00:15:43 talli has quit 00:42:38 davb has joined #openacs 00:54:28 docwolf has joined #openacs 01:16:23 talli has joined #openacs 01:16:35 hey guys 01:25:45 just can't stay away huh? 01:30:10 nope - i'm a loser 01:30:20 although i'm walking away from my computer now... 01:30:24 i am leaving... 01:30:26 see ya... 01:30:30 talli has quit 01:32:50 bye 01:40:59 mio has joined #openacs 01:42:32 mio has left #openacs 02:37:26 til has quit 03:09:29 docwolf has quit 03:25:45 docwolf has joined #openacs 03:41:45 davb has quit 04:29:05 hazmat has quit 04:35:04 blue has joined #openacs 04:36:23 blue is now known as andrew 04:59:11 hazmat has joined #openacs 05:54:03 [GlobalNotice] T-30 seconds to services implosion 05:59:58 ChanServ has changed the topic to: Welcome to the OpenACS Helpline | http://www.openacs.org/ 05:59:58 This channel is logged: http://www.blogspace.com/openacs/chatlogs/ and blogged: http://www.thedesignexperience.org/openacs/ircblog 07:55:17 all openprojects staff, please message me for an invite to #willyoutakethisdance... perpare for liftoff! 08:46:57 [GlobalNotice] This is just a warning that in a few minutes we will start to convert the network to dancer-ircd. 09:08:07 Disconnected from irc.openprojects.net (Connection reset by peer) 09:09:30 loggy has joined #openacs 09:09:30 topic is: Welcome to the OpenACS Helpline | http://www.openacs.org/ 09:09:30 Users on #openacs: loggy andrew docwolf rbm shagster chump GEM AaronSw 09:09:30 This channel is logged: http://www.blogspace.com/openacs/chatlogs/ and blogged: http://www.thedesignexperience.org/openacs/ircblog 09:12:42 rbm has quit 09:12:42 andrew has quit 09:12:42 GEM has quit 09:12:42 AaronSw has quit 09:12:49 GEM has joined #openacs 09:12:55 rbm has joined #openacs 09:14:00 Disconnected from irc.openprojects.net (Connection reset by peer) 09:15:56 Disconnected from irc.openprojects.net (ERROR :Closing Link: loggy[~openacs-l@205.253.219.118] (Ping timeout: 182 seconds)) 09:16:17 loggy has joined #openacs 09:16:17 Users on #openacs: loggy AaronSw chump 09:16:17 month or so. If we have forgotten something, please inform us 09:16:17 soon (due to database drift) especially if important. 09:16:17  09:16:17 For conversion matters please use #Conversion. 09:16:17  09:16:17 We hope you will have a great stay with us! 09:16:17 (Open Projects Net Team) 09:16:17 [#openacs] This channel is logged: http://www.blogspace.com/openacs/chatlogs/ and blogged: http://www.thedesignexperience.org/openacs/ircblog 09:16:19 blue has joined #openacs 09:17:04 shagster has joined #openacs 09:17:37 blue has quit 09:17:37 AaronSw has quit 09:18:14 rbm has joined #openacs 09:19:00 blueseed has joined #openacs 09:19:34 rbm has quit 09:19:35 rbm_ has joined #openacs 09:21:12 AaronSw has joined #openacs 09:25:18 Welcome to the new Open Projects network. We hope you enjoy 09:25:18 the new quality of service that should be provided. If you 09:25:18 have any questions / comments / concerns, please email 09:25:18 support@openprojects.net or drop by #OpenProjects. 09:25:18  09:25:18 Please note that a few items in the services database have not 09:25:18 been converted into the new database. This includes: 09:25:18 - channel access lists (different complex trust metrics) 09:25:18 - memos sent to channels (time reasons) 09:25:18 - items which have no equivalent in the new database 09:25:18 - things which we have forgotten ;) 09:25:18 If there is any information which has not been converted, 09:25:18 it will be available on a yet unnamed server for the next 09:25:18 month or so. If we have forgotten something, please inform us 09:25:18 soon (due to database drift) especially if important. 09:25:18  09:25:18 For conversion matters please use #Conversion. 09:25:18  09:25:18 We hope you will have a great stay with us! 09:25:18 (Open Projects Net Team) 09:25:18 [#openacs] This channel is logged: http://www.blogspace.com/openacs/chatlogs/ and blogged: http://www.thedesignexperience.org/openacs/ircblog 09:28:43 rbm_ has quit 09:28:43 shagster has quit 09:28:48 rbm_ has joined #openacs 09:28:53 shagster has joined #openacs 09:38:47 graszew has joined #openacs 09:50:10 hazmat has joined #openacs 10:02:33 rbm_ has quit 10:02:33 AaronSw has quit 10:02:33 chump has quit 10:03:39 rbm_ has joined #openacs 10:03:56 AaronSw has joined #openacs 10:05:36 chump has joined #openacs 10:11:53 hazmat has left #openacs 10:11:56 hazmat has joined #openacs 10:16:05 hazmat has quit 10:22:13 Disconnected from irc.openprojects.net (Connection reset by peer) 10:22:30 loggy has joined #openacs 10:22:30 Users on #openacs: loggy AaronSw @rbm_ 10:22:43 Disconnected from irc.openprojects.net (Connection reset by peer) 10:22:56 loggy has joined #openacs 10:22:56 Users on #openacs: loggy AaronSw 10:23:19 chump has joined #openacs 10:25:15 chump has quit 10:25:15 AaronSw has quit 10:26:16 chump has joined #openacs 10:27:44 graszew has joined #openacs 10:29:25 Disconnected from irc.openprojects.net (Connection reset by peer) 10:29:33 loggy has joined #openacs 10:29:33 Users on #openacs: loggy graszew chump 10:29:33 [#openacs] This channel is logged: http://www.blogspace.com/openacs/chatlogs/ and blogged: http://www.thedesignexperience.org/openacs/ircblog 10:34:39 AaronSw has joined #openacs 10:46:06 ola has joined #openacs 11:28:54 [UrgentNotice] 11:33:43 :[UrgentNotice] Sorry. Please can you report servers not listening on usual ports ie 6665-7000 to me immediately. 11:36:33 ola has quit 11:36:38 ola has joined #openacs 11:37:08 graszew has quit 11:45:11 :[GlobalSpam] "We have normality. I repeat, we have normality. Anything you still can't cope with is therefore your own problem." But if you do have problems please come to #conversion where your problems will be solved. Thankyou for bearing with us during these difficult times :) 11:52:43 graszew has joined #openacs 12:24:06 Disconnected from irc.openprojects.net (Connection reset by peer) 12:24:21 loggy has joined #openacs 12:24:21 Users on #openacs: loggy graszew 12:26:38 AaronSw has joined #openacs 12:27:08 chump has joined #openacs 12:36:41 til has joined #openacs 12:50:51 graszew has quit 13:21:28 til has quit 13:27:03 [GlobalNotice] Hi all. We're slowly struggling into the new code. If you're having problems, please message me a short detail message describing the problem. That should get us going for now. Thanks. 13:31:20 [GlobalNotice] Hi all. I've gotten surprisingly few messages so far. As other people have mentioned previously, we are aware there are problems. Keep letting us know your needs, we'll respond as quickly as we can. Thanks. 13:34:46 Disconnected from irc.openprojects.net (Connection reset by peer) 13:34:50 loggy has joined #openacs 13:34:50 Users on #openacs: loggy AaronSw chump 13:34:50 [#openacs] This channel is logged: http://www.blogspace.com/openacs/chatlogs/ and blogged: http://www.thedesignexperience.org/openacs/ircblog 13:38:48 [GlobalNotice] Hi all. Apologies for the global notice. You'll probably see a few more. Be aware that I may not have replied to several messages, but I'm logging and intend to hit each one in turn. Thanks for your patience during the cutover. 13:42:08 Disconnected from irc.openprojects.net (Connection reset by peer) 13:43:56 Disconnected from irc.openprojects.net (ERROR :Closing Link: loggy[~openacs-l@205.253.219.118] (Ping timeout: 181 seconds)) 13:44:01 loggy_1 has joined #openacs 13:44:01 Users on #openacs: loggy_1 @AaronSw 13:49:10 chump has joined #openacs 13:50:11 Welcome to the new Open Projects network. We hope you enjoy 13:50:11 the new quality of service that should be provided. If you 13:50:11 have any questions / comments / concerns, please email 13:50:11 support@openprojects.net or drop by #OpenProjects. 13:50:11 [#openacs] This channel is logged: http://www.blogspace.com/openacs/chatlogs/ and blogged: http://www.thedesignexperience.org/openacs/ircblog 13:50:34 talli has joined #openacs 13:50:35 whoa. all alone. no one here to love but me. 13:54:47 [GlobalNotice] Hi all. In a few moments we will be reverting to the old servers and stabilize there while we analyze the various issues produced by our conversion attempt. So expect the usual (relatively) stable environment in a few minutes. Please bear with us and thanks again for your patience. 13:56:46 davb has joined #openacs 13:56:59 Good morning. 14:01:38 davb has quit 14:01:38 talli has quit 14:01:38 chump has quit 14:01:38 AaronSw has quit 14:02:05 Disconnected from irc.openprojects.net (Connection reset by peer) 14:06:27 loggy has joined #openacs 14:06:27 Users on #openacs: loggy AaronSw davb talli 14:06:27 This channel is logged: http://www.blogspace.com/openacs/chatlogs/ and blogged: http://www.thedesignexperience.org/openacs/ircblog 14:06:36 brb 14:07:16 chump has joined #openacs 14:07:29 nevermind. 14:07:46 go loggy! 14:13:07 AaronSw is now known as Aaron 14:13:15 Aaron is now known as Aaron2 14:13:53 * Aaron2 disconnects 14:13:56 * Aaron2 reconnects 14:15:08 Aaron2 is now known as AaronSw 14:19:46 who is this AaronSW guy and why is he always here? 14:20:00 You don't know AaronSw? 14:20:08 I'm a secret spy from Zope. 14:20:10 hey aaron, isn't it time for school? 14:20:27 Ooh, that gets you a smack. 14:20:29 aren't you late for the bus? 14:20:30 * AaronSw slaps talli around a bit with a large trout 14:20:39 train 14:20:51 whatever. get your ass to class, boy! 14:21:25 You east-coasters... I've got a whole 'nother hour still. 14:21:34 oh... right 14:21:36 good point 14:21:54 Heh. 14:21:56 BTW, I'll be in Washington D.C. next week, in case anybody is around there. 14:21:58 well, don't you have any homework to do in this time? 14:22:14 what for? meeting with some alQaeda operative? 14:22:50 http://dabbler.org/stories/storyReader$88 14:22:52 A: http://dabbler.org/stories/storyReader$88 from davb 14:22:59 Homework? Why bother 14:23:03 A:| .NET Signals an Industry Shift 14:23:05 titled item A 14:23:05 DC: P2PCon2 14:23:06 Label DC not found. 14:23:26 AaronSw: do you go to all those things? 14:23:35 all what things? 14:24:28 You seems like you are visiting quite a few conferences. Well more than me anyway :) 14:25:24 Really? I can only think of like 2. 14:25:43 OK, I jsut have no life...:) sorry. 14:25:44 what about parent-teacher-student conferences? 14:25:54 been to many of those lately, aaronsw? 14:26:03 Not for years. 14:26:09 How about you? 14:26:09 truant! 14:26:19 double truant! 14:29:06 A:[OS Independence|http://dabbler.org/stories/storyReader$22] 14:29:07 commented item A 14:29:39 A:[Open Pragmatism|http://dabbler.org/stories/storyReader$23] 14:29:39 commented item A 14:30:22 i hear that C# is an awesome language 14:30:29 much better than java, which isn't saying a whole lot 14:30:40 but C# apaprently is supposed to hold a lot of promise 14:30:43 Interesting. 14:31:13 But will it run on non-MS systems? 14:31:17 i've heard that from many people, even within the open source community (including luke and lars) 14:31:51 i believe so. there were always rumors that the beta versions of some of the .NET IDEs had Linux in their platform choices 14:32:05 also, Mono and dotGNU are trying to build open source CLRs 14:32:22 Right. Did they say what it was good for? 14:32:41 Is it a replacement for C/C++ in other words? 14:34:23 chump has quit 14:34:40 Looks like I still have a memory leak on that server... 14:35:34 it is more of a replacement of java, which i guess is a replacement of C/C++ 14:35:53 chump has joined #openacs 14:35:55 then again, C/C++ still have their considerable place. witness all the work we still do in C to extend AOLserver, 14:36:10 which is still the best fucking web software anyone has written 14:36:12 IMHO 14:36:22 For some things, yes. But for most applications, a nice scripting language is better. FOr low-level stuff, C is more efficient. 14:36:38 on the web, i agree 14:36:59 but all the heavy duty stuff on the server side in aolserver is in C, right? 14:37:10 Right. 14:37:46 C is for the hard stuff. Everything else should be in a high-level language. Most applications can be programmed this way. 14:38:05 right, i agree 14:38:28 I am going to work on the new keyword stuff today. 14:38:36 killer! 14:38:46 What do you think about fixing bboard to use the CR keywords? 14:38:49 that stuff will be awesome. it will be a great addition to the toolkit 14:38:53 i think that would be amazing 14:40:23 ola has joined #openacs 14:41:15 Hi all! 14:41:43 Hi ola. 14:42:13 hey ola 14:44:16 brb-breakfast 15:05:25 chump has quit 15:10:23 chump has joined #openacs 15:17:55 markd2 has joined #openacs 15:22:10 Hi markd2 15:22:27 greetings 15:22:28 * markd2 is back home, and slogging through the work mailbox 15:29:32 yo markd 15:31:47 hiya 15:37:01 Anyone know how I would create a package with no SQL files? 15:39:11 aha, nevermind. 15:39:40 Psychephylax has joined #openacs 15:41:51 hi people 15:41:54 Howdy 15:41:58 hey mark 15:42:15 hey Sporkephylax 15:43:03 I need Oracle help =) 15:47:57 sure 15:47:58 whazzup? 15:48:40 I can't insert stuff into oracle 15:48:40 what happens? 15:48:40 I get a number to string error or sumfing 15:52:40 hold on..I need to get my butt in gear here 15:52:50 multitasking and daylight savings times don't mix 15:53:33 Should I create seperate UI for site-administrators and users to assign keywords to items? 15:53:43 til has joined #openacs 15:58:07 Initialization error: X server not responding 15:58:11 Why is that not working!!! 15:58:55 Oracle doesn't use an X server for insert operations... 15:59:23 :D 16:00:08 ora8.c:3568:ora_tcl_command: error in `OCIStmtExecute ()': ORA-06502: PL/SQL: numeric or value error: character to number conversion error 16:00:08 ORA-06512: at line 1 16:02:14 try the query in SQL*PLUS 16:02:30 it's not a query it's an insert 16:02:42 ok. try the insert... 16:03:38 how do I run a stored function ? 16:05:21 Try select stored.functionname(params); ... 16:05:55 or begin(CR) stored.functionname(args); end; / 16:06:03 spiffy! 16:06:26 CR? 16:06:58 <-ENTER 16:07:01 oh 16:07:05 :) 16:07:51 * Psychephylax tries 16:09:34 if I have defaults for a few of them then I don't need to specify them? 16:10:14 right 16:10:21 aha 16:10:22 Well, wait 16:10:30 say I have something like this 16:10:55 pl/sql proc that takes (sysname, query, sysuptime, syscontact) 16:11:02 where query is sysdate 16:11:17 how will it know to insert that when I have this: 16:11:47 ('OracleBox','10 days','me@oraclebox.com'); 16:11:59 wouldn't it try to insert 10 days as the time? 16:12:09 as sysdate rather 16:16:27 anyone? 16:16:55 talli has quit 16:17:23 What are you trying to accomplish? 16:19:07 i think the problem arrived when I omitted a value 16:19:20 causing oracle think it's inserting something else 16:19:32 everything to shifted by 1 or however many omissions i made 16:19:39 makes sense to me...but I'm no oracle genius 16:19:47 * Psychephylax waves a cookie around 16:20:09 yeah 16:20:54 so how would I signify I want to use a default in the space? 16:20:57 just go , ,? 16:21:54 don't think you can do that 16:22:05 ok 16:22:06 it'll fill in defaults after it uses the existing arguments 16:22:09 you can name them 16:22:20 explain 16:22:26 e.g. somefunction (arg1 => 'value', arg17 => 'value2', give_mark => 'more cookies'); 16:22:33 it'll pass those values for those arguments 16:22:37 and then everything else will take its default value 16:22:38 right 16:22:49 ok this is what I have: 16:23:01 set incident_id [db_exec_plsql insert_info "begin :1 := snmp_incident.new(mac_address => :modem_mac,comments => :comments, 16:23:27 before mac_address I have a incident_id which is calculated from acs_objects 16:23:36 and between modem_mac and comments there's a sysdate 16:28:12 if you don't have any unlabeled paramters, it should then be doing the right thing with the default valus 16:28:30 very odd 16:29:38 then add one 16:29:40 then it becomes very even 16:29:56 i will just fill in the values that don't have defaults 16:31:19 aha 16:32:32 PLS-00306: wrong number or types of arguments in call to 'NEW' 16:33:56 double-check your types 16:34:02 that your'e not passing a character string where a number is expected 16:34:41 I checked 16:34:51 I made it so everything that has a default value is using it 16:34:56 ANd it seems to work 16:34:58 but : 16:35:06 ora8.c:3568:ora_tcl_command: error in `OCIStmtExecute ()': ORA-02291: integrity constraint (MYOPENACS.ACS_OBJECTS_OBJECT_TYPE_FK) violated - parent key not found 16:35:51 that means you're trying to insert a value into a 'references' column, and ther's not a correspnding value in the referenced table 16:36:08 judging from the name of the constraint (ACS_BJECTS_OBJECT_TYPE_FK), you probably don't have a correspnding object type 16:36:16 hmm? 16:37:50 You are trying to create an object of type 'foo' when there is no entry in acs_object_types for that type. 16:39:22 i just did select object_name from user_objects; 16:39:41 and I have snmp_incident x2 and snmp_incidents 16:40:27 blueseed has joined #openacs 16:40:41 I don;t even have a user_objects table. 16:41:28 acs_objects 16:41:37 user_objects is part of the oracle data dictionary 16:42:12 Thats the objects name. Not its type. You need to create a type if you are using a custom type instead of cr_item. 16:42:17 or whatever/ 16:42:28 ummm 16:42:31 i'm confused 16:42:41 what tables are the ones I am going to be playing with 16:42:56 Do you have in your SQL file acs_object.create_type (my type) 16:43:05 let me check 16:43:33 acs_object_type.create_type ( 16:43:33 supertype => 'acs_object', 16:43:33 object_type => 'snmp_incident', 16:43:33 pretty_name => 'SNMP Incident', 16:43:33 pretty_plural => 'SNMP Incidents', 16:43:33 table_name => 'snmp_incidents', 16:43:35 id_column => 'incident_id' 16:43:37 ); 16:43:48 Ok. good. 16:44:04 make sure its there select object_type from acs_object_types; 16:44:05 SQL> select * from acs_objects where object_type = 'snmp_incident'; 16:44:05 no rows selected 16:44:16 ok. 16:44:25 am I missing something?> 16:44:42 The log of when you ran the create script? 16:44:46 :) 16:44:48 ummm 16:44:54 lemme check 16:45:01 Try creating it now maybe... 16:45:19 nope 16:45:43 umm 16:45:46 spool something 16:49:11 that did nothing 16:49:24 no errors? 16:49:53 no 16:51:16 you want the log? :) 16:53:28 Not very verbose is it. 16:53:35 no 16:54:00 What if you try just that acs_object.create_type in SQL*PLUS? 16:55:28 markd2 has quit 16:56:32 it works 16:56:35 it does create my stuff 16:56:37 or so it seems 16:57:42 ok. What is the exact acs_object.new call you are making? 16:58:33 it's very long :) 16:58:38 I'm not pasting it 16:59:32 ok. Check to make sure that you are setting object_type in there to the object type you have created. and make sure that is what is really getting passed to oracle. 17:00:01 aha 17:00:03 wait 17:00:07 I think I found a problem 17:00:29 heh, as I suspected. You should not work before lunchtime on Mondays. :) 17:02:01 heh 17:02:09 that eliminates 10% of my workweek 17:03:05 Ok, maybe not... 17:06:51 i see nothing 17:07:15 darn. 17:09:33 graszew has joined #openacs 17:09:39 how do I do a plain sql block 17:09:47 I'm going to try manually inserting into the thing 17:10:17 markd2 has joined #openacs 17:10:20 geeze 17:10:23 just like this 17:10:23 lol 17:10:26 varley netsplit off, and never rejoined 17:10:31 insert into blah 17:10:41 object id, 17:10:42 etc, 17:10:42 don't I need do do ns_? 17:10:45 etc 17:10:51 oh inside AOLserver. 17:10:52 yes/ 17:10:54 db_dml 17:11:03 I think. 17:15:55 umm 17:16:16 you are correct 17:16:21 but when I use it I get an error 17:16:33 nm 17:17:36 ora8.c:3930:ora_tcl_command: error in `OCIStmtExecute ()': ORA-01400: cannot insert NULL into ("MYOPENACS"."SNMP_INCIDENTS"."INCIDENT_ID") 17:17:47 Oh right. do 17:17:52 so stop inserting null for incident_id :-) 17:18:00 nextval(acs_object_id_seq) 17:18:01 it gets calculated by openacs object! 17:18:07 oh wait...it used to 17:18:27 This is not MS Access. 17:18:33 :) 17:18:36 thank ghod 17:19:25 insert into snmp_incidents(incident_id, mac_address, problem_id) values (nextval(acs_object_id_seq),'00:E0:6F:01:54:B2', \ 17:19:32 it didn't like my nextval 17:19:47 I am wrong 17:19:52 nextval.sequencename 17:19:55 duh 17:20:00 * Psychephylax gives Mark Dave's cookie 17:20:09 argh! not again! 17:20:17 * davb goes to study his oracle again... 17:20:44 sequencename.nextval 17:20:57 Ok. at least I was close the second time. 17:21:03 column not allowed here 17:21:08 lol 17:21:09 ok 17:21:26 third time is the cahrm 17:21:38 Well first time with the correct information :) 17:23:21 heh 17:24:02 Will this get expanded into the proper values: 17:24:04 '$modem_ip', '$cmts_ip', '$cmts_dns' 17:25:27 it should. 17:26:35 graszew has quit 17:28:22 ola: were you around yesterday? Static-pages is serachable, at least in postgresql. 17:35:35 hazmat has joined #openacs 17:35:46 hello 17:35:54 hi hazmat 17:36:09 hi davb 17:36:54 i spent my last night tracking down dependencies for the search stuff, we got get a unified distro with all this stuff included :) 17:37:10 That is a good idea. 17:37:29 My main problem is debian puts everything in a different spot. 17:37:50 hmm.. so you're installing via apt, does apt include all the nesc headers? 17:38:07 Yes. What it won;t do is compile intarray. 17:38:28 it just installs the source for intarray in /usr/lib/postgresql/lib/contrib 17:40:49 i compiled it all from source (minus tcl), it wasn't so bad, except for getting hold of everything. 17:40:49 which is useless because it won't compile without the postgresql source. Which I could not find as a package. 17:40:49 It is much easier to just compile postgresql. 17:40:49 so to do a quick recap openacs4 requires like 10 different packages, ns_xml (depends on libxml), ns_sha1, ns_cache, aolserver, postgresql, intarry, openfts, nstcl, libpgtcl. i'm probably missing something :) 17:40:49 oh yeah and the pg-driver 17:40:49 Disconnected from irc.openprojects.net (ERROR :Closing Link: loggy[xcdfddb76.ip.ggn.net] by carter.openprojects.net (Ping timeout for loggy[xcdfddb76.ip.ggn.net])) 17:44:10 loggy has joined #openacs 17:44:10 Users on #openacs: loggy k2pts hazmat markd2 blueseed til Psychephylax chump ola AaronSw davb 17:44:13 This channel is logged: http://www.blogspace.com/openacs/chatlogs/ and blogged: http://www.thedesignexperience.org/openacs/ircblog 17:44:18 docwolf has joined #openacs 17:44:23 hi docwolf 17:44:24 Hi docwolf 17:44:30 hi, everybody! 17:44:41 Ok. I need some advice here on building a new keywords admin/user package. 17:44:58 Do I need to create permissions like keyword_create etc...? 17:45:18 davb: i'm confused. what are you trying to do? 17:45:35 So that an item, say a bboard posting or something can be set to allow visitors to assign categories to it. 17:45:42 or just the original author. 17:46:16 k2pts: OK. I am building a general package to offer a UI to manage keywords. To add keywords to the list of keywords and to assign them to items in the CR. 17:46:31 i thought that keywords were already implemented in the CR 17:46:36 ok 17:46:45 ok: ui on top of the CR API 17:46:47 right? 17:46:51 That is correct. 17:47:08 CMS has one, but this would be general so any package could take advantage of it. Like general-comments. 17:47:15 hello 17:47:28 hi hazmat. 17:47:34 hazmat: wf is nearly done 17:47:51 i think i installed a bunch of extra stuff for, openfts oh well. wf is nearly done. awesome! are you going to include graphviz? 17:48:14 you mean in the package? 17:48:19 yes 17:48:33 i have no idea what the license is... 17:48:39 probably not, I am not sure that the license is combatible 17:49:26 bummer, it def. add some visual appeal. is anyone familiar with request processor internals? 17:49:42 Ok, here is my other question. My package will have pages like keyword-add keyword-remove etc. 17:50:29 I want to allow other packages to be able to link to those pages with the proper variables to use them. Should I make a seperate proc for each like keyword_add_link, keyword_remove_link? 17:50:55 davb: first give some thought to the API. portlets would be a better idea 17:51:14 hazmat: go ahead. i'll see what i can do 17:51:18 Hmmm. Ok. I will have to look into that. 17:51:50 davb: i meant the dotlrn stuff, but if you give some thought to the API it will be easier in the long run 17:52:34 Aha. Right. 17:53:36 the general idea is to think of you functions as datasources (providing info for the portlet) 17:54:09 and think of the portlet like a widget that you will call to be rendered within the other package 17:54:31 i've been working on hooking up python to a running openacs installation, after having it working well with standalone aolserver, i think the problem is the request processor is messing up the handling, so i'm trying to figure out the differences between the various styles of procs/filters compared to a map directive in an aolserver config file, my understanding of the rp, was that it was a single filter/proc that allowed regist 17:55:51 ok let me see 17:57:01 ok what is the problem exactly: a) not parsing (python) the pages correctly b) not finding them at all 17:57:27 doh! 17:58:19 hazmat? 17:59:36 adler has joined #openacs 17:59:49 i'm investigating, more info in a sec. 17:59:57 greetings 18:00:28 IIRC, the RP registers some filters at startup and then uses those filters to map them into the right parser. 18:00:38 s/parser/handler 18:00:48 you will need to do something like this: rp_register_extension_handler $type $handler 18:01:05 this is done in: acs-tcl/tcl/request-processor-init.tcl 18:01:37 ack 18:03:08 hi adler 18:03:19 bbl - lunch 18:03:29 davb: enjoy 18:03:41 mmmm... nourishment 18:03:47 howdy k2 18:04:06 hey 18:05:00 ever work with acs-mail/messaging? 18:05:36 nope 18:06:28 I think that's vinod's bag. 18:06:37 yes me too 18:06:41 and jong 18:06:52 JonG 18:07:08 I'd love it if oacs could receive mail (from auction sites, pay pal, etc) 18:07:23 but I think queue_incoming remains unimplemented. 18:07:48 henry did some work on that (email-handler) 18:07:56 but I've never tried 18:08:23 I think it needs porting/fixing to work in oacs4 but i am not sure 18:08:25 is it in the alpha tarball? 18:08:31 let me check 18:08:54 i don't know if it's in the alpha tarball but you can get it from cvs 18:09:39 it seems ported but IIRC there were some extra stuff that needed some work 18:09:57 probably queue-message.pl and queue-message.sh 18:10:20 i think i stumbled across those at one point 18:10:45 email-handler big picture: 18:10:46 You can build a gateway for handling incoming email messages for your application using a Perl script called queue-message.pl. That Perl script will accept an 18:10:46 incoming email message from the mailer and insert its contents into a queue table in the database. A procedure can be scheduled to sweep the queue at some interval to 18:10:46 process the messages. 18:12:17 supposed to use procmail to trigger the perl script? 18:16:01 k2pts: is there anything I can read that will help me design the API to plan for future portlets? 18:17:58 I guess not. But, it would be nice to know what the acs-sc package does. the new portals package is based on that. Kapil wrote an excellent guide to acs-sc issues 18:18:39 you can find it in: acs-service-contract/www/doc/index.html 18:18:41 Ok. 18:20:21 davb, if you have any questions, let us know and i'll add them to the faq section of the doc.. 18:20:38 sure. 18:21:14 hazmat: Can you stick some tags around that so its not all monospaced? :) 18:22:14 k2pts: so should all packages that provide services like this use acs-sc? I was modeling this more after general-comments. which works, but I can't tell you how well designed it is. 18:22:16 my fault 18:22:37 well, i like the acs-sc way more :) 18:22:41 davb, sure the format, is structured text, i was going to convert it to docbook and html, but i ended up just giving it raw to k2pts, i'll convert it over now. 18:23:00 No problem really :) 18:24:50 k2pts: did you have URL for henry's work? 18:25:18 it's in the cvs 18:25:21 I think I will have to post in the bboard to solicit ideas on how this package will be used before I go casting an API in stone. 18:25:33 like you said :) 18:25:35 yes, that's a good idea 18:27:04 well, well, need to go off guys, later. 18:27:12 bye. 18:27:21 k2pts has left #openacs 22:52:59 loggy has joined #openacs 22:52:59 Users on #openacs: loggy rbm Psychephylax hazmat shagster blueseed AaronSw 22:53:03 This channel is logged: http://www.blogspace.com/openacs/chatlogs/ and blogged: http://www.thedesignexperience.org/openacs/ircblog 23:01:47 toddfx has joined #openacs 23:07:10 hi todd 23:07:14 welcome to #openacs 23:07:23 hello 23:09:05 how's the project coming? 23:24:29 talli has joined #openacs 23:24:37 hey guys 23:24:47 Make way! It's talli on the channel! 23:24:56 all hail talli! 23:25:10 wait, can i say that for myself? 23:25:22 * rbm twacks talli for self indulgence 23:25:29 ouch 23:25:37 i've gotten smacked around today on this channel 23:25:49 * rbm twacks talli again, for good measure 23:26:01 aaronsw hit my with a trout this morning for suggesting he get his ass to school. damn truant. 23:26:09 heheh 23:26:18 And I'll do it again too! 23:26:19 how's it going rbm? 23:26:22 I can see Aaron doing that with that kind of suggestion :) 23:26:43 talli: Trying to solve something in our legacy AOLserver app at work, so I can get out of here 23:26:56 fun! 23:27:17 hi guys 23:27:25 hi haz 23:27:29 hey hazmat 23:28:22 hey, i hooked up python to the acs-templating system :) 23:28:38 hazmat: COOL! 23:28:45 * rbm hugs python 23:28:46 whoa 23:28:50 dope 23:29:04 rbm: you prefer python to tcl? 23:29:15 I like Tcl too. 23:29:26 But Python seems to have a much better design than Tcl. 23:29:34 true enough 23:29:44 i spent *way* too much time today doing it, but i'm psyched... i like tcl sorta of but python allows be to do much larger programming projects in all sorts of settings. 23:29:49 It looks like executable pseudocode 23:30:35 I'm slowly digging my way through Programming 23:30:37 so hazmat, you can have full acs-templating functionality with python code? 23:30:43 I'm slowly digging my way through Programming Python, 2nd Ed. 23:31:04 yup... and lisp fans can make even use it in a functional style 23:31:23 whoa 23:31:28 that's dope 23:32:17 * rbm calls the DEA to go after talli 23:32:32 * talli calls the INS to go after rbm 23:32:43 talli: I'm already in the concentration camp. 23:32:50 bummer 23:32:57 they got you fenced in? 23:33:15 talli: yes its fully functionality, master templates, variable insertion etc. i need to refactor it a little, cause right now i have to have a tcl page do the nesc. upvars, but righting a wrapper function around the templating call will fix that. 23:33:29 I'm still waiting for something they should have completed in 2 weeks. It's been 10 weeks. 23:33:40 christ. 23:33:47 hazmat: that's totally awesome 23:35:12 so hazmat, i have a question... 23:35:46 given that object oriented code can have a serious problem when used in the context of a relational db 23:36:05 but given the fact that OOP can be real nice when doing a lot of coding... 23:36:40 does python provide a way of bridging this gap as it is an OOP scripting language? 23:40:15 talli: i'm not sure why oopl nesc. have problems with relational systems... i think thats a more interesting concern considering the natural mismatch of rich object data vs. a pure relational system and bridging them with some sort of relational mapping system. how does python handle it? in terms of pure relational work... 23:40:59 the relational mapping issue is the problem i am talking about. it's a real bitch to get right. i would prefer avoiding it all together 23:41:28 you can put in as little abstraction as you want conn = dbadaptor(connection_string) ; cursor = conn.cursor() ; cursor.execute("select now from dual"); return cursor.fetchmany() 23:42:29 i actually wrote a zope product that sucks up all the openacs query files and allows you to use them from zope via engine.execute('query_name')... iotw there is a lot of freedom in terms of possible answers. 23:43:14 cool 23:44:02 I read about DODS in enhydra (Java). It looked... slow. DODS had to write all your queries for you dinamically 23:44:11 talli: (just a statement of facts not bragging) the zope corporation guys where trying to have me do work on fitting an or mapping system underneath the zodb, so you could transparently access objects that were stored relationally. i don't really think that this way is terribly efficient, but its something i'm interested in exploring a little more. 23:45:03 the key concept is a model which maps the class structure to the relational structure, i think the openacs system lends itself *very* well to an or mapping system mainly because of the amount of object meta info lying around 23:45:58 i wrote up a half-finished paper on zope relational integration a little while ago, its a bit dated, at http://www.zope.org/Members/k_vertigo/Design/smartobjects 23:46:11 i'm not entirely qualified to argue one way or the other here, but the mapping problem has always seemed like one of unneeded complexity without any clear benefit 23:46:37 yes, you get OOP, but AOLservers approach seems like a better kind of code reuse 23:46:40 talli: i think a certain amount of metadata is very helpful 23:47:16 but how is metadata related to the object-relational mapping problem? 23:47:18 talli: at least for formulating common queries, the key to an or mapping system, is that it shouldn't get in your way when you need to drill past to make a query to the db. 23:48:11 i think that my question is usually related directly to OACS 23:48:38 IMHO, aolserver plus an RDBMS makes the or mapping problem disappear 23:48:42 i think you're question is focusing on python vs. java and or techniques at the acs datamodel 23:49:19 so i guess what i'm asking is what do you gain by using an OOP language over aolserver? 23:49:32 i'm not being confrontational, just wondering 23:50:21 You can use Tcl with one of its OO extensions in AOLserver BTW 23:50:54 rbm: i was wondering about that , i looked around but i couldn't figure out which ones were compatible? which ones work? 23:51:18 hazmat: iTcl should work. XOTcl should work too. 23:51:37 I haven't tested either, but a friend of mine has tested iTcl I think 23:52:39 rbm: what are TclObjects? and what's the big deal about them being integrated into AOlserver4? 23:53:15 hey guys! i just got an email from monpost6969 telling me about an amazing breakthrough in weightloss!!!! 23:53:17 talli: you'll also get access to all the resources and libraries that the language has to offer, i grant that aolserver is getting an impressive number of libraries and applications (and i hope to add to that list). 23:53:43 but you can't do everything from aolserver... 23:53:54 out of curosity talli, how long did etp take to build? 23:54:18 thanks rbm, i'll look into it more when i got some spare cycles :) 23:55:13 the basic functions or etp were probably built in a week or so 23:55:20 talli: In Tcl < 8, everything was a string. That was very inneficient. In Tcl 8 there are Tcl Objects, which makes things a lot faster. 23:55:23 but that's because Luke, the God-like Machine, did it 23:55:51 AOLserver internally still uses the Tcl Strings 23:55:55 (method) 23:56:12 but we've been talking about trying to see what etp would look like if it were written in python. just to prototype python in aolserver and openacs 23:56:35 so what is a Tcl Object? 23:56:41 just a thing? 23:56:44 I'm going to implement a graphical app accessing PostgreSQL pretty soon. I'm excited about it. 23:56:45 string or otherwise? 23:56:49 actually it uses ns_dstrings? are those tcl? i thought they were naviserver c stuff 23:57:12 talli: It's like a structure/class, but it's internal to Tcl. 23:57:19 talli: i was using etp the other day, and wondering how long it took cause i was thinking how long it would take me in zope/python... 23:57:39 hazmat: They hook up to the Tcl strings I think. I don't know that much from the AOLserver internals. 23:58:04 it didn't take long. the difference between apps in zope and apps in openacs is you don't need to datamodel stuff in zope, right? 23:58:15 at least not in the same way or as anally 23:58:17 aolserver actually has some really clean source, its refreshing to look at well written c. 23:58:39 hazmat: I know! I love it too. 23:59:30 how much noise is generated on the aolserver mailing list? is it worth being on it? 23:59:32 plus you get a whole lot more flexibility wrt to searching and all sorts of other stuff, like making it web services scriptable, and so much more.... sigh. i should stop before i get on this track 23:59:46 talli: not much, its very worthwhile to be on, rob mayoff is a god