X-Git-Url: https://git.ucc.asn.au/?a=blobdiff_plain;f=irc%2Flog;h=ae50d0ac7ccf19041d0403bac941ac3cd210601d;hb=dd5ae3e50b2d77cf0cc1a2805c91d518270c8537;hp=d799f2dec9152253ff08a1048f6923baa6998ed0;hpb=19c48bb0707a0a92aa713e125cbd070295621383;p=matches%2FMCTX3420.git diff --git a/irc/log b/irc/log index d799f2d..ae50d0a 100644 --- a/irc/log +++ b/irc/log @@ -5530,3 +5530,252 @@ 15:46 < jtanx> okay 15:50 < jtanx> finally... back to actually coding the control page 22:59 -!- jtanx [~asfa@220-253-133-241.dyn.iinet.net.au] has quit ["ChatZilla 0.9.90.1 [Firefox 24.0/20130910160258]"] +--- Day changed Sun Oct 20 2013 +15:56 -!- Callum [~Callum@106-68-161-33.dyn.iinet.net.au] has joined #mctxuwa_softdev +15:59 < Callum> hey sam, got a link to the dropbox? +16:06 < sam_moore> https://www.dropbox.com/sh/km90dmbdrgin3mg/OimGp0qopv +16:06 < Callum> thanks +16:07 < sam_moore> I think I might be able to hack together a user management system based on UserCake +16:08 < sam_moore> Someone else will have to make it look pretty though +17:25 < Callum> opencv is fucking retarded. +17:25 < Callum> C++ has a bunch of different functions for each individual type of blur +17:25 < Callum> C has one function but you pass it an ID for which type of blur you want to use.. +17:30 < sam_moore> OpenCV used to be in C, but they moved it to C++ +17:30 < sam_moore> But it still supports the C API +17:30 < sam_moore> Through some wierd magic +17:30 < sam_moore> At least, if you run our program in valgrind, it reports a lot of stuff happening with things like "std::string" and "cv::Mat"; ie: C++ namespaces +17:32 < Callum> yea. its a bit of a bitch to code in C...all the documentation is for c++ pretty much +17:32 < Callum> like right now im trying to find out how to simply set all elements to 0.. where in c++ its just Mat::Scalar(0) +17:32 < sam_moore> If you really want you can probably compile in C++, but I'm not sure if it will cause issues elsewhere +17:33 < Callum> il just keep going like this. just a couple of things atm i need to figure out. hopefully thats it +17:33 < sam_moore> Cool +17:34 < sam_moore> The server now supports 3 entirely different authentication methods :S +17:34 < Callum> haha. a little excessive. +17:34 < Callum> hopefully they'll be happy with atleast one of them.. +17:34 < sam_moore> If we modify UserCake we can run our own user management system +17:35 < sam_moore> I need to change it so that instead of users registering themselves it's the admin that adds users manually though +17:35 < sam_moore> Also we'd have to get email working +17:57 < Callum> uuh. so using one of his test images for the microscope, running it through Canny gives a bunch of squiggly lines all over the place +17:57 < Callum> its funny, the part which is actually the edge...has the most black +17:57 < Callum> il try some of his other images and see what it does +20:22 -!- Callum [~Callum@106-68-161-33.dyn.iinet.net.au] has quit [EOF From client] +20:50 -!- MctxBot_ [~twang@106-68-93-93.dyn.iinet.net.au] has joined #mctxuwa_softdev +21:05 -!- MctxBot [~twang@220-253-133-241.dyn.iinet.net.au] has quit [Ping timeout] +21:06 -!- Irssi: #mctxuwa_softdev: Total of 2 nicks [0 ops, 0 halfops, 0 voices, 2 normal] +21:08 -!- jtanx [~asfa@106-68-93-93.dyn.iinet.net.au] has joined #mctxuwa_softdev +21:10 < jtanx> working on the report the day before... +21:10 < jtanx> what is this madness +22:23 -!- jtanx [~asfa@106-68-93-93.dyn.iinet.net.au] has quit ["ChatZilla 0.9.90.1 [Firefox 24.0/20130910160258]"] +--- Day changed Mon Oct 21 2013 +07:40 -!- jtanx [~asfa@106-68-93-93.dyn.iinet.net.au] has joined #mctxuwa_softdev +08:04 -!- jtanx [~asfa@106-68-93-93.dyn.iinet.net.au] has quit ["ChatZilla 0.9.90.1 [Firefox 24.0/20130910160258]"] +09:15 -!- jtanx [~asfa@130.95.131.138] has joined #mctxuwa_softdev +11:00 -!- jtanx [~asfa@130.95.131.138] has quit [Ping timeout] +11:03 -!- jtanx_ [~asfa@130.95.54.13] has joined #mctxuwa_softdev +11:03 -!- jtanx_ is now known as jtanx +11:22 < jtanx> mctxserv[10834]: FATAL: Data_Save (data.c:80) - Error seeking to end of DataFile test/sensor_1 - File too large +11:23 < jtanx> jeremy@pickle:~/git/MCTX3420/server/test$ du -sh * +11:23 < jtanx> 0 actuator_0 +11:23 < jtanx> 2.0G sensor_0 +11:23 < jtanx> 2.1G sensor_1 +11:23 < jtanx> after running overnight +11:23 < jtanx> hahaha +11:36 < sam_moore> The 32G SD card might have been a good thing to buy... +11:36 < sam_moore> Just in case +11:36 < jtanx> I tried removing the line i made to fstab +11:36 < jtanx> still didn't work +11:36 < sam_moore> Damn +11:36 < jtanx> was there anything important on it +11:36 < jtanx> could just flash it with the internal memory +11:37 < sam_moore> I'd be very hesitant to flash it to the internal memory if it doesn't boot +11:37 < sam_moore> There may be something else wrong with it +11:37 < jtanx> I mean +11:37 < jtanx> there's a script +11:37 < jtanx> to copy the internal memory to the sd card +11:37 < sam_moore> Oh, that's cool +11:37 < sam_moore> We can do that +11:37 < jtanx> Okay +11:37 < sam_moore> Since Ubuntu magically solved our image problems, we should use it +11:38 < jtanx> yeah, that was weird +11:38 < jtanx> just as long as there's nothing important on the sd card +11:38 < sam_moore> Probably change the phrasing in the report from "We don't know why this fixed it" to "We fixed it through our 1337 haxor skills" +11:38 < jtanx> >.> +11:39 < jtanx> how did you set up usercake? +11:39 < jtanx> like where's this db-settings.php that it mentions +11:39 < jtanx> oh right +11:39 < jtanx> never mind +11:41 < sam_moore> It comes with an install directory that I removed +11:41 < sam_moore> We should add that because we might want to modify the database structure +11:41 < jtanx> yeah +11:41 < sam_moore> Eg: It's a real pain that user permission levels are in a seperate table to everything else +11:41 < sam_moore> And doesn't make sense... +11:41 < jtanx> normalised database design +11:41 < jtanx> :P +11:42 < sam_moore> We can go with the "strcmp(user, "admin")" for now +11:42 < sam_moore> Did you get the email about LEFT vs RIGHT can? +11:42 < jtanx> yeah +11:42 < sam_moore> I think I should ask +11:42 < jtanx> pneumatics wants left +11:43 < sam_moore> "Are you looking at the box from the FRONT or BACK" +11:43 < jtanx> everyone else thought right +11:43 < jtanx> hahahaah +11:43 < sam_moore> "Also, which side is the FRONT" +11:43 < jtanx> the case team got blasted this morning +11:43 < sam_moore> Uh oh +11:43 < jtanx> apparentl +11:43 < jtanx> y +11:43 < sam_moore> Then again, we all get blasted every week +11:43 < jtanx> true +11:43 < sam_moore> I think we manage to get the least blasting, but still +11:43 < jtanx> oh yeah +11:44 < jtanx> apparently the report IS individual +11:44 < jtanx> per group +11:44 < jtanx> I'm getting confused +11:44 < sam_moore> Haha +11:44 < jtanx> and it's due next monday??? +11:44 < sam_moore> -_- +11:44 < jtanx> confusion over the due date +11:44 < jtanx> adam apparently didn't know, and thought sparkplus would send out the date +11:44 < jtanx> i don't know what sparkplus has got to do with this though +11:45 < sam_moore> How long does it have to be? +11:45 < jtanx> I have no idea +11:45 < sam_moore> Sparkplus is for our peer evaluation +11:45 < jtanx> yeah +11:45 < jtanx> but the relation to report due date? +11:45 < sam_moore> Pretty lazy if you ask me, and I don't care that they might read this (frankly I doubt it) to get the students to mark each other +11:46 < jtanx> Yeah +11:46 < jtanx> urgh +11:46 < jtanx> getting 1045 (access denied) from mysql +11:46 < jtanx> pretty sure I've got the pwd right +11:46 < sam_moore> It might be what happens in the Real World (TM), but this is not some company, this is university, my grades should not depend on someone else +11:47 < sam_moore> Try with `mysql` from the command line and check the password? +11:50 < jtanx> herp derp +11:50 < jtanx> no password +11:53 < jtanx> now that I think about it, it's almost exactly like the django idea, except in php +11:53 < sam_moore> Yep +11:53 < jtanx> when you install it, is there any default login? +11:54 < sam_moore> No; you register an account and that account becomes the admin +11:54 < sam_moore> (of course you can also manually update the database) +11:54 < jtanx> okay +11:54 < sam_moore> You could change the install script to do that +11:54 < jtanx> how are usernames and passwords stored in the db? +11:54 < jtanx> crypt? +11:54 < jtanx> (do you specify the algorithm?) +11:55 < sam_moore> passwords are crypted with SHA6 +11:55 < sam_moore> The other stuff is plain text +11:55 < sam_moore> UserCake default is SHA1 or something +11:55 < jtanx> sha6? +11:55 < sam_moore> I changed it to SHA6 +11:55 < jtanx> does it use something called pcks +11:55 < jtanx> pcsk2 +11:55 < jtanx> pcks2* +11:56 < sam_moore> Wait... by SHA6 I mean "SHA-Whatever-you-get-with-$6$-in-the-salt" +11:56 < sam_moore> Which is actually SHA-512 +11:56 < sam_moore> Derp +11:56 < jtanx> hmm +11:57 < sam_moore> I couldn't find any references to security issues with doing it this way, but that doesn't mean there aren't any +11:57 < sam_moore> However there's always going to be some security issues with anything we use +11:57 < jtanx> yeah +11:57 < jtanx> about the stuff in login.c +11:57 < jtanx> you'll definitely want to change tat +11:57 < jtanx> "SELECT password FROM %s WHERE user_name = \"%s\";", +11:58 < jtanx> sql injection right there +11:58 < sam_moore> I know +11:58 < sam_moore> That's why Login_Handler removes all non alpha-numeric characters from the user name +11:58 < jtanx> the password? +11:58 < sam_moore> The password isn't part of the MySQL query +11:58 < sam_moore> Only the user name +11:58 < jtanx> oh right +11:58 < sam_moore> So the password can contain any characters +11:59 < jtanx> yeah +11:59 < jtanx> I know with sqlite you can use placeholders +11:59 < jtanx> then you 'bind' stuff to those placeholders +11:59 < sam_moore> I think you can with MySQL too +11:59 < jtanx> probably +11:59 < sam_moore> Feel free to change it if you want, I figured it was safe if username was already checked for bad characters +12:00 < jtanx> yep +12:03 < jtanx> Pneumatics and electronics wants to test the system this wednesday, starting from 10am +12:03 < sam_moore> That provides a convenient reason to not work on ENSC1001 +12:03 < sam_moore> So I'll be there +12:03 < jtanx> Hahaha +12:08 < jtanx> nneded php5-gd +12:08 < jtanx> these username/password restrictions are attrocious +12:57 < sam_moore> The UserCake ones? +12:57 < jtanx> yeah +12:57 < sam_moore> Also, sorry I'll be late today +12:57 < jtanx> that's ok +12:58 < sam_moore> We're getting there... +12:59 < sam_moore> The question is, do they want a finished project, or do they want a report on an unfinished project +12:59 < jtanx> hahaha +12:59 < sam_moore> Because we'll have to devote a lot of time to the report +12:59 < jtanx> yes +13:00 < sam_moore> We've looked into every layer of software from linux kernel drivers to databases and human computer interaction... +13:01 < jtanx> o.o +13:01 < sam_moore> I think we can conclude that they are all horrible hacks tied together with shoelaces +13:01 < jtanx> about usercake +13:02 < jtanx> is it just +13:02 < jtanx> you have a set of pages +13:02 < jtanx> and they're either accessible or not +13:02 < jtanx> to the user? +13:02 < sam_moore> Yep, and I know you're thinking we could put the API as one of those pages :P +13:02 < jtanx> haha +13:02 < sam_moore> But I think it's better to keep the API seperate from user management +13:02 < jtanx> didn't think that +13:02 < jtanx> but was just trying to understand usercake +13:02 < sam_moore> Fair enough +13:02 < jtanx> but since everything's in php +13:03 < jtanx> we can do some templating +13:03 < jtanx> instead of hacking it together with javascript +13:03 < sam_moore> Sure +13:03 < sam_moore> The main modification I wanted to make was a page that lets the admin upload a list of users and get rid of the registration page (or restrict it somehow) +13:04 < jtanx> yeah +13:04 < jtanx> but some things are better done in php +13:04 < jtanx> like that whole 'load the sidebar in jaascript' is crap +13:04 < jtanx> and the whole 'if not logged in, redirect the user with javascript' too +13:04 < sam_moore> Feel free to add php to the main gui, it is probably a good choice +13:05 < jtanx> okay +13:05 < jtanx> once I figure out how stuff works +13:05 < sam_moore> I'm just not experienced enough with this sort of web development stuff +13:05 < jtanx> I can probably say it's a learing experience for everyone +13:05 < sam_moore> Yes +13:06 < sam_moore> I am pretty happy with how we've done this, even if it doesn't get finished +13:06 < jtanx> Yeah, it's quite ok +13:06 < sam_moore> I think we made good design choices +13:06 < sam_moore> Probably the best one was when you convinced me to use FastCGI instead of writing a custom HTTP server :P +13:07 < sam_moore> Or we'd be well and truly screwed when all the requirements for security and user management came along +13:07 < jtanx> hahaha +13:07 < jtanx> php +13:07 < jtanx> custom webserver +13:07 < jtanx> well +13:07 < jtanx> it might have worked +13:08 < jtanx> call php5-cli with system +13:08 < jtanx> hahaha +13:08 < sam_moore> I could probably do something like that with enough time +13:08 < sam_moore> But it wouldn't really be worth it +13:14 < jtanx> oh right +13:14 < jtanx> you changed usercake to use crypt +13:15 < jtanx> I'll change the password for the sd card image back to what we used before +13:16 < jtanx> username will be ubuntu though +13:18 < sam_moore> ok +13:18 < jtanx> the internal memory still uses temppwd though +13:20 < jtanx> if we use usercake we'd have to use php throughout the stuff anyway +13:20 < jtanx> unless you want it to be publicly visible +--- Log closed Mon Oct 21 14:11:57 2013 +--- Log opened Mon Oct 21 14:40:20 2013 +14:40 -!- sam_moore [matches@motsugo.ucc.gu.uwa.edu.au] has joined #mctxuwa_softdev +14:40 -!- Irssi: #mctxuwa_softdev: Total of 2 nicks [0 ops, 0 halfops, 0 voices, 2 normal] +14:40 -!- Irssi: Join to #mctxuwa_softdev was synced in 3 secs +14:43 -!- jtanx [~asfa@130.95.54.13] has joined #mctxuwa_softdev +--- Log closed Mon Oct 21 14:49:35 2013 +--- Log opened Mon Oct 21 15:08:16 2013 +15:08 -!- sam_moore [matches@motsugo.ucc.gu.uwa.edu.au] has joined #mctxuwa_softdev +15:08 -!- Irssi: #mctxuwa_softdev: Total of 1 nicks [0 ops, 0 halfops, 0 voices, 1 normal] +15:08 -!- Irssi: Join to #mctxuwa_softdev was synced in 2 secs +15:17 -!- MctxBot [~twang@106-68-93-93.dyn.iinet.net.au] has joined #mctxuwa_softdev +15:22 -!- jtanx_ [~asfa@130.95.54.13] has joined #mctxuwa_softdev +15:22 -!- jtanx_ is now known as jtanx +15:22 -!- jtanx [~asfa@130.95.54.13] has quit [EOF From client] +18:17 -!- jtanx [~asfa@106-68-93-93.dyn.iinet.net.au] has joined #mctxuwa_softdev +18:36 -!- MctxBot [~twang@106-68-93-93.dyn.iinet.net.au] has quit [Ping timeout] +22:14 -!- jtanx [~asfa@106-68-93-93.dyn.iinet.net.au] has quit ["ChatZilla 0.9.90.1 [Firefox 24.0/20130910160258]"] +22:57 -!- MctxBot [~twang@106-68-93-93.dyn.iinet.net.au] has joined #mctxuwa_softdev