00:01:42 loggy has joined #openacs 00:01:42 topic is: Welcome to the OpenACS Helpline | http://www.openacs.org/ 00:01:42 Users on #openacs: loggy abbaJ davb rbm AaronSw Psychephylax shagster chump 00:01:43 This channel is logged: http://www.blogspace.com/openacs/chatlogs/ and blogged: http://www.thedesignexperience.org/openacs/ircblog 00:48:59 greetings 00:49:02 * Psychephylax pokes dave 01:15:56 AaronSw has quit 01:22:06 hey! 01:32:34 AaronSw has joined #openacs 01:47:00 http://www.bris.ac.uk/ISC/cms/ttw.html 01:47:00 B: http://www.bris.ac.uk/ISC/cms/ttw.html from davb 01:47:19 B:|List of web-based WYSIWYG Editors 01:47:20 titled item B 01:51:08 http://www.labour.org.uk/lp/new/labour/docs/LONGSPEECHES/TBCONF2001SPEECH.html 01:51:08 C: http://www.labour.org.uk/lp/new/labour/docs/LONGSPEECHES/TBCONF2001SPEECH.html from davb 01:51:29 C:|Transcript of Tony Blair's Speech 01:51:29 titled item C 01:51:38 about what? 01:53:21 I didn't read the whole thing. About 9-11 and what we should do. 01:55:52 heh 01:55:54 dave is awake 01:56:52 Not just about that, but really how it all ties in with how the world is changing and we have to wake up because it's not going to stop. 01:57:52 I'm working on my resume 01:58:00 Have to describe what my project is all about 01:58:38 Somehow "Bugging dave and mark on IRC to get tips on how to structure my program" is not the right thing to write down :/ 01:59:26 Hey we just helped you with the syntax. You learned the system and build extensions. You took a web-based application toolkit and customized it. 02:00:08 But do I really know the system? 02:01:38 i can't even write a damn resume lol 02:02:09 You know more than alot of us. I haven't actually used the system yet. 02:06:25 that's impossible 02:07:49 ·Development of a support tool for the Customer Service and Technical Support Groups of Cablevision. The tool is based on the ArsDigita Community System with a custom data model and user interface as well as additional functionality incorporated into the system through the use of TCL and AOLServer. 02:08:38 cool. 02:08:43 gonna add more stuph 02:09:33 tell what the "added functionality" is. 02:10:13 Beautiful 02:10:16 ·Development of a support tool for the Customer Service and Technical Support Groups of Cablevision. The tool is based on the ArsDigita Community System with a custom data model and user interface as well as additional functionality that allows for cable modem queries incorporated into the system through the use of SNMP, TCL and AOLServer. Upon completion the tool will support approximately a million subscribers of broadband cable servi 02:10:46 good. 02:11:08 davb has changed the topic to: Welcome to the OpenACS based resume enhancement channel 02:11:13 rofl 02:11:19 can anyone change the topic? 02:11:56 yeah. I guess so. 02:12:22 Psychephylax has changed the topic to: Welcome to the OpenACS based resume enhancement channel, soon to be an OACS module 02:12:26 rofl 02:12:48 Cool! I'll tell Don you signed up to work on it 02:13:12 hehehehe 02:15:48 Oh 02:15:51 I might not be around tomorrow 02:16:01 I think they want me to rebuild a server rack 02:17:11 is that good or bad? 02:18:33 Good because i get to do something different 02:18:44 bad because it sounds like a lot of work I probably won't want to do :P 02:18:52 I hear it involves crimpers and stuph 02:19:05 moo 02:19:21 ¿oom 02:19:49 heh, someones gotta do it. Thats what I do at work now. The network admin has me do all the work he doesn't want to do. 02:22:08 hehehe 02:23:56 ok 02:24:04 Who's up for looking at my resumee 02:28:15 Anyone? 02:30:38 puhleeeeeze? 02:30:40 dave? 02:39:44 heh 02:46:39 Roberto? wanna seeeeee? :) 02:52:18 where is it? 02:52:32 umm http://www.neverhere.com:1085/resume2.pdf 02:52:39 lol 02:52:41 " 02:52:47 "twain assisted data entry" 02:52:54 sounds a lot better than "scan monkey" lol 02:55:54 ack pdf. I don't know if I am equipped to handle that. 02:56:10 heh 02:56:13 What formad you want 02:56:34 Probably plain text or html. 02:57:27 ugh 02:57:39 html or text always mess up the formatting 02:58:05 hmmm.... Really? Most places only want text resumes by email these days. 03:01:26 I have ghostscript. does that do pdf? 03:01:55 umm 03:01:56 not sure 03:02:18 most take word or pdf too 03:04:40 cool. 03:05:02 I got it. gs filename x11 03:06:45 You gotta do something with the twain-assisted thing. Noone who is hiring for the type of position you are looking for is gonna be fooled by that one. Unless they have a really good sense of humor. 03:11:29 Good luck. I have to go though. 03:11:44 davb has quit 05:58:18 abbaJ has quit 06:00:36 abbaJ has joined #openacs 08:34:49 ola has joined #openacs 09:47:16 sinaptico has joined #openacs 09:48:07 sinaptico has left #openacs 10:59:50 ola has quit 11:06:53 ola has joined #openacs 11:24:38 davb has joined #OpenACS 11:25:42 davb has changed the topic to: Welcome to the OpenACS helpline http://openacs.org 11:26:01 morning. 11:26:06 Hi 11:27:34 hmm...I'm trying to open up a shell in emacs so I can do "tail -f"... 11:29:03 M-x shell 11:30:02 thanks. 11:31:37 np 11:31:41 davb has quit 12:00:36 davb has joined #openacs 12:03:50 Psychephylax has quit 12:40:46 dave, when I disable or delete my package from the admin pages, will the system execute the drop script? 12:41:17 It will try. 12:41:35 I am not sure about disable, delete runs it. 12:42:17 ok. 12:43:15 everything installed without errors the first time I tried. I find that hard to believe.. 12:54:30 grr. "database support" only shows oracle8. 13:10:17 hmmmmm 13:10:39 i wonder if that is specified in the .info file 13:13:40 postgresql is specified, but not the sql/postgresql/package-create.sql , the drop file and so on... 13:14:21 perhaps I need to "Manage File Information" and "Scan for new files" first? 13:15:42 see, I haven't edited the info file at all. 13:17:48 sorry to bug you like this, dave;-) 13:22:37 np 13:22:49 i am not sure. try it 13:34:09 that was it. now I'm beginning to see some errors. 13:51:00 markd2 has joined #openacs 13:51:39 cool! that means you are making progress. 13:51:51 progress is good 13:52:00 I was always happy to get a new error. it meant the old one was fixed. 13:53:11 or introduced a brand new one :-) 13:53:18 * markd2 is feeling particularly cynical this morning 13:56:05 heh, i did that too! 13:57:58 markd2: it promises well. 13:58:21 you feeling x-tra cynical, that is:-) 14:01:50 heh 14:37:48 what is wrong with the following view: 14:37:58 create view mp3_mp3_playlist_map_view as 14:37:58 select m.mp3_id, 14:37:58 m.playlist_id, 14:37:58 m.sort_key, 14:37:58 coalesce(v.total,0) as total 14:37:59 from mp3_mp3_playlist_map m 14:38:01 left join (select mp3_id, 14:38:03 playlist_id, 14:38:05 sum(vote) as total 14:38:07 from mp3_votes 14:38:09 group by mp3_id, playlist_id) v 14:38:11 using (mp3_id) 14:38:13 left join v using (playlist_id); 14:38:23 I get: ERROR: Relation 'v' does not exist 14:39:03 the original view looks like this: 14:39:25 create or replace view mp3_mp3_playlist_map_view as 14:39:25 select m.mp3_id, 14:39:25 m.playlist_id, 14:39:25 m.sort_key, 14:39:25 nvl(v.total,0) as total 14:39:26 from mp3_mp3_playlist_map m, 14:39:28 (select mp3_id, 14:39:30 playlist_id, 14:39:32 sum(vote) as total 14:39:34 from mp3_votes 14:39:36 group by mp3_id, playlist_id) v 14:39:38 where m.mp3_id = v.mp3_id (+) 14:39:40 and m.playlist_id = v.playlist_id (+); 14:41:17 ack! I just entered a password into my windows machine by accident. I don't know what it is... 14:41:34 it's probably stored in cleartext in several files 14:41:45 i am not sure about the join syntax... 14:41:51 markd2: i am sure... 14:43:38 Psychephylax has joined #openacs 14:43:46 morning 14:44:58 hi 14:45:26 markd2 has quit 14:49:45 :) 14:50:38 lets seee...where was I 14:50:53 I'm getting moved again 14:51:47 oh...Dave, what do you think I should replace my "Twain assisted data entry" with...ehehe 14:52:27 no idea 14:52:41 ola: looks like you are missing a ( somewhere in there 14:52:43 brb 14:52:50 davb has quit 14:53:06 davb has joined #OpenACS 14:54:09 never mind about the ( 14:54:28 :) 15:00:23 i guess I need a custom search page now 15:01:39 fun! 15:01:53 Well 15:01:57 maybe I can hack this one 15:02:15 it just has to go find a person by e-mail address 15:03:36 That is definitely in ACS already. 15:04:36 yep 15:04:45 I just have to figure out what the hell %5 represents 15:04:49 I know %20 is a space 15:06:10 my bad 15:06:13 it's %5F 15:06:54 ola: try this 15:06:58 left join ( 15:06:58 select mp3_id, 15:06:58 playlist_id, 15:06:58 sum(vote) as total 15:06:58 from mp3_vots 15:06:59 group by mp3_id, playlist_id ) v 15:07:01 on (m.mp3_id = v.mp3_id and 15:07:03 m.playlist_id = v.playlist_id ) 15:07:04 aha 15:07:12 %5F is an underscore 15:07:34 I am just winging it though. 15:08:09 http://www.blooberry.com/indexdot/html/topics/urlencoding.htm 15:08:09 A: http://www.blooberry.com/indexdot/html/topics/urlencoding.htm from Psychephylax 15:08:20 A:| All you ever wanted to know about URL Encoding 15:08:20 titled item A 15:08:22 B 15:08:39 Psychephylax: OpenACS/AOLserver does url encoding for you. 15:08:39 A: Now with it's own converter! 15:08:39 commented item A 15:11:22 I know 15:11:25 I'm trying decode it! 15:11:27 lol 15:12:36 for fun? ad_page_contract should stuff the URL variables into Tcl variables for you 15:12:40 :) 15:13:20 no 15:13:38 I'm trying to search for an email by manually passing it parameters in a URL 15:13:58 Instead of coding the page first and then finding out that it doesn't work I'd much rather try the search manually first 15:14:22 And it searches for everyone still even though I tell it to find people with specific email 15:17:01 dont you need something like this: select * from users where email = :email_passed_in_form_variable 15:17:25 How many users will you have? If its not too many a drop down box might be better. 15:17:54 it'll be many 15:18:05 Ok, so thats a bad idea. 15:18:34 we are assuming the user knows the exact email address? 15:18:55 yeah 15:19:08 of course knowing my boss that will change once i mention this to him 15:19:14 heh 15:19:17 What fun! 15:19:30 not really 15:19:37 I think I'll use this Quick search thing 15:19:46 it seems to use keywords and can find multiple things 15:20:50 cool. 15:21:33 now I just have to figure out how it works 15:21:33 shagster has quit 15:21:34 lol 15:22:14 aha 15:22:16 it calls search 15:22:37 my package is gonna be huuuge 15:22:48 cuz I'm duplicating all the stuff 15:24:40 davb: thanks! it works. 15:24:42 yeah but your package is really "the system" 15:24:50 shagster has joined #openacs 15:25:17 ola! I can't believe it. what luck. Maybe one day I'll actually understand what all the stuff means. I copied off a join DanW helped me port. 15:25:25 lol 15:25:38 I rule 15:26:54 actually..."Eureka!" :) 15:28:06 davb: cool! 15:32:45 muahaha 15:32:50 unfortunately I posted to the forum before... 15:33:29 hi psyche! 15:33:29 shagster has quit 15:33:34 So post back 15:33:37 Hi Ola 15:33:55 Post back with a solution to your problem this way if someone has a similar problem they might learn 15:34:09 there should be an "unpost" option. 15:34:43 shagster has joined #openacs 15:35:06 yes sir, but I want to wait out other possible solutions first:-) 15:36:40 Well 15:36:47 maybe not unpost but definetly an "edit" 15:37:45 shagster has quit 15:38:08 two replies already... 15:38:14 cool 15:38:44 shagster has joined #openacs 15:40:15 shagster has quit 15:40:44 shagster has joined #openacs 15:41:42 ola: There is a difference between ON and USING 15:41:56 but I have no idea what or if it matters for that query 15:43:48 alrighty! 15:44:10 hmmm 15:44:14 this won't be fun 15:44:18 of course not! 15:45:12 no, I have to figure out how to make it so that only the people who you work for and you can edit your information 15:45:49 that is a privilege (write) and a permission I think 15:47:00 davb: both ways are fine but USING is less code. 15:47:16 Interesting. 15:47:57 see Don's and Gilbert's replys. 15:48:11 replies 15:49:13 ON a.blah = b.blah apparently is the same as using ( blah) with on you have to specify which column matches in each table so they could have different name. Using they have to have the same name. 15:49:14 hmm 15:49:15 ok, I got a question 15:49:29 thats my new thing I learned today. 15:49:54 and this : " A side-effect of both 15:49:54 USING and NATURAL is that only one copy of each joined column is emitted into the result table (compare the 15:49:54 relational-algebra definition of JOIN, shown earlier). 15:49:54 " 15:51:14 coolio 15:51:56 This is more of a double check 15:52:12 if I want to pass a variable from page A to page C 15:52:23 wher A links to B first and B goes to C 15:52:24 You just do the ad_permission_p on both pages. 15:52:29 davb: on is good if the tables uses different keys 15:52:31 I have to page_contract the variable in B 15:52:37 and then pass it into C from B again? 15:52:47 Thats what edit-this-page does. 15:53:03 A and B are ADP pages? 15:53:12 A,B,C are all ADP pages 15:53:16 you need something like form_vars 15:53:26 they're not form vars 15:53:29 you set form_vars in the TCL scripts. 15:53:29 shagster has quit 15:53:35 I have a user clicking on add or remove links for a user 15:53:52 the you do @form_vars@ in the ADP page. It encodes them as hidden form fields 15:53:53 I mean edit/remove 15:54:06 They're already hidden sorta 15:54:15 If the user hits edit I don't want the "delete" link showing up 15:54:23 if he hits delete I don't want the edit 15:54:45 shagster has joined #openacs 15:54:47 Oh. Ok. So the TCL page checks and sets a variable say display_edit_link 15:54:55 then check for it in the ADP template. 15:54:57 yes 15:55:14 and you want to pass that around between the pages? 15:55:20 but that page is like 3 pages inward from the page where you click on "edit/remove" 15:55:27 hmmmm... 15:55:30 so each page has to keep the state 15:55:35 Hang on 15:55:39 lemme try my idea :) 15:56:20 ok 15:57:12 Since you claim I know ACS now I should be able to figure it out 15:57:13 hehehe 15:58:00 rbm has quit 16:02:13 vinod has joined #openacs 16:02:46 heh 16:02:52 Hi vinod! 16:03:00 hey dave! 16:03:15 what's new? 16:03:26 Hey vinod! 16:03:42 Not much. Sick kids all week so I had to stay home and haven't got much work done. 16:03:46 hey Psychephylax! 16:03:50 bummer 16:04:00 Arthur's Preschool and Tonka Construction don't run in Linux. 16:04:11 :-) 16:04:13 ..yet 16:04:21 Hey, stupid question 16:04:25 in a page contract 16:04:35 I have the decode block listen for a "action" 16:04:51 can it also have a property to return to adp of "action" or is that a scope problem? 16:05:36 Don't know. Try it. If ad_page_contract doesn't like it, it will tell you. 16:05:46 ok 16:06:05 It should be OK. if action is defined. you should be able to change the value in the tcl code if you need to. 16:06:54 markd2 has joined #openacs 16:07:02 hmmm 16:07:04 Hey Mark 16:07:11 does "acs_object_type__drop_type" take two arguments. if so, what are the second for? 16:07:53 Psychephylax: see how easy it is to make it look like I actually know what I am talking about when I really am totally clueless! 16:07:54 can't find it in the api-browser. 16:07:55 hiya 16:08:08 hmmm. 16:08:37 same here, not in the API 16:09:15 uhmmm 16:09:20 ola: it's a cascade parameter. if 't', it's supposed to drop all types which are dependent on it (i think) 16:10:34 although looking at the code, it's completely ignored... (defined in packages/acs-kernel/sql/postgresql/acs-metadata-create.sql 16:10:57 On the API browser go to acs-kernel then SQL files. You can search through the code to see how it really works. 16:11:39 thanks vinod and dave! 16:11:51 aD planned ahead. That is why there are so many 46 parameter functions. 16:11:53 oooh. vinod is in the haus too 16:11:58 shouldn't you be out saving lives, or something like that? 16:12:11 vinod: you missed the outer join! 16:12:26 should, shouldn't - those are all very subjective terms 16:12:39 ugh - i hate outer joins :-) 16:12:41 subjunctive, too 16:12:49 help! I've got subjunctivitis! 16:13:01 hey, don't go grammar-school on me 16:13:06 w00t 16:13:48 vinod hates outer joins. then how does the rest of us feel about them;-) 16:14:13 haha - well, i love them once they're ported 16:14:20 haha 16:15:07 so mark, how's java treating ya? 16:16:01 i could use some java right about now 16:16:13 the project I was on fell through 16:16:21 doh! sorry to hear that 16:16:30 worked out best for both parties 16:16:45 so I'm taking advantage of my "on the bench time" do get more oracle studliness 16:16:56 hehe 16:16:57 Right now I'm reading a wacky book on Backup & Recovery 16:16:58 Oh 16:17:07 I had a test on Databases yesterday 16:17:14 THere is WHOLE book on backup? 16:17:27 There's like 2 paragraphs for Postgresql. 16:17:28 My oracle l33tness helped me I think 16:17:29 this one is 688 pages 16:18:10 We had to draw an E-R diagram of items with like 5 properties and a relation stored and another entity shelf which had 2 properties 16:18:16 Ah, a small Oracle book. 16:18:17 Then convert that to a schema 16:18:26 yeah. with large type 16:18:34 I am not sure if my syntax was correct 16:19:20 did you use boxes and lines? 16:20:17 yes 16:20:17 and ovals 16:20:24 Well, not for the schema 16:20:32 cuz he wanted us to create tables out of tose 16:20:43 Is this valid 16:20:47 oooh.. ovals 16:20:54 the sophisticated ER tools 16:21:08 I dropped out before we hit the parallelograms 16:21:09 * markd2 shudders 16:21:10 constraint validshelf check (size=100 OR size=150) 16:21:21 I think those are for the transverse normal form 16:21:36 parallelograms are relations :P 16:22:59 hmmm 16:23:00 I see 16:23:05 passthrough_parameters 16:23:09 Need to figure out what this does 16:37:58 john has joined #openacs 16:38:43 does anyone know how to port plsql output parameters to pgplsql? 16:39:54 john: see: http://www.brasileiro.net/postgres/porting/ 16:40:13 I've read it three times. Nothing about output parameters 16:40:48 ok. sorry I don't know much.. 16:41:19 that is the best doc I've found on pgplsql ... you'd think there'd be something more comprehensive somewhere 16:41:34 welcome to the wonderful world of ACS development 16:41:38 lol 16:41:43 where nobody knows anything 16:41:56 There are no output parameters in pl/pgsql I think. 16:41:59 Well, Mark knows oracle 16:42:19 * Psychephylax gives Mark a cookie 16:43:15 davb - I think you're right. I was hoping someone would say otherwise. 16:43:25 In the acs-worfklow module, there are lots of comments that say 16:43:31 TODO: output parameters 16:43:31 Label TODO not found. 16:43:57 hehe 16:44:02 I assume that means it's not straightforward, but my pgsql experience is measured in days. 16:44:02 A: 16:44:02 http://www.blooberry.com/indexdot/html/topics/urlencoding.htm 16:44:03 All you ever wanted to know about URL Encoding 16:44:04 (Psychephylax) Now with it's own converter! 16:45:35 I don't have too much experience either. too bad rbm is not around. 16:46:45 so, what could this be: 16:46:46 psql:mp3jukebox-create.sql:30: ERROR: Function 'acs_object_type__create_type(unknown, unknown, unknown, unknown, unknown, unknown)' does not exist 16:46:46 Unable to identify a function that satisfies the given argument types 16:46:47 You may need to add explicit typecasts 16:47:47 ola: go back in the log a little farther and see what values the function is getting. 16:47:57 what's this mean: {passthrough ""} in a page_contract? 16:48:12 Also, the parameters must be in order. They usually are not becuase Oracle takes named parameters. 16:49:09 I just looked up the definition for that function. It looks like it takes 10 args, not 8 16:49:23 fill in the missing ones, should work 16:49:45 Yeah. if there are defaults you need to fill them in. Postgresql cannout have default parameters. 16:49:54 In the oracle they are probably left out. 16:51:51 EUREKA! 16:51:58 I got it to pass the action! 16:54:27 cool! 16:55:41 yeah 16:55:43 I fscking rule 16:55:44 lol 16:58:51 yes! thanks davb and john. 10 args not 6. 16:59:14 hehe 17:00:41 ola: np. you will see that ALOT in porting. 17:00:54 cool! 17:03:51 where was that Expr thing on the web? 17:03:59 I gots to find out how to compare stings again hehe 17:04:34 john: the pg docs suggest that you can create a function with a return type of 'setof' that lets you return multiple params in a set. it doesn't give examples though 17:04:38 http://www.postgresql.org/idocs/index.php?sql-createfunction.html 17:04:38 B: http://www.postgresql.org/idocs/index.php?sql-createfunction.html from vinod 17:05:45 B:| PG docs for 'create function' 17:05:45 titled item B 17:08:55 Psychephylax: http://www.aolserver.com/docs/ 17:09:54 gracias 17:10:51 np 17:11:11 It is the easier place to find Tcl docs. Also Tcl for web nerds is handy. 17:11:46 yeah 17:11:48 I like that page 17:11:50 i bookmarked it 17:11:55 vinod has quit 17:12:42 plus it has the AOLserver tcl api stuff which is handy. 17:13:36 woohoo! no errors when I load the datamodel. 17:13:50 great! 17:15:18 heh 17:15:28 I'm having so much fun torturing one of my old x-friends 17:16:49 hmmmm 17:18:23 it's a long story that I prefer not to have logged :) 17:18:29 does anyone know if AOLserver sends a user-agent when it does an HTTP GET or POST? 17:18:34 k 17:19:38 if {[action == "edit"]} { 17:19:43 it's telling me that's not valid 17:19:49 you can have stuff not be logged by prepending "#" to the line 17:19:51 action s/b $action? 17:20:03 lose the square brackets 17:20:07 that too. 17:20:29 oh 17:20:30 right on 17:20:51 davb: check out $NS_HOME/modules/tcl/http.tcl 17:21:05 cool. 17:21:22 Aha! 17:21:35 I am using util_httppost. it sets it to Mozilla 1.01 17:23:10 psyche: you should use [string_compare $action "edit"] instead. 17:24:51 Is there an advantage? 17:25:22 if $action is a number, you'l get an error 17:26:17 well, it's hardcoded it can only be remove or edit 17:26:28 users can't choose :) 17:28:43 hmmm 17:28:57 I would like to hide the url stuff later 17:29:14 john has quit 17:34:37 vinod has joined #openacs 17:34:50 wb 17:36:26 aha 17:36:43 vinod: I found something I think should be changed in "A Day in the Life of a Porter" 17:38:25 vinod has quit 17:38:52 Woah! you scared him away! 17:38:58 lol 17:39:07 darn. 17:42:12 markd2 has quit 17:49:36 alrighty 17:49:55 I think I'm in alpha of my User interface 17:51:02 now I have to figure out all this extra permissions stuff 17:52:54 go gadget, go! 17:53:32 heh 17:53:37 I guess this is the hard part 17:53:39 :( 17:54:02 not only that 17:54:12 I'm not sure if my tables are right 17:55:23 rbm has joined #openacs 17:56:16 hey roberto 17:56:17 hi. 17:56:35 Psychephylax: the datamodel? 17:56:41 or HTML tables? 17:57:22 datamodel 17:57:30 I had to add 3 more tables 17:57:35 CM_PINGS 17:57:37 CMTS_PINGS 17:57:43 EMPLOYEE_HIERARCHY 17:57:50 Ok. 17:58:02 But I am not sure if I need to register any of them with Oacs 17:58:17 I think I might have to with EMP_H 17:58:22 I don't think so. They are for your application. As long as your code knows about them. 17:58:43 OACS doesn't care who is whose boss. 17:58:59 Yeah but I'm gonna have to check things like: Is the current user able to update this persons data? 17:59:00 heh 17:59:34 Oh. right like you said only someones boss can edit thier data. 17:59:52 YOu should probably model that in the OACS user/group structure somehow. 17:59:52 Yyes 18:00:01 well 18:00:04 I sort of did 18:00:05 lol 18:00:21 ACS knows there are Managers and Supervisors 18:00:35 but it only allows for these types of constraints: 18:00:38 Right. You almost need a group for EACH manager/supervisot. 18:00:42 All managers can edit all regular users 18:00:50 yeah 18:00:54 DISCLAIMER: davb has no idea what he is talking about. 18:00:54 Label DISCLAIMER not found. 18:00:58 it would be Manager A group 18:01:02 No, you're right 18:01:23 I would have to create a new relation for every new Manager and Supervisor 18:01:28 Well you can have the ALL managers permissions and the individual manager permissions. 18:01:31 and then users would hae to go to that 18:01:45 It would be too complex to code I think 18:01:45 I think your type of application is exactly why the permission system is so complex. 18:01:55 lol 18:02:00 Isn't that what the group relationships stuff is? 18:02:06 yah 18:02:11 I would have to do this 18:02:12 Most web sites have a webmaster and maybe a content editor or two. 18:02:19 yeah 18:02:41 Groups types: Manager, Supervisor, User 18:02:58 Groups: Manager (manager type), Supervisor (supervisor type), USer (main site user) 18:03:50 Sub Group Manager: John Smith (of Manager type) 18:04:03 Then if he creates a user they work under him 18:26:25 markd2 has joined #openacs 18:26:46 Anybody has libclntsh.so to mail to me please? (it's the oracle client library) 18:28:22 what platform / version? 18:28:32 Linux. 8i should do 18:28:38 8.1.6 or 8.1.7? 18:28:45 Hmmm. good question 18:28:45 lemon or tuna flavored? 18:29:03 I'll put 'em both up 18:29:17 If you could mail both, that gets rid of the possibility of me having to nag you again :-) 18:29:39 Stupid Oracle! Why not release a tarball jusst with the client libs?!??! 18:30:06 BTW, tune with a touch of lemon please :) 18:30:46 Oracle has to keep up their minumum pain factor. 18:31:32 markd2: Thanks a zillion! 18:31:38 m'pleasure 18:31:54 find out where Psychephylax gets his cookies. They're pretty good 18:32:37 now to dismantle my PC and see what's wrong with the network card 18:33:02 AaronSw: Wha?? 18:49:52 Hmmm, someone please remind me how to convert a cpio file to a .tgz? 18:50:12 <-- thinks RPM is an ugly hack 18:50:50 expand the cpio, then tar it up? 18:50:53 * markd2 grins weakly 18:52:12 please help me port this: 18:52:18 markd2: No... 18:52:34 FIX ME PLSQL 18:52:34 begin 18:52:34 :1 := mp3_playlist__vote_cast ( 18:52:34 mp3_id => :mp3_id, 18:52:34 playlist_id => :playlist_id, 18:52:34 user_id => :user_id, 18:52:36 vote_date => current_timestamp, 18:52:36 how do you expand the cpio (I guess you're right) 18:52:38 vote => :vote_num 18:52:40 ); 18:52:42 end; 18:53:42 heh 18:53:52 who registers a domain called agentpp 18:54:38 ola: select mp3_playlist__vote_cast ( params ) 18:54:58 ola: db_exec_plsql will do create an inline function for you. Just call db_exec_plsql (with the appropriate PG function syntax) and assign the result to :1 (there's a switch if I am not mistaken) 18:55:23 OH, I saw something different. davb is right 18:55:41 I never would have believed it. 18:55:41 * rbm twacks self for thinking the hard way 18:55:48 Sometimes it is easier in Postgresql than oracle. 18:55:49 aha, thanks. 18:56:55 gilbertw has joined #openacs 18:58:01 do I need the begin-end stuff? 18:58:03 Almost time for my very (Yawwwwwnnn) exciting class. 18:58:08 no 18:58:13 heh 18:58:18 thanks 18:58:40 Same professor of last semester's mega-lame (tm) class 18:59:53 * rbm wonders why all P2 lab computers were replaced with P3s with 500 Mb of RAM when all 99% of the students do is read their Yahoo/Hotmail mail and use AIM/MSN Messenger. 18:59:53 gilbertw has quit 19:00:31 So? 19:00:39 Gotta get them buzzwords in 19:00:50 Just amusing 19:01:19 does this look ok: 19:01:21 19:01:22 select mp3_playlist__vote_cast ( 19:01:22 :mp3_id, 19:01:22 :playlist_id, 19:01:22 :user_id, 19:01:22 current_timestamp, 19:01:24 :vote_num 19:01:26 ); 19:01:28 19:01:30 19:01:34 ? 19:02:02 Yeah. I'm worried about the assignment to ":1" though 19:02:19 ok. 19:02:44 ola: what does the Tcl that calls it look like? 19:02:46 share your worries...:-) 19:02:59 * davb has been wrong before 19:03:09 wait... 19:03:43 "But WAIT! Call now and receive a free foobar! A $50 value absolutely free!" 19:03:59 davb: like this: 19:04:01 db_exec_plsql vote_cast " 19:04:01 begin 19:04:01 :1 := mp3_playlist.vote_cast ( 19:04:01 mp3_id => :mp3_id, 19:04:01 playlist_id => :playlist_id, 19:04:02 user_id => :user_id, 19:04:04 vote_date => sysdate, 19:04:06 vote => :vote_num 19:04:08 ); 19:04:10 end; 19:04:12 " 19:04:29 So it doesn't return the value to a Tcl variable? 19:05:36 no. I don't think so. it's inside a catch statement, though. 19:06:01 OK. 19:07:53 hmm... the plpgsql proc returns "v_l_vote_sum". 19:08:15 mp3_playlist__vote_cast that is. 19:08:53 look at the syntax for db_exec_plsql. There's a return switch IIANM 19:10:38 ugh! 19:10:52 It would be nice to have something in the readme on how to compile the library 19:10:55 * rbm curses cpio 19:11:21 * Psychephylax curses C3PO 19:12:12 ummm 19:12:21 there's a Makefile.solaris 19:12:28 but no way to start it compiling 19:12:41 Get R2D2. He's smarter. 19:12:55 "and r2d2 said 'weeeeeeeeeeeeeee.... bleep!'" 19:13:12 heh 19:13:52 * rbm now feels like watching Star Wars 19:14:04 sigh 19:14:05 ok 19:14:11 this wants a cc compiler and I have gcc 19:14:20 that's ok 19:14:23 alias cc gcc 19:14:46 or there may be a makefile tweak to change your CC var to be gcc 19:15:13 yeah 19:15:15 i did that 19:15:17 change cc to gcc 19:15:22 It still bombs out 19:15:33 * rbm gives up and grabs a source file 19:15:36 rbm: how about: select mp3_playlist__vote_cast (blargh) into :1 ? 19:15:43 Nope 19:15:55 okay. 19:15:55 cc1plus: Invalid option `t' 19:16:16 At least I don't think that would work (unless the driver is handling such cases, which I doubt) 19:16:27 Mark, solaris 2.6 is Pre-SunOS what? 19:16:37 or is it post 19:16:41 Postgres doesn't have PL/SQL, of course, but it does have PL/pgSQL and 19:16:41 other procedural languages. Rather than assign the result to a bind variable which is then returned to the caller, the Postgres version of 19:16:41 OpenACS requires the caller to perform a select query that returns 19:16:41 the value of the function. 19:16:41 We are no longer calling db_string, which screws up the bind variable 19:16:43 stuff otherwise because of calling environments. (ben) 19:18:31 Oh crap 19:18:36 * rbm goes to class 19:18:43 Solars 2.6 is SunOS 5.6 19:18:53 so it's Solaris-new-improved land 19:20:01 So, if I have SunOS 5.7 19:20:08 Can I still compile stuff for 5.6? 19:20:25 ola: what tcl file are you in? 19:20:40 usually it works the other way around 19:20:47 stuff compiled on 5.6 runs on 5.7 19:20:52 but it's worth a try 19:21:11 no no 19:21:13 I have the source 19:21:20 but it says it's Solaris 2.6 19:21:29 I need to compile it for SunOS 2.7 19:21:35 not a problem 19:21:43 I doubt it says SunOS 2.7. it's probably solaris 2.7 19:21:56 if it says sunos 2.7, and was written in the last 10 years, it's solaris 2.7 19:22:56 No, it says Solaris 2.6 19:23:05 but uname -a says SunOS 5.7 19:23:19 then you're fine 19:23:23 Solaris 2.6 == SunOS 5.6 19:23:28 Not if I can't get it compiled :P 19:28:15 wait a second 19:28:22 cc is a c compiler or a c++ compiler? 19:28:28 C 19:28:33 davb: mp3jukebox/www/vote.tcl 19:29:53 k 19:30:21 i have no fscking idea 19:30:37 I think the :1 is a dummy variable. 19:32:12 oh. maybe I don't have to worry about it.. 19:34:38 ok 19:34:40 I have no clue 19:34:51 so I'm open to suggestoins 19:35:44 Do I have to go get the SPROcc package installed on the solaris box? 19:35:50 ola that seems to count the number of times someone tries to vote. 19:35:53 gcc should work ok 19:35:56 how is it failing? 19:36:16 it's telling me it doesn't know what -t is 19:36:27 "it"? 19:37:14 $ make 19:37:14 gcc -mt -D_XPG4_EXTENDED -Dsun -D__unix -I../include -I./ -g -pta -o address.o -c address.cpp 19:37:14 gcc: unrecognized option `-pta' 19:37:14 cc1plus: Invalid option `t' 19:37:14 make: *** [address.o] Error 1 19:37:46 ok. now this: FIX ME ROWNUM ??? 19:38:30 * ola searches the openacs forum 19:39:19 guess you're out of luck 19:39:32 ola see this: set folder_id [db_exec_plsql create_new_folder { 19:39:32 begin 19:39:32 :1 := static_page.new_folder ( 19:39:32 name => :directory, 19:39:32 label => :directory, 19:39:32 try seeing if there are any porting note out on the net to gcc for whatever software package this is 19:39:35 parent_id => :parent_folder_id, 19:39:35 description => 'Static pages folder' 19:39:36 ); 19:39:38 end; 19:39:40 }] 19:39:56 it sets folder_id to the value returned by the pl/sql function. 19:40:08 Mark any ideas? 19:40:26 Yours is just throwing away the result. 19:40:38 the flags its passing aren't compatible with gcc 19:40:48 so either get the sun for-pay compiler, or troll the net for porting advice 19:41:13 You mean go ask my boss to install the package that has cc in it? 19:41:18 hmmmmm 19:41:45 yeah 19:42:07 ok 19:42:08 if it's a popular piece of software, there may be a gcc port out there already 19:42:11 brb 19:43:01 davb: yeah. it should return the total vote sum. 19:44:27 maybe it's someting strange ybos did. 19:44:34 * ola ducks 19:45:12 * markd2 quacks 20:05:01 ugh 20:05:07 he didn't know either 20:05:55 must find out what -t does in CC 20:06:06 what software is this? 20:06:18 snmp++ 20:06:23 I need it for nssnmp 20:07:40 Psychephylax: is that the AOLserver module? 20:08:51 yeah 20:09:16 Trying to integrate it with your package huh? 20:09:45 yes 20:10:29 hmmm.. good luck. 20:10:40 heh 20:10:45 my friend jsut told me what -t does 20:10:59 cool. 20:11:05 whazzit dooz? 20:11:21 * davb has to search for information in the ........ PHONE BOOK!!! 20:11:57 it prints out the files as the linker processes them 20:12:02 But it still does not solve my problem 20:12:22 $ make 20:12:22 gcc -mt -D_XPG4_EXTENDED -Dsun -D__unix -I../include -I./ -I../../libdes -g -pa -o address.o -c address.cpp 20:12:22 gcc: unrecognized option `-pa' 20:12:22 cc1plus: Invalid option `t' 20:12:22 address.cpp: In method `int IpAddress::parse_dotted_ipstring(const char *)': 20:12:23 address.cpp:446: implicit declaration of function `int strtok_r(...)' 20:12:25 address.cpp:446: assignment to `char *' from `int' lacks a cast 20:12:27 address.cpp:471: assignment to `char *' from `int' lacks a cast 20:12:29 make: *** [address.o] Error 1 20:12:31 $ 20:12:35 if i used Java I probably would have finished 20:13:43 so go use java 20:14:55 yeah good luck convincing my boss to switch 20:15:29 i guess I'll post on the bboard to that guy asking him what the hell he did to compile it 20:15:59 maybe he'll send you a binary if you give him a cookie 20:16:22 maybe 20:17:15 e-mail or post? 20:18:29 email 20:18:37 that way he won't be besieged by request for the binaries 20:19:00 heh 20:19:10 Well, I think he's russian I can encode my post in russian 20:19:11 muahaha 20:26:49 how's this: 20:26:52 Hello, 20:26:52 This is regarding your post on the OpenACS BB and your SNMP module for AOLServer. 20:26:52 I have started trying to set up AOLServer so that I can do the SNMP part of my project, however, I am encountering some problems along the way. First question I have is regarding the version of SNMP++ you used for your module. I have tried compiling both 2.8 and 3.1 to no avail. Also, what operating system did you use to compile? I'm trying to use SunOS 5.7 and it does not have CC installed so I am trying to tell the makefiles to use gcc 20:27:15 ugh 20:27:18 stupid client 20:27:32 looks good to me 20:30:40 oh 20:30:43 ok. done with the first round 20:30:46 forgot to ask him about the binder.c patch 20:30:51 Any of you heard of it? 20:34:10 ola: cool 20:34:24 Psychephylax: you can use Java inside AOLserver 20:35:57 ? 20:36:07 no, I would write a servlet 20:36:17 OK. never mind then. 20:36:18 Eh 20:36:25 I'd figure out a way to get it to work 20:36:48 I'm extra lousy dealing with RIGHT JOIN... 20:37:07 Opposite of LEFT :) 20:37:12 heh 20:37:43 I have no idea how to figure out which is which. 20:38:00 the bad news is that I don't exactly understand these outer joins. 20:38:01 one is when you're taking stuff out of the RIGHT table 20:38:05 and the other one is the left table 20:38:09 afaik 20:38:47 whats this. 20:39:08 a page saying...MP3 Jukebox.. 20:39:15 :-) 20:39:38 porting is fun. 20:40:07 Java Servlets can be run in aolserver using the nstomcat module, see nstomcat 20:40:20 http://www.cs.pub.ro/~gaburici/nstomcat/ 20:40:20 C: http://www.cs.pub.ro/~gaburici/nstomcat/ from davb 20:40:25 C:|NS Tomcat 20:40:25 titled item C 20:40:34 C: Java Servlets in AOLserver 20:40:35 commented item C 20:40:48 yeah 20:41:04 ola: what query? 20:41:06 but that means I'd have to rewrite the whole user interface and the data model and so forth and so on etc 20:41:39 Oh yeah. I thought you were think just about the SNMP stuff. 20:43:10 no 20:43:18 I would have done the whole thing in Java from scratch 20:43:35 I have some problem with ROWNUM in this query: 20:43:51 FIX ME ROWNUM 20:43:51 select title, mp3_id, creation_date 20:43:51 from (select title, mp3_id, creation_date 20:43:51 from mp3_mp3s_not_deleted, acs_objects 20:43:51 where mp3_id = object_id 20:43:51 and creation_date > current_timestamp - 1 20:43:53 order by creation_date desc) 20:43:55 where rownum < 10 20:44:32 I know somewhere this is something about porting ROWNUM. 20:46:06 Dan Wickstrom talks about using LIMIT instead of ROWNUM...? 20:46:10 I think you can use LIMIT 10 20:46:30 I'll try. 20:47:46 abbaJ has quit 20:50:23 davb: do you mean; where LIMIT 10 ? 20:51:27 I think you can skip the WHERE 20:51:36 rownum is a fake column. 20:51:55 aha. 20:52:41 http://openacs.org/bboard/q-and-a-fetch-msg.tcl?msg_id=0000XF&topic_id=11&topic=OpenACS 20:52:41 D: http://openacs.org/bboard/q-and-a-fetch-msg.tcl?msg_id=0000XF&topic_id=11&topic=OpenACS from davb 20:52:56 D:|Porting ROWRUM from Oracle to Postgresql thread on Openacs.Org 20:52:56 titled item D 20:56:58 cooooool! dave, it works. 21:00:16 great! 21:03:46 * Psychephylax twiddles thumbs 21:04:31 i'm about to start looking at porn here 21:05:04 [06/Oct/2001:01:06:22][2711.5125][-conn1-] Error: dbinit: error(localhost::openacs4,ERROR: Function 'mp3_playlist__new(unknown, unknown, unknown, unknown, unknown, timestamp, unknown)' does not exist 21:05:04 Unable to identify a function that satisfies the given argument types 21:05:04 You may need to add explicit typecasts 21:05:04 ): ' 21:05:04 select mp3_playlist__new( 21:05:05 '7135',-- playlist_id 21:05:07 'Ola',-- name 21:05:09 '-3',-- remove_threshold 21:05:11 '127.0.0.1',-- creation_ip 21:05:13 '2215',-- creation_user 21:05:15 current_timestamp,-- creation_date 21:05:17 '7128'-- context_id 21:05:19 ); 21:05:36 hmm.. 21:06:27 Is that one missing a couple parameters? 21:08:33 there should be nine parameters. 21:10:58 how come there are missing parans in the query? 21:11:14 can oracle cope with that? 21:11:45 s/parans/params 21:13:58 defaults. 21:14:03 ( i think ) 21:15:48 moo 21:18:20 * rbm 's brother is on a plane back to Brazil :-( 21:19:06 question: can I use current_timestamp in place of now() ? 21:19:44 * ola hopes the best for rbm's bro 21:22:40 ola: In 7.1, yes. In fact, you should. 21:23:29 aha. why? :-) 21:24:21 Because it's part of SQL3. 21:24:37 (the standard). Hopefully other products will also adopt it soon. 21:25:23 so now() is postgres specific, right? 21:25:48 yes. 21:25:59 Just like sysdate is to Oracle. 21:27:42 thanks, rbm. 21:28:43 np 21:31:21 i smell popcorn 21:32:02 buttery? 21:32:22 very 21:32:30 I think that's what's making it smell really tasty 21:33:16 this is the exact error I'm getting: http://www.katz.dusnet.de/agent++Archiv/0107/0071.html 21:33:27 Mark, you awake? 21:33:33 * Psychephylax gives Mark a cookie 21:33:36 barely 21:33:39 hehe 21:33:45 Take a look at that 21:34:12 It's a solution Makefile but I don't think it will work for be....because....it's for x86 and uses gmake 21:34:16 looks good 21:34:29 i don't have gmake 21:34:29 gmake is easy to compile 21:34:38 and solarisx86 and solaris are nearly identical 21:34:45 ok 21:34:49 where can i get gmake? 21:34:49 the port of aolserver 2.3.3 to solaris x86 was literally a recompile 21:34:55 www.gnu.org 21:35:44 * Psychephylax runs a find on the box to make sure it's definetly not installed 21:38:11 abbaJ has joined #openacs 21:40:11 * ola gotta go 21:40:17 bye ola 21:40:34 ola has quit 22:03:11 * markd2 puts his head back in the backup & recovery book 22:03:14 markd2 has quit 23:10:43 docwolf has joined #openacs 23:10:57 yo 23:26:44 hey Adam! 23:26:49 what's up? 23:27:23 moo anerd 23:27:27 roehuoeur 23:42:09 hi 23:42:32 anyone here a prize winner? 23:43:12 AaronSw is 23:43:23 Rats! mark is not here anymore 23:44:10 d'oh 23:44:42 i heard some horrible shit 23:44:51 about their server 23:47:13 right.. i've heard from a couple of people 23:47:17 docwolf: got Oracle in the machine you're in? 23:47:20 that aD is taking them back 23:47:32 nope. i'm oracle-free :-) 23:47:50 hmm 23:50:01 i just wanted to confirm this 23:50:09 so i could perhaps convey to the right people within the company 23:50:18 that it's probably not a nice thing to do 23:50:31 (hey, i'm a VC, and even I know _that_ .. ) 23:51:28 but.. i'm not sure that it's even true. 23:51:30 so we'll see. 23:51:53 heheh 23:52:05 It would be a pretty low thing to do that. 23:52:10 yeah