IRC Chat : 2012-12-17 - OpenMRS

00:01:17 *** harshadura_ has joined #openmrs
00:01:20 *** harshadura has quit IRC
00:03:28 *** harshadura_ has quit IRC
00:03:37 *** harshadura has joined #openmrs
00:06:17 *** yony258 has quit IRC
02:45:56 *** harshadura has quit IRC
04:15:49 *** kishoreyekkanti has joined #openmrs
05:09:32 *** Hasintha has joined #openmrs
05:28:26 *** hasintha has joined #openmrs
05:44:30 *** hasintha has quit IRC
06:38:22 *** djazayeri has joined #openmrs
06:38:22 *** ChanServ sets mode: +o djazayeri
07:33:54 *** dkayiwa has joined #openmrs
07:44:16 *** dkayiwa has quit IRC
08:27:48 *** kavuri has joined #openmrs
08:37:09 *** kavuri has quit IRC
08:58:44 *** kavuri has joined #openmrs
09:00:51 *** kavuri has quit IRC
09:02:36 *** kavuri has joined #openmrs
09:09:14 *** djazayeri has quit IRC
09:18:25 *** dkayiwa has joined #openmrs
09:35:38 <kavuri> dkayiwa: hi
09:35:45 <dkayiwa> kavuri: hi
09:35:58 <kavuri> dkayiwa: found a strange bug - https://tickets.openmrs.org/browse/RESTWS-314
09:36:16 <kavuri> dkayiwa: unit testcase passes, response is correct. But looks like the data is not updated to database
09:36:23 <kavuri> dkayiwa: so could be some problem in DAO object
09:36:34 *** sagar_ has joined #openmrs
09:36:53 <dkayiwa> kavuri: ok
09:37:22 <kavuri> dkayiwa: any inputs on what could be the problem?
09:37:32 <dkayiwa> kavuri: let me look into it now
09:37:41 <kavuri> dkayiwa: thank you
09:40:25 *** rafa has joined #openmrs
09:40:25 *** ChanServ sets mode: +v rafa
09:41:32 *** kavuri has quit IRC
09:43:20 <rafa> dkayiwa: good morning!
09:43:28 <dkayiwa> rafa: good morning
09:43:33 <rafa> dkayiwa: feel like pairing?
09:44:03 <dkayiwa> rafa: no problem as long as it is after something like an hour from now :)
09:44:19 <rafa> dkayiwa: okay, let me know when you're ready
09:45:14 <dkayiwa> rafa: no problem will surely do so :)
09:56:23 *** rafa has quit IRC
09:58:03 *** rafa has joined #openmrs
09:58:03 *** ChanServ sets mode: +v rafa
10:00:26 *** sagar_ has quit IRC
10:00:55 *** sagar_ has joined #openmrs
10:17:59 *** akmad has joined #openmrs
10:19:24 *** ibewes has joined #openmrs
10:21:36 <ibewes> hello all!
10:23:17 <dkayiwa> ibewes: hi
10:26:49 *** ibewes_ has joined #openmrs
10:30:09 <ibewes> dkayiwa: hrm, well I'm not sure why I'm in here twice... switched over to Ubuntu this weekend and still figuring things out
10:30:34 <dkayiwa> ibewes: :)
10:30:52 <dkayiwa> ibewes: IRC will sort itself out later :D
10:31:54 <ibewes> dkayiwa: here's hoping!
10:35:12 *** ibewes__ has joined #openmrs
10:35:51 *** ibewes__ has quit IRC
10:38:48 <dkayiwa> hi rafa
10:39:00 <rafa> dkayiwa: ready?
10:39:07 <dkayiwa> rafa: oh yes
10:39:23 <rafa> dkayiwa: let's do this with teamviewer
10:39:29 <dkayiwa> rafa: ok
10:39:33 *** kavuri has joined #openmrs
10:39:49 <kavuri> dkayiwa: sorry! got logged out for sometime
10:40:06 <dkayiwa> kavuri: was looking for you :)
10:40:22 <kavuri> dkayiwa: internet outage at my place.. :(
10:42:05 <kavuri> dkayiwa: anything you found about the problem?
10:42:24 <dkayiwa> kavuri: was waiting for you to come back because i had some questions for you
10:42:30 <kavuri> dkayiwa: sure
10:44:20 <dkayiwa> kavuri: am in the middle of something with rafa now. But will surely get back to you after :)
10:44:28 <kavuri> dkayiwa: ok
10:46:56 *** k-joseph has joined #openmrs
10:47:00 <OpenMRSBot> Recent updates in the world of openmrs: Shazin Sadakath: SpringObjectFactory Error in creating application in Struts 2.3.7 <http://shazsterblog.blogspot.com/2012/12/springobjectfactory-error-in-creating.html>
10:47:49 *** cpower has quit IRC
10:49:29 *** lh has joined #openmrs
10:49:29 *** lh has joined #openmrs
11:10:01 <dkayiwa> k-joseph: is internet working well for you?
11:15:46 <k-joseph> dkayiwa: yes it is
11:15:59 *** ibewes_ has quit IRC
11:15:59 *** ibewes has quit IRC
11:16:01 <dkayiwa> k-joseph: ok good
11:17:05 <k-joseph> dkayiwa: please come again, on clean install
11:17:29 <dkayiwa> k-joseph: after clean install. run the app
11:19:48 <k-joseph> dkayiwa: should i run it manually in the current directory, when i do mvn clean and then mvn install from eclipse, am getting a build failure
11:20:28 <dkayiwa> k-joseph: can you right click on the project in eclipse and do team - reset- HARD
11:20:37 <dkayiwa> k-joseph: then rerun clean install
11:23:02 <k-joseph> dkayiwa: reset- HARD, is missing team
11:23:37 <dkayiwa> k-joseph: do you see the RESET under Team?
11:23:47 <k-joseph> dkayiwa: no
11:23:59 <dkayiwa> k-joseph: did yuo share the projects?
11:24:06 <k-joseph> dkayiwa: yes
11:24:17 <dkayiwa> k-joseph: what do you see under Team?
11:25:15 <k-joseph> dkayiwa: apply patch, share project, share projects and Upgrade projects
11:25:41 <dkayiwa> k-joseph: when you see share project, it means you have not yet shared the project
11:28:04 <k-joseph> dkayiwa: shared now but under team- reset is not available, am running again clean install
11:28:27 <dkayiwa> k-joseph: what do you have under Team?
11:29:19 <k-joseph> dkayiwa: commit, advanced, Merge, ignore.......
11:29:58 <dkayiwa> k-joseph: can you lost all
11:30:03 <dkayiwa> lost - list
11:33:13 <k-joseph> dkayiwa: lost - list is not included in the list of choices, is it either == untrack, 0r disconnect
11:34:05 <dkayiwa> k-joseph: when you right click on say the api project, right click and choose Team, what items do you see under it?
11:36:14 <k-joseph> dkayiwa: i see a list of alternatives, like untract, disconnect, create patch, apply patch, among others
11:36:36 <dkayiwa> k-joseph: can you list all of them?
11:40:32 <k-joseph> dkayiwa: commit, remote, switch to, advanced, syncronize work space, merge, create patch, apply patch, ignore, add index, remove from index, untract, show in repository view, show in history and disconnect, others are not highligted
11:40:59 <dkayiwa> k-joseph: how did you share the project?
11:42:09 <k-joseph> dkayiwa: shared as git
11:42:16 <dkayiwa> k-joseph: ok
11:42:25 <dkayiwa> k-joseph: can you run mvn clean install from command line?
11:42:52 <k-joseph> dkayiwa: ok
11:46:23 <k-joseph> dkayiwa: i get the error: http://pastebin.com/r4uvk0ak from command prompt
11:48:09 <dkayiwa> k-joseph: from which dirrectory are you running that command?
11:48:23 *** lh has quit IRC
11:49:37 <k-joseph> dkayiwa: the git repository for openmrs, created where the six projects were moved to
11:49:55 <dkayiwa> k-joseph: whats the complete path for it?
11:50:30 <k-joseph> dkayiwa: D:\git\openMRS
11:51:43 <dkayiwa> k-joseph: what folders are listed there?
11:53:09 <k-joseph> dkayiwa: .git, openmrs-api, openmrs-release-test, openmrs-test, openmrs-tools, openmrs-web, openmrs-webapp
11:53:43 <dkayiwa> k-joseph: can you clone the project again to a new folder?
11:53:54 <k-joseph> dkayiwa: dkayiwa: ok
11:58:28 *** k-joseph has quit IRC
12:03:06 *** k-joseph has joined #openmrs
12:12:05 *** sagar_ has quit IRC
12:19:54 *** robbyoconnor has quit IRC
12:25:15 *** ibewes has joined #openmrs
12:41:04 <k-joseph> dkayiwa: i have finished cloning, then done mvn clean install from command prompt, am getting only 3 projects successful
12:41:31 <dkayiwa> k-joseph: pastebin the error
12:43:13 <k-joseph> dkayiwa: http://pastebin.com/6sFnsHXZ
12:45:34 <dkayiwa> k-joseph: which command are you running?
12:46:10 <k-joseph> dkayiwa: mvn clean install
12:46:18 <dkayiwa> k-joseph: from which folder?
12:46:39 <k-joseph> dkayiwa: from where i had clone openmrs from
12:46:57 <dkayiwa> k-joseph: can you run the command again?
12:47:06 <k-joseph> dkayiwa: ok
12:47:57 *** suranga has joined #openmrs
12:47:57 *** ChanServ sets mode: +v suranga
12:50:42 *** kishoreyekkanti has quit IRC
12:53:21 <k-joseph> dkayiwa: still failing with the same error, repeated for now the third time
12:54:59 <dkayiwa> k-joseph: did you make any changes to the code?
12:55:18 <k-joseph> dkayiwa: not yet
12:56:19 <dkayiwa> k-joseph: can you pastebin again?
12:57:46 <k-joseph> dkayiwa: http://pastebin.com/QUDwbMUh
12:59:22 <dkayiwa> k-joseph: are you running from the normal windows command prompt?
13:00:24 <k-joseph> dkayiwa: no, am running from a git command prompt, am now going to run from cmd
13:03:25 <k-joseph> dkayiwa: the same failure message with the normal window command prompt
13:03:53 <dkayiwa> k-joseph: is it possible to pastebin the entire log?
13:05:41 <k-joseph> dkayiwa: http://pastebin.com/f1hTD5pp
13:09:24 <dkayiwa> k-joseph: can you send this to the dev list?
13:09:33 <k-joseph> dkayiwa: ok
13:11:08 *** ibewes has quit IRC
13:14:28 *** kavuri has quit IRC
13:21:26 *** jkeiper has joined #openmrs
13:21:26 *** ChanServ sets mode: +v jkeiper
13:25:57 <jkeiper> hi y'all
13:26:28 <suranga> jkeiper, howdy howdy :)
13:26:51 <jkeiper> suranga++
13:28:08 <suranga> !karama suranga
13:28:09 <OpenMRSBot> suranga: Error: "karama" is not a valid command.
13:28:18 <suranga> !karma suranga
13:28:18 <OpenMRSBot> suranga: Karma for "suranga" has been increased 12 times and decreased 2 times for a total karma of 10.
13:31:53 <dkayiwa> k-joseph: which version of maven are you running?
13:32:57 <k-joseph> dkayiwa: version-3.0.4
13:33:17 <dkayiwa> k-joseph: is that for the windows command prompt?
13:34:19 <k-joseph> dkayiwa: Apache Maven 3.0.4 from command prompt
13:35:02 <dkayiwa> k-joseph: can you clone this module? https://github.com/openmrs/openmrs-module-appointment/
13:35:03 <OpenMRSBot> <http://ln-s.net/+HdB> (at github.com)
13:35:11 <dkayiwa> k-joseph: then try mvn clean install for it?
13:40:16 <k-joseph> dkayiwa: mvn clean install is successfull for this module
13:44:00 *** wyclif_ has quit IRC
13:53:15 <dkayiwa> k-joseph: and which version of java are you using?
13:54:10 <k-joseph> dkayiwa: java 7
13:54:32 <dkayiwa> k-joseph: can you try with lower?
13:55:37 <k-joseph> dkayiwa: ok
14:03:39 <rafa> dkayiwa: the easiest way to get my changes it to do: git fetch https://github.com/rkorytkowski/openmrs-module-registrationcore.git RC-3:RC-3
14:03:45 <OpenMRSBot> <http://ln-s.net/+Ilp> (at github.com)
14:03:54 <rafa> dkayiwa: then: git checkout RC-3
14:04:48 *** ibewes has joined #openmrs
14:05:35 *** hasintha has joined #openmrs
14:07:19 <suranga> hi rafa , when u guys are releasing a module, do you run the git commands from a branch, or trunk itself ? :)
14:07:37 <rafa> suranga: from master
14:07:54 <suranga> rafa, aha. so no branching ?
14:08:09 <suranga> rafa, ok, so thats probably why it wasnt working for me :P
14:08:10 <dkayiwa> rafa: ok
14:08:25 <rafa> suranga: it will create a new tag but no branch
14:08:40 <rafa> suranga: assuming you use mvn release plugin
14:09:23 <suranga> rafa, thanks, will re-try that way :)
14:10:14 <hasintha> hi all,i tried to run openmrs in eclipse it gives NoPluginFoundForPrefixException this can anyone help me.
14:10:44 <rafa> hasintha: is it when you run a maven build?
14:11:13 <rafa> hasintha: please describe the steps you follow
14:13:38 <hasintha> first i run mvn clean install then in webapp i run jetty:run then it gives that exception
14:14:43 <rafa> hasintha: do you specify jetty:run in the goals field for the webapp project?
14:15:29 <hasintha> yes..i dd so...but it gives it..
14:16:03 <rafa> hasintha: let's try again. right click on the webapp project and do Debug As -> Maven Build... and then specify goals: jetty:run
14:17:13 <hasintha> i tried that way also but same thing happend..:(
14:18:03 <rafa> hasintha: what is the full exception?
14:20:32 <hasintha> No plugin found for prefix 'jetty' in the current project and in the plugin groups [org.apache.maven.plugins, org.codehaus.mojo] available from the repositories [local (C:\Users\Hasintha\.m2\repository), openmrs-repo (http://mavenrepo.openmrs.org/nexus/content/repositories/public), central (http://repo.maven.apache.org/maven2)]
14:20:33 <OpenMRSBot> <http://ln-s.net/8-2X> (at mavenrepo.openmrs.org)
14:21:00 <rafa> hasintha: it looks as if you're trying to run it from a different project than webapp
14:23:14 <rafa> hasintha: when you import openmrs you should see 6 or 7 projects in eclipse like: openmrs, openmrs-api, openmrs-test, openmrs-web, openmrs-webapp
14:23:25 <rafa> hasintha: is it what you see in project explorer?
14:25:58 <hasintha> ya,do i wanna run jetty:run comand in openmrs-webapp??
14:27:21 <rafa> hasintha: yes
14:28:46 <hasintha> it's ok now takz +rafa...:)
14:29:44 <rafa> hasintha: you're welcome
14:33:56 *** wyclif_ has joined #openmrs
14:34:12 *** suranga has quit IRC
14:35:02 *** hasintha has quit IRC
14:38:20 *** kishoreyekkanti has joined #openmrs
14:39:41 *** k-joseph has quit IRC
14:43:18 *** k-joseph has joined #openmrs
14:57:23 *** ibewes has quit IRC
14:59:34 *** cpower has joined #openmrs
14:59:40 <cpower> Good day everyoen
14:59:43 <cpower> everyone
15:00:20 <cpower> 5 days till the end of the world...have you said your goodbyes?
15:01:42 <cpower> Well ok the early morning silence tells me to get to something that is always a constant...even when the world is not...SCRUM!
15:01:48 <cpower> !scrumon cpower
15:01:48 * OpenMRSBot says the DAILY SCRUM MEETING is STARTING. This meeting should not last longer than 15 minutes. Please hold other comments until the end of the meeting, or message someone privately. Thank you! ScrumMaster cpower- you may begin when ready.
15:02:49 <cpower> Order: jkeiper, dkayiwa, wyclif_, andrea, Darius, and whomever else would like to contribute
15:03:03 <jkeiper> Until now:
15:03:03 <jkeiper> - worked with wyclif on RC-1
15:03:03 <jkeiper> - discussed need for a new EVNT ticket
15:03:03 <jkeiper> - meetings
15:03:03 <jkeiper> Until next sprint:
15:03:03 <jkeiper> - work with wyclif on RC-1, RC-2
15:03:05 <jkeiper> - CALC-47
15:03:07 <jkeiper> Blockers:
15:03:09 <jkeiper> - None
15:03:44 <rafa> jkeiper: Until next *scrum* :)
15:03:49 <jkeiper> HAH!
15:03:52 <jkeiper> that's what you think!
15:03:52 <cpower> ha
15:03:57 <jkeiper> ;-)
15:03:57 <cpower> dkayiwa, you are up
15:04:03 <rafa> lol
15:04:07 *** andrea_ has joined #openmrs
15:04:27 *** djazayeri has joined #openmrs
15:04:27 *** ChanServ sets mode: +o djazayeri
15:04:28 <cpower> ...
15:04:31 <dkayiwa> Paired with rafa on: Add service method to search for similar persons and patients - RC-3
15:04:31 <dkayiwa> Trying to fix a failing unit test for it
15:04:32 <dkayiwa> No Blockers
15:04:43 <cpower> wyclif_
15:04:54 <wyclif_> Friday:
15:04:54 <wyclif_> curating tickets
15:04:54 <wyclif_> Had a couple of discussions with jeremy about the registration module tickets
15:04:54 <wyclif_> RC-1 - Create a model for a registration event
15:04:54 <wyclif_>
15:04:55 <wyclif_> Monday:
15:04:57 <wyclif_> Complete and commit code for RC-1
15:04:59 <wyclif_> Pick anotehr sprint ticket
15:05:01 <wyclif_>
15:05:03 <wyclif_> Blockers: None
15:05:39 <cpower> andrea_ go for it
15:06:12 <andrea_> Friday
15:06:13 <andrea_> TRUNK-2768
15:06:13 <andrea_> Found controllers
15:06:13 <andrea_> PersonFormController
15:06:13 <andrea_> dwr/PatientListItem
15:06:13 <andrea_> Still tests in Error
15:06:15 <andrea_> Monday
15:06:17 <andrea_> TRUNK-2768
15:06:19 <andrea_> Pull request
15:06:21 <andrea_> Find problem causing errors in tests
15:06:25 <andrea_> Blocker still errors in tests on TRUNK-2768
15:07:18 <cpower> andrea_, work with jkeiper and see if you can get on some tickets that are in this sprint (I'm assuming the trunk tickets are bugs)
15:07:40 <cpower> djazayeri you are up
15:07:46 <andrea_> cpower, ggreat will do
15:08:03 <andrea_> cpower, yes they are bugs
15:08:18 <djazayeri> Working on Mirebalais, and doing advising on ITECH-kenya
15:08:30 <djazayeri> no blockers
15:08:35 <djazayeri> (sorry, on another call...)
15:08:38 <cpower> rafa your turn
15:08:45 <rafa> Today:
15:08:45 <rafa> * Paired with Daniel on RC-3: Add service method to search for similar persons and patients
15:08:45 <rafa> https://tickets.openmrs.org/browse/RC-3
15:08:45 <rafa> Blockers: an issue with querying by address, it should work the same way names do, but it finds nothing
15:08:45 <rafa> (see https://github.com/rkorytkowski/openmrs-module-registrationcore/compare/RC-3)
15:08:45 <rafa> Discussion points: getPreferred in PersonName and PersonAddress return FALSE instead of null if the field is null, which makes it impossible to ignore the preferred flag when searching by example (it's either false or true)
15:08:49 <OpenMRSBot> <http://ln-s.net/+Ims> (at github.com)
15:10:48 <cpower> Ok a rafa has our only discussion point for today so please see above and help him out. As a heads up, I'm on vacation next week, but will try to make it online for scrums to see how things are going. If you need me email/google IM is always the best ways to reach me
15:10:59 <cpower> !scrumoff
15:10:59 * OpenMRSBot says the DAILY SCRUM MEETING has ENDED. This channel is now returned to normal hacking operations. Post-scrum meeting follow-up conversations may now begin.
15:11:51 <djazayeri> tricky.
15:12:19 <djazayeri> I don't know how to get around that, in a compatible way
15:12:40 <jkeiper> hmm
15:13:19 <rafa> one way is to add Boolean preferred to searchForSimilarPeople in our module
15:13:50 <rafa> and we would ignore what is specified in PersonName or PersonAddress
15:15:22 *** travis-ci has joined #openmrs
15:15:22 <travis-ci> [travis-ci] [openmrs/openmrs-core] [e8e7921] [Jeremy Keiper] The build passed. - http://travis-ci.org/openmrs/openmrs-core/builds/3704195
15:15:22 *** travis-ci has left #openmrs
15:15:25 <OpenMRSBot> <http://ln-s.net/+In6> (at travis-ci.org)
15:15:32 <jkeiper> thanks travis
15:15:35 <rafa> :)
15:15:35 <jkeiper> ;-)
15:16:07 <dkayiwa> k-joseph: https://wiki.openmrs.org/display/docs/Using+Git
15:17:10 <rafa> djazayeri: or Boolean ignorePreferredFlag
15:17:47 <rafa> djazayeri: alternatively we could adjust hbms for PersonName and PersonAddress to access the field directly instead of using getter
15:20:48 <rafa> djazayeri, jkeiper: also for searchForSimilarPeople queries to be performing well we'll need to say in docs that indexes on columns you use for searching need to be created in the db
15:21:16 <rafa> I don't think we want the module to be creating them
15:21:19 <djazayeri> what's the method signature, and/or ticket?
15:21:30 <rafa> https://github.com/rkorytkowski/openmrs-module-registrationcore/compare/RC-3
15:21:31 <OpenMRSBot> <http://ln-s.net/+Ims> (at github.com)
15:24:21 <djazayeri> rafa, jkeiper: peeking at that code, isn't this going to be too specific?
15:24:43 <djazayeri> jkeiper: actually, per Burke's original suggestions, don't we need two methods:
15:24:51 <djazayeri> 1. findSimilarPatientsSpecific
15:24:56 <djazayeri> 2. findSimilarPatientsSensitive
15:24:58 <djazayeri> ?
15:25:26 <rafa> djazayeri: what is the difference between them?
15:25:54 <rafa> djazayeri: I mean how the behavior differs ;)
15:26:35 <djazayeri> one will err towards returning more results, to make sure that it at least finds the relevant patient
15:26:45 <djazayeri> the other will err towards retruning fewer results.
15:26:58 <djazayeri> sorry, let me try that again
15:27:30 <djazayeri> So, the idea is that there's one search that we're doing as each additional field about a patient is entered
15:28:01 <djazayeri> the other search is, after we've entered all fields, and we're just about to create a new patient, do a last-ditch search to make sure we really don't create a duplicate
15:28:28 <djazayeri> unfortunately I forget which is which!
15:28:55 <djazayeri> one of them is supposed to ensure that it definitely finds the matching patient (even if it returns lots of false negatives too)
15:29:46 <djazayeri> the other one should try to return a short list, without many false positives
15:29:52 <djazayeri> (sorry, I meant to say false positive both times)
15:30:04 <djazayeri> jkeiper: do you have the notes from the manila discussion?
15:30:37 <rafa> djazayeri: so one method should be doing AND queries whereas the more broad method should be doing ORs
15:30:53 <rafa> djazayeri: something in that direction?
15:31:26 <djazayeri> rafa: I think that either way it should all be fuzzy searching, rather than exact searching
15:32:09 <rafa> djazayeri: it can't be really done performant without lucene
15:32:30 <djazayeri> I.e. we're not searching for a patient with the identical birthdate, but within a couple years, etc
15:33:17 <djazayeri> and that we'd look for people with similar names (perhaps mis-spelled, perhaps out of order) rather than identical names
15:33:20 <djazayeri> same for addresses.
15:33:25 <rafa> djazayeri: heh that adds a lot to RC-3
15:33:35 <djazayeri> jkeiper: you around?
15:33:35 <rafa> djazayeri: I wish it was specified there :)
15:36:51 <rafa> djazayeri: okay, so going further one of those methods would do born in a month and the other born in a day?
15:38:31 <rafa> dkayiwa: givent that ^ I think we can stop investigating why query by example doesn't work for address ;)
15:38:34 <djazayeri> rafa: the original idea was that the module would actually provide a mechanism for modules to plug in implementations of these. :-)
15:38:43 <djazayeri> :-)
15:38:47 <dkayiwa> rafa: oh what a relief!!! :)
15:38:47 <jkeiper> djazayeri, yes
15:38:51 <djazayeri> also not written on the ticket, as far as I can tell.
15:38:52 <jkeiper> also djazayeri: https://tickets.openmrs.org/browse/EVNT-25
15:39:10 <jkeiper> djazayeri, i was not in manila ... which discussion?
15:39:20 <djazayeri> but definitely we don't want to be using a hibernate query-by-example
15:39:39 <djazayeri> it's find _similar_ patients, not find _identical_ patients
15:39:43 <djazayeri> jkeiper: oh, right...
15:39:59 <djazayeri> jkeiper: we had a session about registration
15:40:10 <jkeiper> djazayeri, ah!
15:40:12 <djazayeri> jkeiper: led by burke, and brainstormed a bunch of features
15:40:22 <djazayeri> jkeiper: I assumed that Burke would have made sure you saw this.
15:40:36 <jkeiper> djazayeri, nope ... probably assumed i knew
15:40:44 <jkeiper> djazayeri, downey just said it might be on youtube
15:41:45 <jkeiper> mm, don't see it right away
15:41:54 <jkeiper> djazayeri, anyways ... there should be a notes page right?
15:42:02 <djazayeri> jkeiper: yeah...
15:42:56 <jkeiper> ah, downey said lanyrd site should have note links
15:42:59 * jkeiper looks
15:43:17 <djazayeri> jkeiper: this I guess...
15:43:19 <djazayeri> http://notes.openmrs.org/omrs12-registration
15:43:21 <djazayeri> though I thought that we
15:43:26 <djazayeri> I thought that we had written more
15:43:39 <jkeiper> hrm
15:44:29 <jkeiper> so, rafa ... this is one reason why i thought just building a lucene index of patients would make it easier to search for potential matches ;-)
15:44:39 <jkeiper> djazayeri, would you agree or is that going too far?
15:45:58 <rafa> jkeiper: it's going as far as 1.10 ;)
15:46:38 <rafa> jkeiper: lucene jars did make it to 1.9.x
15:46:45 <rafa> jkeiper: did not
15:46:57 <djazayeri> jkeiper, rafa: this is where having a reporting core API that allows for this algorithm to be plugged in by another module is a good thing.
15:47:14 <jkeiper> rafa, i thought hibernate-search could be added by a module ;-) ...
15:47:20 <djazayeri> I think our goal is to get the method signature right, and start off with a hacky implementation
15:47:33 <djazayeri> so initially we have _something_ :-)
15:47:44 <jkeiper> djazayeri, agreed ... pluggable ftw, but also we need to let people rely on it eventually
15:48:09 <djazayeri> then we'd work on doing something cool via lucene later...
15:48:17 *** rfriedman has joined #openmrs
15:49:08 <rfriedman> jkeiper, wyclif_ , how do you intend to get the payload information?
15:49:26 <jkeiper> manually ...?
15:49:40 <rfriedman> the constipated mathematician worked it out with a pencil
15:50:25 <rfriedman> i guess what i'm asking is, do you intend to increase the scope of what the event notification contains, or will it be on the event recipient to check for its own information?
15:51:35 <rafa> djazayeri: jkeiper: so this method should take a person object and a recipe for what is considered similar
15:52:14 <djazayeri> rafa: I don't think so
15:52:49 <jkeiper> possibly follow the model of Comparator
15:52:50 <jkeiper> ?
15:53:10 <djazayeri> rafa: the client making the call to findSimilarPatients shouldn't be the one choosing the algorithm
15:53:21 <djazayeri> thought they might specify a "specificity level" parameter
15:53:46 <jkeiper> a customized registration module should be essentially overriding or registering a search method?
15:54:08 <rfriedman> jkeiper, wyclif_ , this is in response to your message about the fire method, sorry if I came into the middle of a different conversation
15:54:29 <rafa> djazayeri: "specificity level" is just a constant or a recipe which can be adjusted for different fields?
15:55:05 <djazayeri> rafa, jkeiper: the idea is that the registration UI module just knows that it wants to call "get similar patients", and it's up to some plugin/registration mechanism, and admin screen to configure whether the implementation of that method is a sucky default one, or something fancy provided by the patient matching module, etc.
15:55:13 <jkeiper> rfriedman, one sec ..
15:55:38 <rafa> djazayeri: ok
15:55:53 <jkeiper> djazayeri, ok .. so first run is to just perform minimal comparisons and return _something_
15:55:55 <djazayeri> rafa, jkeiper: there's no actual need to do this plugin/registration/admin idea here
15:56:17 <djazayeri> as long as we're getting the right method signatures to assume that eventually other modules should be able to plug in better algorithms
15:56:48 <jkeiper> djazayeri, agreed
15:57:07 <jkeiper> rafa, the take-away is to not worry about "preferred" for now :-D
15:57:17 <djazayeri> I seem to recall Burke suggesting a float "how specific", which between 0 and 1.
15:57:22 <djazayeri> but I don't remember the details
15:57:25 <jkeiper> rafa, even just implementing what mirebalais or ampath has in it will be fine to begin with
15:57:28 <djazayeri> jkeiper: are you in a position to ask him?
15:57:44 <jkeiper> djazayeri, that makes sense to match against relevancy scores if we were to calculate them
15:57:52 <jkeiper> djazayeri, he's not nearby :(
15:57:57 <jkeiper> djazayeri, pretty sure he's in a mtg now
15:58:40 <djazayeri> jkeiper: oh, right, it wasn't a specificity score, it was a relevancy score
15:59:13 <djazayeri> anyway, jkeiper, rafa, can you please email burke (via the list) and say I said he would have an opinion on the method signature of the "find similar patient" method or methods?
15:59:55 <jkeiper> sure, i'll do it rafa
16:00:25 <rafa> jkeiper: ok
16:00:56 *** cpower has quit IRC
16:06:19 <rfriedman> jkeiper, u there?
16:06:55 <jkeiper> rfriedman, yep. mail sent rafa
16:07:02 <jkeiper> rfriedman, so ... for the events ...
16:07:48 <jkeiper> rfriedman, i think it would be nice for a particular target to have a known contractual format of Message payloads
16:07:54 <jkeiper> rfriedman, but until we get there ...
16:08:39 <rfriedman> seemed to me like there were 3 possible approaches:
16:08:39 <jkeiper> rfriedman, i think the event module should let me fire an event of my own naming (or at least close to it) without having to rely on AOP, and that i should be allowed to communicate more than just basic persisted object metadata like uuid and classname
16:09:23 <rfriedman> (1) expand the contract as to what the message will contain
16:09:41 <rfriedman> (2) have the subscription define what the message will contain
16:09:57 *** downey has joined #openmrs
16:10:03 <rfriedman> (3) keep the message minimal and let the subscriber figure out whether/how it wants to deal with the message
16:10:21 *** downey is now known as Guest39736
16:10:49 <rfriedman> jkeiper, which are you proposing?
16:11:13 *** Guest39736 has quit IRC
16:11:13 *** Guest39736 has joined #openmrs
16:11:13 *** ChanServ sets mode: +o Guest39736
16:12:12 <jkeiper> hmmm
16:13:08 <jkeiper> rfriedman, well ... a message is essentially a hashmap, so finding what information is sent can be found by getKeys
16:14:20 <rfriedman> jkeiper, that's not what i'm asking about as much as allocation of work between event and consumer
16:14:55 <jkeiper> rfriedman, in this case I want to simply provide more information in the message ... how we establish that contract, I don't know
16:16:16 <rfriedman> jkeiper, so addressing your additional message content, I don't like the idea of (1) because I think your requests are not all that common
16:16:35 *** Guest39736 has quit IRC
16:16:59 <rfriedman> I don't mind (2) that much because the work will have to be done anyway, but maybe the consumer can be smart about when it asks for what
16:17:08 *** downey_ has joined #openmrs
16:17:08 <rfriedman> jkeiper, so I prefer (3)
16:17:15 *** downey_ has quit IRC
16:17:15 *** downey_ has joined #openmrs
16:17:15 *** ChanServ sets mode: +o downey_
16:18:32 *** downey_ has quit IRC
16:20:15 *** downey_ has joined #openmrs
16:20:15 *** ChanServ sets mode: +o downey_
16:21:18 <jkeiper> rfriedman, so with #3 does that give me the chance to add more content (even though it is outside / in addition to the contract)?
16:22:07 <rfriedman> jkeiper, you should know from the message everything that event knows ... if it is not providing that information, then the contract needs to be changed
16:23:09 <jkeiper> rfriedman, as it stands the contract (if there even is one) is that the event is communicating a persisted object change of some kind (create, update, save, etc)
16:23:41 <rfriedman> jkeiper, right, my undertanding was that it was sending the type of change, the type of object, and the id of the object
16:24:16 <jkeiper> rfriedman, yes ... and since these are not persisted objects I need to communicate more in the actual message
16:24:33 <jkeiper> rfriedman, or implement my own messaging service just for this one notification
16:25:22 <rfriedman> jkeiper, what are not persisted objects? i thought event was based on hiberante interceptors?
16:25:40 <jkeiper> rfriedman, Registration
16:25:45 <wyclif_> djazayeri, any comment about https://tickets.openmrs.org/browse/EVNT-25?
16:25:48 <jkeiper> rfriedman, event module is
16:25:56 <jkeiper> rfriedman, "events" are not necessarily tied to hibernate
16:26:06 <jkeiper> rfriedman, but the basic implementation is
16:26:39 <jkeiper> rfriedman, so i either implement my own messaging system (complete waste of time) or i use the one provided by event module for a purpose outside of AOP
16:26:56 <rfriedman> jkeiper, so you are wanting to use it like an event bus?
16:27:13 <jkeiper> rfriedman, yes. pub/sub
16:29:33 <rfriedman> jkeiper, if you are defining your own message type, e.g. "patient photo just taken", then there probably needs to be a message type identifier, for which I would use event type, leaving the existing ones alone ... unless they're an enum or something awful like that
16:30:33 <rfriedman> jkeiper, it would be easier for compatibility if you included the other current message fields, even if only with dummy values
16:31:12 <jkeiper> rfriedman, exactly ... i agree we should commit to having the default message keys
16:31:33 <jkeiper> rfriedman, as for the "topic", i want to stick with the existing paradigm of "action" + className
16:31:39 <jkeiper> the action is, unfortunately, an enum
16:31:58 <jkeiper> but in this case, the className can be a marker class
16:33:13 <jkeiper> rfriedman, i'm trying to stick with the particular formats that the event module is using so i don't wander too far off the path ... this change is simply to override the message format but could guarantee that certain keys exist (and fill them in with blank values if they do not)
16:33:44 <rfriedman> jkeiper, sounds like you are addressing my concerns
16:36:17 <jkeiper> rfriedman, cool :-D
16:41:22 <rfriedman> jkeiper, not sure if this is what you mean by marker, but the "classname" should actually represent a class, even if it's just a token for the event
16:48:24 *** wyclif_ has quit IRC
16:49:58 *** downey_ has quit IRC
17:17:57 *** andrea_ has quit IRC
17:18:22 *** k-joseph has quit IRC
17:22:42 *** dkayiwa_ has joined #openmrs
17:22:43 *** dkayiwa has quit IRC
17:22:43 *** dkayiwa_ is now known as dkayiwa
17:26:52 *** rafa has quit IRC
18:15:54 <jkeiper> rfriedman, yep that's what i meant :-)
18:31:54 *** wyclif_ has joined #openmrs
18:34:05 *** rfriedman has quit IRC
19:02:12 <jkeiper> djazayeri, yt?
19:02:17 <djazayeri> jkeiper: yes
19:02:24 *** dkayiwa has quit IRC
19:06:29 *** kishoreyekkanti has quit IRC
19:07:55 *** robbyoconnor has joined #openmrs
19:07:55 *** ChanServ sets mode: +v robbyoconnor
19:29:03 *** dkayiwa has joined #openmrs
19:45:53 *** his has joined #openmrs
19:50:02 *** his has quit IRC
19:52:17 *** travis-ci has joined #openmrs
19:52:17 <travis-ci> [travis-ci] [openmrs/openmrs-core] [114faf2] [Jeremy Keiper] The build passed. - http://travis-ci.org/openmrs/openmrs-core/builds/3708574
19:52:17 *** travis-ci has left #openmrs
19:52:21 <OpenMRSBot> <http://ln-s.net/+Itc> (at travis-ci.org)
20:00:54 <OpenMRSBot> Recent updates in the world of openmrs: OpenMRS Kenya Project: KenyaEMR Release 2012.2 - Minimum Implementable Release (Woohoo!) <http://openmrskenya.blogspot.com/2012/12/kenyaemr-release-20122-minimum.html> || OpenMRS Kenya Project: KenyaEMR Release 2012.2.1 <http://openmrskenya.blogspot.com/2012/12/kenyaemr-release-201221.html>
20:09:28 *** dkayiwa has quit IRC
20:23:38 *** dkayiwa has joined #openmrs
20:39:28 *** dkayiwa has quit IRC
20:48:09 *** dkayiwa has joined #openmrs
21:02:36 *** prolificgnosis has quit IRC
21:18:15 *** Mkop has joined #openmrs
21:18:15 *** ChanServ sets mode: +v Mkop
21:25:39 *** andreapat has joined #openmrs
21:26:02 <andreapat> jkeiper, u around?
21:33:21 *** r0bby has joined #openmrs
21:33:21 *** ChanServ sets mode: +v r0bby
21:36:06 *** jkeiper has quit IRC
21:36:52 *** robbyoconnor has quit IRC
21:53:30 *** dkayiwa has quit IRC
22:04:21 *** rafa has joined #openmrs
22:04:21 *** ChanServ sets mode: +v rafa
23:06:50 *** wyclif_ has quit IRC
23:12:29 *** rafa has quit IRC