00:39:52 rbm has quit 00:47:42 rbm has joined #openacs 00:48:27 tdxdave has quit 01:00:28 !@#$%)OO#! 01:02:06 I'm in class, but not busy :-) 01:03:49 tdxdave has joined #openacs 01:06:55 tdxdave has quit 01:12:15 tdxdave has joined #openacs 01:26:03 * rbm wonders if tdxdave is here to stay this time 01:26:44 it looks like I am. 01:26:59 I setup a printer in MS-windows and it won't print... 01:27:45 * tdxdave will use the color laser at work next time 01:29:44 aha an old fashioned IRQ conflict! 01:30:19 looks like I have to temporarily depart to change jumpers on an old soundblaster 01:31:24 tdxdave has quit 01:49:15 tdxdave has joined #openacs 01:55:07 andrews has joined #openacs 01:55:55 phew IRQ conflict resolved, now I am just out of ink... 01:56:25 andrews is now known as andrew_s 02:02:36 rbm: what sort of class? 02:56:05 Wow this is cool. I am playing the new online "game" Majestic. majestic.ea.com 03:10:22 tdxdave has quit 04:12:20 zzzirk has quit 04:12:20 GEM has quit 04:16:27 hey rjr 04:17:07 MadDucks has joined #openacs 04:17:25 GEM has joined #openacs 04:17:25 zzzirk has joined #openacs 04:17:27 zzzirk has quit 04:17:27 GEM has quit 04:17:30 GEM has joined #openacs 04:17:30 zzzirk has joined #openacs 04:19:26 andy__ has joined #openacs 04:31:05 hey andy 04:43:39 hey rbm 04:43:45 if your still here 04:44:09 Hi andy, you from Salt Lake? 04:44:14 that i am 04:44:25 how about yourself? 04:44:28 Same; Cottonwood area. 04:44:32 cool 04:44:58 you in school, work or what? 04:45:10 Just working right now, you? 04:45:16 school 04:45:21 UofU? 04:45:26 westminster 04:45:31 Groovy. 04:45:34 take my first CS class this fall 04:45:41 Woohoo! 04:45:44 i'm excited as hell to learn some programming 04:46:00 It's a lot of fun. Did you do any in high school? 04:46:04 hopefully i can apply it to the tickle stuff 04:46:18 nope, wasn't much into computers in highschool 04:46:18 He, he. When I was in high school they were teaching Pascal. 04:46:53 you working in a CS field? 04:47:04 Yup. 04:47:13 would i have heard of it? 04:47:55 I've recently worked for an ecommerce company and a medical software company. Now I'm just contracting on my own. 04:48:11 fun fun 04:48:31 made any money off the openacs stuff yet? 04:49:16 Not really, most the stuff I have done wasn't involved with the OpenACS. 04:50:20 Have you worked with the ACS at all? 04:50:41 with the help of rbm i got a tiny little server going 04:50:46 that is about it 04:51:05 like i said i don't know any programming 04:51:05 Cool, it's a good start. 04:51:11 i hope so 04:51:29 whoa 04:52:06 hey rbm 04:52:50 lot of talking going on ! 04:53:07 that there is 04:53:57 hey andy, you use your nick "andy__" intentionally or accidently? 04:54:12 acciddently 04:54:19 Did you know you can change it with the "/nick newnickname" IRC command? 04:54:26 try it 04:54:27 nope 04:54:42 andy__ is now known as andyiscool 04:54:50 Congrats! :-) 04:54:54 thanks much 04:55:23 You can also register your nickname, so that people can send you memo's even if you're offline. 04:55:32 And also so no one else can use your nick 04:55:43 try /m nickserv help 04:55:50 that would be cool, but i got to think of a good nickname first 04:55:55 (where /m is abbrev for /msg) 04:56:03 I just use my initials 04:56:18 Roberto? 04:56:37 Yes? 04:57:11 So we three (4 if we count zzzirk) are in UT. 04:57:34 Sorry, I met you once at USU. 04:57:37 We're going to have an OpenACS booth at our upcoming 2nd USU Free Software and Linux Forum, in November 04:57:52 Bruce Perens will be our speaker. 04:58:12 sorry for what? :) 04:58:42 I think I personally know at least 10-15 OpenACS users in UT. 04:58:50 For not recognizing your initials. 04:58:58 andrew_s: no problem 04:59:23 rbm can stand for a lot of things... Retroactive Butt Master, for example :-) 04:59:34 LOL! 04:59:43 What's the "b" for? 04:59:48 Barbosa 04:59:53 My 2nd middle name 05:00:25 andyiscool has quit 05:00:58 I'm glad there are more and more ACS developers from UT. Just met andy. 05:01:28 You've done a lot of proselyting. :) 05:02:21 As soon as 4.x is out I'll have a OpenACS 4.x workshop 05:02:29 At USU? 05:02:34 Yeah 05:02:44 I'll see if I can schedule one in SLC. 05:03:13 Maybe at the U 05:03:33 That would be great, though it's always nice to come up to Logan. 05:04:29 Yeah. 05:04:33 Logan's a nice town. 05:06:23 andyiscool has joined #openacs 05:08:05 i'm back 05:08:08 woohoo 05:08:13 hi 05:08:21 howdy again 05:08:24 woohoo to you too 05:08:59 what was the address for the bug reporting again? 05:09:17 openacs.org/sdm 05:09:26 Has anyone ever had problems with db_source_sql_file in OpenACS 4.2b? 05:09:37 hmm. no 05:09:50 i can't find a link to that anywhere on the openacs page 05:09:56 but i suspect i'm just blind 05:10:50 Emacs is _nice_ 05:11:23 andyiscool: hmmm. that's probably right :-) That's why Musea is redesigning the site 05:16:44 *sigh* none of my keys work anymore 05:17:11 * andrew_s gives Roberto a hug. 05:18:10 * rbm acts hugged 05:19:20 I gotta go shower, then to bed 05:19:28 Almost done with this ACS 4 Oracle app. 05:19:49 g'night 05:20:00 night all 05:21:10 night 05:21:41 off to bed for me, nice to meet you andy. 05:21:44 * andrew_s is away: I'm busy 05:32:10 snacks has joined #openacs 05:32:39 i'm having a file upload problem 05:36:37 snacks: what problems? 05:39:14 snacks: post your problem. I'll read it in the morning, or someone else will in the meantime. 05:39:23 Pascal should be getting up anytime now :-) 05:39:33 * rbm goes to bed for real 05:47:22 ;-) 05:47:37 i will try to figure it out. 06:12:08 snacks has quit 06:22:29 rzolf has joined #openacs 06:43:35 \quit 06:43:37 rzolf has quit 06:47:13 vinod has joined #openacs 06:58:27 vinod has left channel 07:50:35 abbaJ has quit 07:50:37 andyiscool has quit 07:54:39 abbaJ has joined #openacs 12:40:08 tdxdave has joined #openacs 13:10:08 markd2 has joined #openacs 13:31:47 * rbm falls off bed late 13:33:18 so that's what that "whomp" noise was 13:33:47 good thing 13:34:06 went to bed too late last night 13:34:58 did you sleep through stupid class? 13:37:27 Unfortunately no :( It's at 11 13:37:48 OH! It's still 7:37 AM. I thought it was 9:37 when I woke up 13:38:01 s/37/// 13:38:25 you can probably convert your partitions later. 13:38:28 oops 13:38:55 rbm you really gotta work on that :) 13:40:07 Yeah. I'm bad. I need to change the way epic4 displays which window is active 13:40:29 it's 9:37 out on the east coast :-) 13:40:51 Maybe my clock was set to east coast time 13:40:53 :) 13:41:14 So I read there will be an Open Source Database Conference in Rhode Island. 13:41:56 But like almost all open source conferences, you have to be rich or be going through work to attend. Early-bird registration prices start at $495 13:43:41 Thats not too bad. There is a free goverment technology conference coming and one class is $475 13:43:53 PascalS has joined #openacs 13:43:59 hiya 13:44:05 hi PascalS 13:44:16 Apple's WWDC is in the thousands to attend 13:44:18 tdxdave: That price is _just_ for attending. No classes, nothing. 13:44:39 If you want one class it jumps to $995 13:44:44 Aha 13:45:08 Well if people can afford to go Open Source must be making some money somewhere. 13:45:11 ey Pascal 13:45:32 morning to you roberto. 13:45:49 morning. What time is it at the Netherlands? 13:46:09 Or that is why all the dot-coms are having so much trouble... 13:46:18 CET+2 at the moment, 15:47 13:46:18 I have a friend whose dad is from the Netherland. His name is Ilko Minev 13:46:38 We're not *that* small a country 13:46:53 I know a guy from Brazil :) 13:47:07 We're not *that* small a country either :-) 13:48:03 I'd love to visit Europe, including Netherlands. 13:48:27 Well, we can spare a room, if you want one. 13:49:12 heh. thanks. Right now I can't afford even a trip to Yellowstone (a few hours away by car) 13:49:31 likewise, if you want to visit rural redneck western pennsylvania, we've got a room :-) 13:49:50 But please tell me you're not one of those people who thinks the capital of Amsterdam is the netherlands, where everybody is walking on wooden shoes sitting on a windmill blowing pot... 13:50:31 I'm finishing an ACS 4.x (Oracle) package. My job got a $300,000 quote from aD. They decided that was too much, and are paying me $10/hr instead :-) 13:50:34 Maybe wo should setup the OpenACS Bed&Breakfast network :) 13:50:49 s/wo/we 13:50:57 That'd be nice. 13:51:09 Who'll write the module then? 13:51:20 and where everyone raises tulips. plus all those kangaroos in Austria 13:51:27 We will have a big OpenACS booth at our upcoming 2nd Free Software and Linux Forum in November. 13:51:52 Cool. where will that be? 13:52:04 Here in UT. Bruce Perens will be our speaker. 13:52:14 s/speaker/keynote speaker/ 13:52:24 UT? 13:52:27 Utah 13:52:31 sorry 13:52:44 It's better than the french. 13:52:47 I should say Logan, Utah, USA 13:53:01 They just have numbers for their departments. 13:53:23 oh 13:53:40 not that anybody uses them, though. 13:55:13 Has the address book module been ported yet? 13:55:20 Our local OpenACS Users group will be at the forum. Ben and Don might come too. 13:55:33 * rbm shrugs 13:56:03 No. It was assigned, but apparently nothing has been done. 13:57:28 I thought someone was recently asking about it here 13:57:51 * tdxdave wishes someone else in Albany,NY used OpenACS 14:00:47 durn. I'll have to fire up Oracle now. 14:02:07 aha here it is: i am debugging the data model for address-book 14:02:37 good. 14:02:43 I might need it RSN 14:10:43 RSN? 14:13:12 Real Soon Now 14:13:24 or Registered Sadistic Nurse 14:15:54 ah 14:34:49 Someone is asking on the bboard what he should read to get into OpenACS 4 and he already read the aD docs and all the OpenACS 4 guides. I think we should assign him a package and get him coding! 14:37:13 PascalS has quit 14:54:26 derfy has joined #openacs 14:59:43 markd2 has quit 15:07:32 markd2 has joined #openacs 15:07:41 tdxdave: you got that right 15:45:16 Well all that work yesterday, I found out I missed CVS adding some files. 16:02:08 hmmmmm: static-pages breaks on a full CVS checkout... 16:02:26 I thought Dan applied those patches. 16:03:29 aha the patch in CR is in, not in acs-objects-create.sql 16:04:49 Great! 16:06:16 I mean one patch is in, but not the other... 16:06:22 ah 16:06:44 the other patch is to acs-kernel. 16:15:38 back 16:15:41 * andrew_s is back (gone 10:53:56) 16:15:47 welcome back 16:15:52 Oops, thanks. 16:16:55 andrew_s: You can use /quietaway to avoid the "hath returned" messages 16:17:13 derfy has quit 16:17:32 I thought we were trying to improve things a bit. 16:17:35 argh 16:17:46 in what way? 16:18:15 never mind wrong window huh? 16:30:22 anyone know if it's a bad idea to have an update function in a new ACS type package? 16:31:25 update function? 16:31:46 You mean package_name__update like package_name__new? 16:32:04 To change an acs_object? 16:33:08 yeah 16:34:06 There seem to be some already. 16:34:21 I have a type, with new and delete 'methods'. I wanted to do an 'update' method, but thought that might be a bad idea, because if the user does not pass all parameters, a bunch will have default null (the way I did it) 16:34:47 there are? 16:34:50 They all are updating one particular attribute usually such as update_permission or apm__update_paramter 16:35:19 hmmm 16:35:28 That's a better way to go 16:36:04 * tdxdave loves grep 16:36:12 yep 16:36:57 * tdxdave is in for alot of work 16:37:10 static-pages works with sitewide-search 16:37:23 horray! 16:37:24 or rather is supposed to. I'll have to fix it. 16:37:42 I haven't even seen the new sws yet. 16:37:49 markd2 has left channel 16:38:52 sorry for the poor phrasing there :) 16:44:05 argh. now my previous openacs-4 is broken. 16:44:32 I was getting some request errors yesterday 16:45:03 I think its becuase my copy is out-of-sync with CVS 16:45:22 hmm. I think I'm saying "hmmm" too many hmmmming times. 16:45:26 I forgot I can CVS update the whole tree because I committed static-pages 16:45:34 aha fixed. 16:45:41 you can or you can't? 16:46:16 I can now. because I committed my work. I didn't want to have CVS mess up my changes. 16:46:25 * tdxdave is not a CVS expert yet 16:52:06 tdxdave: if you do a CVS update in a directory where you have modified files that you haven't committed yet, CVS will not mess them up. 16:54:40 That is very handy to know. I read alot of CVS docs, but I think using it is the best way to learn. 17:01:17 The CVS Quick reference sheet helps 17:01:25 And the CVS article on ASJ too 17:10:55 quick reference sheet, i'll find that.. 17:11:23 tdxdave: www.quickref.com (or is it quickrefs) 17:11:37 here is another cvs for web sites site: http://www.durak.org/cvswebsites/howto-cvs/howto-cvs.html 17:12:09 I have to get that bot setup if we are going to fling URLs around here. 17:19:54 markd2 has joined #openacs 17:20:41 wb mark 17:20:48 thx 17:20:59 needed to get some grocery shopping done 17:21:12 when does Vinod usually show up around here? 17:21:23 Around 12 I guess 17:21:36 (MST) 17:21:38 loggy, grep 1-2 vinod 17:21:39 I'm logging. I found 372 answers for 'vinod' (showing 1...2) 17:21:40 1) 2001-07-27 06:58:27 vinod has left channel 17:21:41 2) 2001-07-27 06:47:13 vinod has joined #openacs 17:22:01 err. What timezone is that? 17:22:07 UTC, I think 17:23:00 Yep, UTC 17:23:19 coolio. thanks 17:31:31 www.refcards.com 17:33:24 they have emacs too. 17:37:51 hey its lunchtime! 17:37:56 * tdxdave goes to lunch 17:39:04 mmm.... lunchtime refcards 17:39:12 * rbm spits 17:46:11 AaronSw has quit 17:48:02 AaronSw has joined #openacs 18:07:48 * tdxdave returns 18:12:29 Argh! aD makes you fill out a form to DL AOLserver 3.3+aD13 18:12:48 everything's funneled through the same download mechanism 18:13:20 I am lazy and I like to use wget to dl stuff to my server. 18:13:35 * markd2 agrees 18:13:45 I usually end up using lynx on my server to snarf stuff like that 18:14:03 yeah I have that too. I didn;t even think of it. 18:14:42 I compiled 3.3 ad13 and I'm having problem with OACS 4.x 18:15:13 uhoh, well I am trying it for my 3.2.4 site 18:15:36 It works well with my 3.2.5-and-a-half system 18:16:26 outlook express is not outlook? 18:16:40 true 18:16:48 no but it can get most outlook virii 18:17:12 I just heard that. I thought that express was just a trimmed down version of outlook 18:17:23 its just email 18:17:59 I wanted to checkout Evolution, but the way GNOME is going I probably need 60+ libraries to install it 18:18:27 I'm happy with my mutt, but I like to look around once in a while 18:18:48 mutt is faster than any graphical mail client 18:19:53 Ok. somebody explain to me the difference between membership and composition again please 18:20:13 is membership user -> group, and composition group -> group ? 18:20:23 * rbm digs for Dan's post on the topic 18:23:18 should I recompile pg-driver if I upgrade aolserver? 18:24:52 it should probably Just Work 18:25:02 I don't think the C db apis go through too much mutation 18:25:16 cool 18:25:30 if you have problems, save off a copy and recompile :-) 18:37:43 Ok I ran make on aolserver and I can't find where it compiled everything to. This is getting wacky ( or its me) 18:38:33 It should be wherever you compiled. Now you have to install it somewhere with make install INST=/path/to/install/at 18:38:56 see the Simple AOLserver Installation Guide in the OpenACS documentation 18:39:08 ok.... I don't remember doing this before. Weird I am sure I did it correctly last time. 18:40:25 actually I just changed my mind, I am using nsvhr and I do not want to think about changing the server... 18:40:29 if you downloaded the +ad source, it usually puts it in a 'root' directory at the same level with the rest of them 18:42:14 [GlobalNotice] We would like everyone's opinion on this matter. Our new irc daemon is ready to go on line, and we are looking for the least possible intrustive time to make the change. Should we make the change as soon as possible, or wait until the end of the weekend? The change may cause up to 20 minutes of down time on the network. Please tell us your opinion in #openprojects, or mail your opinion to support@openprojects.net. T 18:43:38 rbm has left channel 18:43:45 rbm has joined #openacs 18:53:05 The dancer-ircd server conversion will take place at 3AM EST (7AM GMT). The services database will be exported in another 3 hours. Please make adjustments to your channel lists and nickname passwords. 19:00:05 Clarification about the services database... Changes made after 6PM EST (10PM GMT) will not be reflected in the new services database. 19:10:34 anyone compiled nsxml with xslt support? 19:16:57 I have. 19:17:11 Cool. Which version of libxml2 and libxslt did you use? 19:17:38 libxml2-2.4.1-1 19:17:41 and 19:17:43 I just tried them all. I am getting a parse error in xsltInternels.h 19:17:59 libxslt-1.0.1-1 19:18:09 hmmmm.. must be me 19:18:10 til has joined #openacs 19:18:15 I used the RPMs (i know, i know) 19:18:36 I used the rpms as well, so I don't think that is the problem... 19:19:01 weird. I tried all three versions of libxslt I could find and still got the same error. 19:19:27 have you looked at the xsltInternels.h file? 19:19:44 I am but I doubnt that will help, I don't know C 19:22:22 Hmm... I'll let you know if I have any ideas. 19:23:16 thanks. I suspect it is something with this machine setup. I should start all over with it. 19:24:02 * rbm hands everyone in the channel Debian and apt-get 19:24:17 heh 19:24:35 Slackware for a better world (tm) 19:24:40 LOL! 19:25:47 (as Mother Debian): "Come unto me, all ye that are sick of RPMs, bad security, clunky things; and I'll relieve you of your woes, for I have apt, Debian Policies and Guidelines" 19:28:48 I tried doing apt-get oracle-8.1.7, and it didn't work :-( 19:28:52 * tdxdave is apt-getting libxslt for his debian box 19:28:52 that would be spiffy! 19:29:25 its compiled and installed on debian. 19:29:37 I think I will backup my web sites and redo that machine. 19:30:05 the "unstable" debian is easier to maintain than "mature" redhat 6.2 19:30:37 I'll shutup about it now. 19:31:02 heh 19:31:05 "Debian's unstable is more stable that most other distro's stable" 19:31:10 (rbm) 19:31:20 I'm a SuSE achiever 19:31:35 I heard very good things about SuSE. I actually installed it once. 19:31:49 it's what Oracle uses for their linux development now 19:32:04 Now that RH has RedHat Database :-) 19:32:07 that is a good reason 19:32:09 plus my German security expert is most familiar with it, whcih makes audits of my box easier 19:32:16 But they still maintain a Red Hat version, no? 19:32:31 don't know about the 9i series 19:33:02 given all the bizarro games that RH is playing with compilers and everything 19:34:21 abbaJ has quit 19:37:04 abbaJ has joined #openacs 19:40:15 til has quit 19:57:47 I also did an rbm 19:57:53 s/also/almost 19:58:09 et? 19:58:11 s/et/eh/ 19:58:15 Woohoo XSLT works. http://dave.deepskydesign.com/xml/ 19:58:23 I almost typed in the wrong window 19:58:28 "yeah, and if you get a duck stuck there, it'll take some work to get it out" kind of thing 19:58:30 oh, heheh 20:01:39 but I don't know if I can run the bot on this machine because I sometimes need to switch to windows. 20:02:27 I also probably should program it to convert the file to html every 10 mins or something instead of rendering every time the page is hit. 20:14:18 ~anyone how do I set the "default privilege" on an ACS object? 20:14:27 s/how/know how/ 20:20:47 * rbm heads to class 20:34:48 that is a good question... 20:37:34 you could use security_inherit_p=yes and inherit from the parent object. 20:53:39 be back later 20:53:44 tdxdave has quit 21:08:11 gilbertw has joined #openacs 21:36:55 moo 21:37:01 moo 21:37:14 but how do you setup the default of the parent object? 21:37:22 hi gilbertw 21:38:01 hi rbm 21:40:20 are you setting up permissions or privileges? 21:43:44 yeah 21:44:54 You missed the backlog 21:45:15 I asked: anyone how do I set the "default privilege" on an ACS object? 21:45:26 oh ok 21:48:16 can't you use acs_permission__grant_permission to grant the object a privilege? 21:50:51 But with grant_permission I'm granting to a specific user. 21:51:17 W ant to make the default privilge of a group "read" so that every new user that comes into the group has that privilege. 21:51:26 s/W ant/I want/ 21:51:36 bogie! 21:51:50 hmmm. thought I had it, but no. 21:52:26 can't you grant the privilege to the group? 21:54:09 I think this should work. 21:55:27 hmmm. 21:55:37 and I believe that the permission system knows which group you are a part of 21:55:42 If that's it then it totally scaped me 21:55:56 Well, what if I'm a member of 10 groups? 21:56:21 and? 21:56:38 nm 22:01:31 markd2 has quit 22:03:02 markd2 has joined #openacs 22:26:23 so if I have a group 'A' (with object_id say, 15), and I want to make its default privilege 'read', I do: grant_permission(15,15,'read') ? 22:37:23 acs_permission.grant_permission ( 22:37:23 object_id => v_user_id, 22:37:23 grantee_id => v_user_id, 22:37:23 privilege => 'read' 22:37:24 ); 22:37:28 Looks like it 22:43:54 are you making the group object readable to the group members or are you making some other object readable by the group? 22:45:30 the former 22:46:45 oh. hmmm, that's interesting. yes that would work. 22:47:54 Like I said before, I'm just trying to make "read" the default privilege for the group 22:49:25 are you trying to make "read" the default privilege for the group on all objects? I think I'm missing something here 22:49:58 on all objects that will eventually be owned by the group. 22:50:03 you should also set security_inherit_p=f for that object 22:50:16 eh?? why? 22:53:23 sorry, it depends on how you created the object. if it doesn't have a context_id, then you're ok. 22:53:34 and you don't need to set the security_inherit_p=f. 22:54:54 It does have a context_id 22:55:44 Hmmm. Actually, not. 22:56:02 But if it did, why do you say securit_inherit_p should be 'f'? 22:56:28 Argh! 22:57:18 for instance, if your main site is set so the_public has read permissions on all objects, then the group object will inherit the read permission. you have to set security_inherit_p to f to stop it from inheriting the main site's permissions 22:58:25 Actually it does have a context_id because I'm putting this group in a composition with another group 22:59:18 what is the context_id of the other group? if you don't want that group to be able to read this group's object, then you need to set it to 'f' 22:59:19 So I have to make the top level group for this app have security_inherit_p='f' then. 22:59:37 Let me explain: 22:59:51 yes, I think that permission document explained it really well 23:00:23 This is an application (GPL'd) where schools will be able to keep track of all incidents that happen in that school. The information is highly confidential. 23:01:06 I have 3 main groups: state, district, and school. One state has several districts, which in turn have several schools. 23:02:02 So for my application, the state is the top level group. 'districts' are a composition with 'state', and 'schools' a composition with 'districts' 23:02:35 If you're a user assigned to a 'state' you should be able to see everything under that state. 23:03:08 Does this make more sense now? 23:03:11 yes 23:03:33 so if you're at the school level, you can only view your school. correct? 23:03:38 correct. 23:03:50 But I want everyone in the school to have read-only privs at first. 23:03:50 and if you're at the district level, you can see all schools in your district? 23:04:15 Yes, but same thing for districts. Even if you are at the district level, very few people will have 'write' privs. 23:04:17 so they need read only priv's for their school or for the district, or for the state? 23:04:25 for everything. 23:04:37 (where everything == state, district, school) 23:04:38 oh ok. so everyone can read everything, but only a few can write. 23:04:42 yes 23:04:52 ok. so let me ask a question 23:05:01 not everything. They can read whatever their group has perms to read 23:05:07 ok 23:05:28 do you have a table which stores the incidents? 23:05:31 Yes. 23:05:52 I have an "incident" type, which is a subtype of acs_objects. 23:06:01 ok. that's where you set your permissions 23:06:15 let me think for a sec 23:06:19 ok 23:06:22 to make sure I got that right 23:06:54 cuz what you're really interested in is setting the permission on that incident object, correct? 23:08:31 markd2 is now known as ^ 23:08:32 Right... I guess. 23:08:36 ^ ? 23:08:39 <^> ^ is now known as markd2 23:10:20 so how are you mapping incidents to the schools? is it a column in your incident's table? 23:11:00 I forgot to mention this, but incidents ONLY happen at the school level. 23:11:56 right, does your incidents table have a column to indicate at which school it happened? 23:12:28 Given some of the shenanigans of administrators in this area, maybe having the incidents happen anywhere would work :-) 23:13:28 gilbertw: No. I just thought of setting the context_id of that incident to the school it belongs to, in my __new function. 23:14:30 markd2: heh 23:14:51 ok. so are the schools the groups? 23:15:24 schools are also groups. They are composition with the districts. 23:16:30 so this group 23:16:40 's object_id is the context_id for the incidence 23:16:42 ? 23:16:45 In the pages where the user enters incident information, he/she picks the school where it happened (if he/she is at district level). If he/she is only at school level, then the schol he belongs to is selected 23:16:53 gilbertw: exactly 23:17:35 hmmm... ok. now I understand you question :) 23:17:57 I don't use js much because it's a pain in the neck. And the things I do don't ask for it. 23:18:01 oops 23:18:40 gilbertw: great! :-) 23:18:48 so, am I on the right track? 23:19:02 ok. let me think about it. 23:19:51 let me think aloud. 23:20:13 * rbm hands gilbertw a microphone 23:20:26 thinking aloud is allowed. 23:20:42 or not :) something isn't clicking. gotta sketch this out on paper... 23:22:01 gilbertw: I could probably send you the jpg of our datamodel. 23:22:39 if you want. but I think I understand it 23:25:38 hrmmm... this is tricky. 23:26:33 I'm making a jpg. hang on 23:26:38 ok 23:26:55 [GlobalNotice] Bye bye Espernet services, its been nice knowing you ;) 23:29:30 gwong@orchardlabs.com ? 23:29:47 yeah 23:30:11 on its way 23:30:39 ok 23:33:04 got it 23:33:58 great 23:34:15 http://w4.siemens.de/en2/html/press/pk_eb/pierer_rede.html 23:38:52 hmmm... so at first glance, foreach school, I would grant permission on that school's object to the school group's object 23:39:14 ok 23:39:33 then when quering the incidents table 23:39:52 I would do a security check on school_id for this user. 23:40:02 which should pick up the correct permissions. 23:41:28 I'm not sure I understand your approach 23:41:29 if you are at the school level, the permissions system will recognize that and let you read the incident. And if the user is at the district level where the school is in, then that use should also be able to read it 23:42:50 yeah, I'm thinking aloud. this might not be the best way to do it. 23:44:43 hmmm... So you were setting the context_id for the incident to the school_id, correct? 23:44:54 yes 23:45:22 oh ok. then it should inherit the permissions on schools. nevermind what I said above 23:45:45 pheewww 23:45:53 I was starting to get confused again :-) 23:46:05 It's very easy to lose it with the permissions system 23:46:06 yeah. sorry. I was confusing myself too 23:46:12 tell me about it =) 23:47:49 you would have to set the permissions for each school when you create it. so do you create the group and fill in the sds_schools table at the same time 23:50:32 Yeah. That's what I was doing. 23:50:55 The schools will only be loaded in bulk, or through a function that I create to do that. 23:56:59 ok. so that should work. set the context_id of the incident to the school_id in the sds_schools table. Then give read permission to the school's group on the school_id and make sure the security_inherit_p=f if that object inherits permissions from another object 23:58:14 my brain is starting to hurt 23:59:16 I don't understand why do I need to mess with the security_inherit flag