <@ToxicFrog> | Well...it sets up poly.vector.[ijk] and poly.origin.[xyz], as well as a vertex array, does some additional setup for lighting and suchlike, then calls texture_vertical_polygon |
03:31 | <@Vornicus> | ...oh, god |
03:32 | <@ToxicFrog> | Which is...in both Rasterizer_SW and Rasterizer_OGL |
03:32 | <@ToxicFrog> | But it only seems to be defined for Rasterizer_SW! |
03:34 | <@ToxicFrog> | Oh wait, there it is |
03:34 | <@ToxicFrog> | OGL_RenderWall |
03:34 | <@Vornicus> | Well, really, the OpenGL one /should/ just say "okay, I have a camera, it's pointed this way, I want this segment of the view plane, plop it in this rectangle on the screen, and... go." |
03:35 | <@ToxicFrog> | ha ha ha ha ha ha |
03:35 | <@ToxicFrog> | I don't think it actually converts the map structure into OGL-compliant 3d structures. |
03:35 | <@ToxicFrog> | It does it poly-by-poly at render time instead. |
03:35 | <@Vornicus> | well, yeah, it has to do that bit. |
03:35 | <@Vornicus> | The thing is |
03:36 | <@ToxicFrog> | Aha |
03:36 | <@Vornicus> | It should just be able to go "okay, my wall has corners at these points in space, let the camera figure out where that is on the screen" |
03:36 | <@ToxicFrog> | Take a look at OGL_Render.cpp:1373 |
03:36 | <@ToxicFrog> | RenderAsRealWall() |
03:38 | <@Vornicus> | :( |
03:38 | <@ToxicFrog> | (and by "Aha", I mean "THERE'S FOAM IN MY BRAIN") |
03:38 | <@Vornicus> | OH GOD FOAM |
03:40 | <@Vornicus> | So, uh |
03:40 | <@Vornicus> | Yeah, i think we're looking at "Tear out the rendering engine". |
03:40 | <@Vornicus> | Because it's short-circuiting the pipeline. |
03:41 | <@ToxicFrog> | "Help me unbolt the engine!" |
03:41 | <@Vornicus> | heh |
03:42 | <@Vornicus> | Meanwhile, I've gotten myself lost in Nor'Korh 5, I've found three of the switches overlooking the big lava room, but can't find anywhere else to go. |
03:42 | < TuxRug> | is there a function in javascript that will grab the month as a number, the date, and the year either seperately or as one string formatted yyyymmdd, regardless of the local system's regional settings? |
03:43 | <@ToxicFrog> | I found myself totally disinterested in Nor'Korh shortly after reaching level 4, so I can't really help you there. |
03:43 | <@ToxicFrog> | While the levels have a lot of shiny things in them, I am very much Not Amused by "flip switch, now search entire level top to bottom for the change this resulted in" gameplay. |
03:46 | < TuxRug> | and would it generate a numerical value or a string? |
03:46 | <@ToxicFrog> | Subtly shifting tunnels and lighted power maps are all well and good, but the overall level design is terrible. |
03:46 | <@Vornicus> | There should be something that will give you it as either epoch or a date object. |
03:47 | <@ToxicFrog> | TuxRug: no idea. The ANSI C equivalent is strftime, though. |
03:48 | <@ToxicFrog> | Vornicus: what would it involve if we did it /wrong/? |
03:48 | <@ToxicFrog> | Ie, rather than unbolting the engine and using OGL properly, just spliced it in to the current rendering sequence? |
03:48 | < TuxRug> | basically, what I'm trying to do is use javascript to fetch an image off the web containing the current date in the URL. (so I can have a self-updating comic strip on my desktop) |
03:48 | <@ToxicFrog> | ...why are you using javascript for this? |
03:50 | < TuxRug> | because I want to make a windows active desktop object, from a file on my hard disk, fetching a favorite online comic. I don't want the object to pull the entire site, just the current strip. I figure that means either html-embedded javascript or html-embedded vbscript. |
03:51 | <@ToxicFrog> | Oh. Active desktop. |
03:51 | < TuxRug> | I don't have anywhere near the skill or resources to set up a dedicated CGI server on localhost or whatever :P |
03:51 | <@ToxicFrog> | Good luck. |
03:51 | < TuxRug> | heh |
03:52 | <@ToxicFrog> | Well, you could just set up a scheduled job that downloads to 'background.png' or whatever. |
03:52 | < TuxRug> | I may have to resort to vbscript, but gosh i hate that stuff with a passion |
03:52 | <@ToxicFrog> | Rather than actually making use of AD, which is a path to terror and madness. |
03:53 | < TuxRug> | that would not allow me to position the strip, meaning my icons will cover the strip, and if not, my CPU temperature gauge will |
03:53 | < TuxRug> | and also, that doesn't resolve the issue that the current strip |
03:53 | < TuxRug> | er the current strip's filename is based on the date |
03:54 | < TuxRug> | so today's for example would be kk20070625.gif |
03:54 | < TuxRug> | tomorrow's would be kk20070626.gif |
03:54 | < TuxRug> | etc. |
03:54 | < TuxRug> | there is no "current.gif" or similar |
03:54 | <@Vornicus> | So then use a scheduled job to update a file that AD looks up. |
03:54 | <@ToxicFrog> | The former, ok, hadn't thought of that. At minimum you'd need to do some kind of image manipulation. |
03:54 | <@Vornicus> | I don't know. |
03:54 | <@ToxicFrog> | However, as for the latter...err, what? |
03:55 | <@ToxicFrog> | The scheduled job looks at the current date, uses that to determine the filename on the server, downloads and saves it as the background image. |
03:55 | <@Vornicus> | Not as the original file name. |
03:56 | <@ToxicFrog> | Indeed, the entire point of doing it that way is so that you can use the language of choice, and thus have more powerful tools at hand for that kind of manipulation. |
03:56 | < TuxRug> | ToxicFrog: ok so how could I do that? Only thing I can think of in that case is use a .cmd script and how would I get command.exe to pull a file off the www? |
03:57 | < TuxRug> | yah i do want practice with javascript anyway. |
03:57 | <@ToxicFrog> | Anyways. Vorn makes a good point: you could write a job to download the file, and use a static AD page that handles all the positioning, etc. |
03:57 | <@ToxicFrog> | You definitely wouldn't do it with a batch file. |
03:57 | <@ToxicFrog> | I would probably do it in bash, but in practice you'd do it in whatever you wanted to. |
03:57 | <@ToxicFrog> | But I digress. |
03:58 | <@ToxicFrog> | If you're set on using JS, there are functions that do what you're after. |
03:58 | <@ToxicFrog> | I just have no idea what they're called. |
03:58 | <@ToxicFrog> | I recommend digging up a reference manual and raiding the index. |
03:58 | < TuxRug> | ToxicFrog: bash would be my choice, but i'm forced into windows being my pymary os |
03:58 | <@ToxicFrog> | ...and? |
03:58 | <@ToxicFrog> | Cygwin and MSYS both come with bash. |
04:00 | < TuxRug> | true, I do have cygwin, but in such a case it'd be a huge runaround script. |
04:01 | <@Vornicus> | Cygwin I think comes with Cron, too. |
04:02 | <@ToxicFrog> | Does it actually run cron as a service, though? |
04:02 | <@ToxicFrog> | And in what way would it be a huge runaround script? |
04:02 | <@Vornicus> | ...hell if I know. |
04:02 | <@Vornicus> | TF: I can answer that - you need a batch script to call the bash script if you're using Windows scheduled tasks. |
04:03 | <@ToxicFrog> | Right. Which is like a three-line springpad into the cygwin environment which can trivially be adapted from cygwin.bat |
04:03 | <@ToxicFrog> | And the download script itself is something like: |
04:03 | < TuxRug> | and what in bash would I use to pull the file? |
04:04 | <@ToxicFrog> | curl http://foo.bar.com/comics/$(date +%Y%m%d).png > /cygdrive/c/background.png |
04:04 | <@ToxicFrog> | That's the entire script |
04:04 | <@ToxicFrog> | You can also use wget in place of curl. |
04:04 | < TuxRug> | ah. |
04:04 | <@ToxicFrog> | In which case it turns into: |
04:05 | <@ToxicFrog> | wget -O /cygdrive/c/background.png http://foo.bar.com/comics/$(date +%Y%m%d).png |
04:05 | < TuxRug> | i will try that if I can't get this idea i got to work in javascript. |
04:05 | <@ToxicFrog> | As for the wrapper, I think it's just: |
04:05 | <@ToxicFrog> | CHDIR C:\cygwin\bin |
04:06 | <@ToxicFrog> | bash --login /bin/updatecomic.sh |
04:06 | <@ToxicFrog> | (you might not even need the --login, I'm not sure which parts of the rc files are responsible for setting up the vital parts of Cygwin) |
04:06 | <@ToxicFrog> | Hell, you can even do the entire thing from there: |
04:07 | <@ToxicFrog> | bash -c "wget -O /cygdrive/c/background.png http://foo.bar.com/comics/$(date +%Y%m%d).png" |
04:15 | < TuxRug> | ok after `var date = new Date()` will date.getDate() return a number as an integer or a string? |
04:20 | | GeekSoldier|Sleep is now known as GeekSoldier|Work |
05:03 | < TuxRug> | yay it works! |
05:26 | < TuxRug> | here's my code, I have it set on my desktop as an active desktop entry so i can move it. I tested it with different dates by changing my computer clock. |
05:26 | < TuxRug> | http://pastie.caboo.se/73505 |
05:26 | < TuxRug> | quite a bit of it is copy&paste from reference sites though :P |
05:27 | < TuxRug> | except I changed some variable names so i can quicky and easily expand on it, and tweaked it for my purposes |
05:28 | | takyoji [~Takyoji@Nightstar-25812.dhcp.roch.mn.charter.com] has joined #code |
05:28 | < TuxRug> | thanks |
05:28 | | TuxRug [~lm_lappy@Nightstar-23958.clsp.qwest.net] has left #code [] |
05:28 | < takyoji> | Say, does Perl come with Apache commonly? |
05:29 | < takyoji> | Because I'm trying to make something for a very crappy server that doesn't have RoR, PHP, ASP.NET, etc |
05:30 | < takyoji> | Because I might need to make a small script for recieving orders from my server on http://www.graystonecreations.com/ |
05:30 | < takyoji> | And all I have is WHOIS data of the server, no FTP access or anything |
05:33 | <@Vornicus> | Apache doesn't guarantee any languages. |
05:34 | < takyoji> | hmm? |
05:35 | < takyoji> | Doesn't come with any by default? |
05:35 | <@Vornicus> | right |
05:35 | < takyoji> | ahh |
05:35 | < takyoji> | crap |
06:21 | | ChAfk is now known as Chalain |
06:40 | | Chalain is now known as ChAfk |
07:12 | | Vornicus is now known as Vornicus-Latens |
07:37 | | You're now known as TheWatcher |
08:05 | | Forjadon [~Forjadon@Nightstar-15747.ue.woosh.co.nz] has joined #code |
08:05 | | mode/#code [+o Forjadon] by ChanServ |
08:05 | | Forjadon is now known as ChalcyLaptop |
08:15 | | ChalcyAFK [~Chalcedon@Nightstar-15747.ue.woosh.co.nz] has quit [Quit: Gone] |
08:16 | | Raif [~corvusign@Nightstar-5406.hsd1.ca.comcast.net] has quit [Connection reset by peer] |
08:17 | | Pi [~sysop@Nightstar-6875.hsd1.wa.comcast.net] has quit [Ping Timeout] |
08:19 | | Raif [~corvusign@Nightstar-7926.hsd1.wa.comcast.net] has joined #Code |
08:20 | | Raif_ [~corvusign@Nightstar-7926.hsd1.wa.comcast.net] has joined #Code |
08:22 | | Raif [~corvusign@Nightstar-7926.hsd1.wa.comcast.net] has quit [Ping Timeout] |
08:31 | | ChalcyLaptop [~Forjadon@Nightstar-15747.ue.woosh.co.nz] has quit [Ping Timeout] |
08:54 | | takyoji [~Takyoji@Nightstar-25812.dhcp.roch.mn.charter.com] has quit [Quit: Leaving] |
10:16 | | Pi [~sysop@Nightstar-6875.hsd1.wa.comcast.net] has joined #code |
10:16 | | mode/#code [+o Pi] by ChanServ |
10:45 | | gnolam [lenin@Nightstar-13557.8.5.253.se.wasadata.net] has joined #Code |
10:45 | | mode/#code [+o gnolam] by ChanServ |
11:00 | | GeekSoldier|Work is now known as GeekSoldier |
11:14 | | Reiver [~reiver@Nightstar-12195.worldnet.co.nz] has quit [Quit: Is it can be slep now pleeze?] |
11:25 | | Thaqui [~Thaqui@Nightstar-26828.jetstream.xtra.co.nz] has left #code [Leaving] |
12:09 | | GeekSoldier_ [~Rob@Nightstar-3947.pools.arcor-ip.net] has joined #code |
12:10 | | GeekSoldier [~Rob@Nightstar-3947.pools.arcor-ip.net] has quit [Ping Timeout] |
13:15 | | GeekSoldier_ is now known as GeekSoldier |
14:23 | | ToxicFrog is now known as ToxicFrog|W`rkn |
15:04 | | GeekSoldier_ [~Rob@Nightstar-3373.pools.arcor-ip.net] has joined #code |
15:06 | | GeekSoldier [~Rob@Nightstar-3947.pools.arcor-ip.net] has quit [Ping Timeout] |
15:12 | <@ToxicFrog|W`rkn> | Ok, Lunar Linux - which I have burbled about before - has a very slick interface for submitting module updates. |
15:12 | <@ToxicFrog|W`rkn> | # lunar set ZLOCAL_OVERRIDES on |
15:13 | <@ToxicFrog|W`rkn> | # lvu edit module |
15:13 | <@ToxicFrog|W`rkn> | <edit the module in /var/lib/lunar/moonbase/zlocal/module> |
15:13 | <@ToxicFrog|W`rkn> | # lvu submit |
15:13 | <@Vornicus-Latens> | lvu? |
15:13 | <@ToxicFrog|W`rkn> | Part of the package manager. |
15:13 | <@ToxicFrog|W`rkn> | lvu is used for information retrieval and manipulation. |
15:13 | <@ToxicFrog|W`rkn> | lin is the installer, lrm is the uninstaller. |
15:14 | <@ToxicFrog|W`rkn> | lunar is the configurator and curses frontend. |
15:14 | <@ToxicFrog|W`rkn> | Anyways, submit prompts you for some additional information, parses the edited module for the rest, then automatically packages it and submits the update to the lunar-dev mailing list. |
15:15 | | * ToxicFrog|W`rkn is seriously considering installing it on his laptop, now |
15:16 | <@ToxicFrog|W`rkn> | I've got a spare partition with nothing on it but XP, after all. |
15:16 | <@ToxicFrog|W`rkn> | ...oh wait! I even have an alternate root partition for exactly this kind of experimentation. |
15:25 | < jerith> | Has anyone here used netbeans, perchance? |
15:42 | | GeekSoldier_ is now known as GeekSoldier |
16:13 | <@ToxicFrog|W`rkn> | Oh, hot. |
16:13 | <@ToxicFrog|W`rkn> | Lunar includes fglrx and Opera in the moonbase. |
16:13 | <@ToxicFrog|W`rkn> | Although it's kind of sad that a source-based distro has better binary module support than some binary-based ones... |
16:15 | <@Vornicus-Latens> | "moonbase" being the repository? |
16:15 | <@ToxicFrog|W`rkn> | Yes. |
16:16 | <@ToxicFrog|W`rkn> | Sadly, one thing Lunar doesn't support is mixed-bitness systems. |
16:16 | <@Vornicus-Latens> | 32/64, etc? |
16:16 | | * TheWatcher ponders |
16:17 | <@ToxicFrog|W`rkn> | Yeah. |
16:17 | <@ToxicFrog|W`rkn> | Eg, running Opera or NX on a 64-bit system. |
16:17 | <@ToxicFrog|W`rkn> | It /does/ chroot nicely, but that's very workaround. |
16:18 | <@Vornicus-Latens> | Okay, I've only used chroot once, and it didn't seem to do anything near what I thought it did, considering allthe people who use it for, like, security |
16:19 | <@ToxicFrog|W`rkn> | What did you think it did? |
16:19 | <@Vornicus-Latens> | Well, I don't know |
16:19 | <@Vornicus-Latens> | But what it looks like it does is just change the location of / |
16:20 | <@ToxicFrog|W`rkn> | Yeah, pretty much. |
16:20 | <@Vornicus-Latens> | I thought it was a lot more, uh, comprehensive. |
16:20 | <@ToxicFrog|W`rkn> | ...how much more comprehensive does it need to be? |
16:20 | <@Vornicus-Latens> | I don't know. |
16:20 | <@Vornicus-Latens> | But, uh |
16:21 | <@Vornicus-Latens> | It seems like you'd never be able to even use a program that has a chroot attached to it. |
16:21 | <@ToxicFrog|W`rkn> | "attached to it"? |
16:21 | <@Vornicus-Latens> | uh |
16:21 | <@Vornicus-Latens> | let me try to explain a situation. |
16:21 | <@Vornicus-Latens> | Python, it keeps its libraries in /usr/lib/python<version> |
16:22 | <@ToxicFrog|W`rkn> | (bear in mind that stuff like ftp chroot doesn't actually use chroot(1); it just maps the / visible to the client to some non-/ directory on the server) |
16:22 | <@Vornicus-Latens> | but if I'm using chroot to keep it from thinking about all the sensitive bits of the system, it can't see its own libs. |
16:23 | <@ToxicFrog|W`rkn> | Yes. This is why the chroot needs to contain a working system. |
16:23 | <@ToxicFrog|W`rkn> | No kernel etc, but it needs all the /lib and /bin contents relevant to whatever you're running. |
16:23 | <@Vornicus-Latens> | aha. |
16:23 | <@ToxicFrog|W`rkn> | Typically, /dev and similar are bind-mounted into it. |
16:23 | <@ToxicFrog|W`rkn> | But generally, the way you build a chrooted system is you -construct a functioning install-, chroot into it, and then install the thing you want to run in it. |
16:24 | <@ToxicFrog|W`rkn> | In this case, what I'd probably do is create a /chroot32, install the package manager and all of its dependencies into it, and then chroot into it and go from there. |
16:24 | <@Vornicus-Latens> | aha |
16:24 | <@Vornicus-Latens> | okay, it makes much more sense now. |
16:27 | <@ToxicFrog|W`rkn> | Yey |
16:29 | | Vornicus-Latens is now known as Vornicus |
17:09 | | You're now known as TheWatcher[afk] |
18:22 | | You're now known as TheWatcher |
18:23 | | GeekSoldier_ [~Rob@Nightstar-3373.pools.arcor-ip.net] has joined #code |
18:25 | | GeekSoldier [~Rob@Nightstar-3373.pools.arcor-ip.net] has quit [Ping Timeout] |
18:46 | | GeekSoldier_ is now known as GeekSoldier |
20:23 | | GeekSoldier is now known as GeekSoldier|Sleep |
21:35 | | AnnoDomini [AnnoDomini@Nightstar-18228.koszalin.cvx.ppp.tpnet.pl] has joined #Code |
21:35 | | mode/#code [+o AnnoDomini] by ChanServ |
22:12 | | ChAfk is now known as Chalain |
22:29 | | Chalain is now known as ChAfk |
22:33 | | Chalcedon [~Chalcedon@Nightstar-15747.ue.woosh.co.nz] has joined #code |
22:33 | | mode/#code [+o Chalcedon] by ChanServ |
22:47 | | You're now known as TheWatcher[T-2] |
22:51 | | ToxicFrog|W`rkn [~ToxicFrog@Admin.Nightstar.Net] has quit [Operation timed out] |
22:52 | | ToxicFrog|W`rkn [~ToxicFrog@Admin.Nightstar.Net] has joined #code |
22:52 | | You're now known as TheWatcher[zZzZ] |
22:52 | | ToxicFrog|W`rkn is now known as ToxicFrog |
23:13 | | AnnoDomini [AnnoDomini@Nightstar-18228.koszalin.cvx.ppp.tpnet.pl] has quit [Quit: Astro Cat will play for you... the Symphony of Space.] |
23:24 | | Thaqui [~Thaqui@Nightstar-26828.jetstream.xtra.co.nz] has joined #code |
23:24 | | mode/#code [+o Thaqui] by ChanServ |
23:57 | | Chalcedon is now known as ChalcyAFK |
--- Log closed Wed Jun 27 00:00:12 2007 |