Open Source Integrated Library System

Evergreen on IRC

#evergreen Logs for Tuesday, November 8th, 2011

< Monday, November 7th, 2011Raw Log FileWednesday, November 9th, 2011 >
#TimeNickMessage
#00:33:53dbs has quit IRC
#00:35:30enhancin has joined #evergreen
#00:37:11enhancinwe've imported a few thousand electronic records, but they're not showing up in the public list only inside the staff client. any ideas?
#00:49:06LBA has quit IRC
#01:05:00pinesol_green` has joined #evergreen
#01:05:19pinesol_green has quit IRC
#01:49:42darshan has joined #evergreen
#03:18:51darshan has quit IRC
#03:21:20darshan has joined #evergreen
#03:35:46darshan has quit IRC
#03:40:54darshan has joined #evergreen
#05:39:21darshani want to set an action trigger for hold.capture,and so i have set it with command
#05:43:40darshan./action_trigger_runner.pl --process-hooks --osrf-config /openils/conf/opensrf_core.xml --hooks hold.capture --debug-stdout
#05:43:56darshanbut everything is f9
#05:44:07darshanbut i am not recieivng an mail
#05:44:21darshani even have set emailid of my patron
#05:44:51darshanand even had configure my server host in opensrf_core.xml file
#06:02:44wlayton has joined #evergreen
#06:04:08jamesrf has quit IRC
#06:06:03jamesrf has joined #evergreen
#06:34:35enhancin has quit IRC
#06:37:06tfaile has joined #evergreen
#06:42:44wlayton has quit IRC
#06:49:27kmlussier has joined #evergreen
#07:10:46jamesrf has quit IRC
#07:11:03jamesrf has joined #evergreen
#07:11:28csharpdarshan: you'll have to set up an email server (something that's beyond the scope of this channel's support)
#07:16:38darshan<csharp> i have
#07:27:55darshancan u give me some advice on how to set mail server in linux
#07:32:34darshani have set it but then too want to check it
#07:48:21kivilahtioOk, I have one odd problem. I have defined organization units and some aquisition funds/providers as eg_admin. As a separate user with global admin priviledges, I can't see any OU's in context dropdowns nor any Funds.
#07:49:27kivilahtioI should definetely have OU's visible. I reran the autogen.sh script as well and restarted the osrf-container and apache2 server
#08:09:00fortin has joined #evergreen
#08:12:42Dyrcona has joined #evergreen
#08:13:16kivilahtioas such I fail to get OU's from anywhere else
#08:28:09tsberekivilahtio: Perhaps you have a problem with your org tree in regards to OU types. Or perhaps you should pop open the javascript console in the client and see if any obvious errors come up when you open one of the pages (clear everything, select the errors view, then load one of the pages that isn't working)
#08:31:00kivilahtiotsbere: do I need to be logged in as a eg_admin to open the js-console? The option is grayed out as a global admin.
#08:31:11kivilahtioas eg_admin I get no errors
#08:31:22kivilahtioconsidering OU's not being visible
#08:31:31tsberekivilahtio: New idea! Maybe you never assigned a work ou to the global admin?
#08:31:52kivilahtiotsbere: it should have one branch, but ill see
#08:37:03gmcharlt@later tell dbs agreed; I've merged and pushed the release notes branch
#08:37:03pinesol_green`gmcharlt: The operation succeeded.
#08:39:59darshan has quit IRC
#08:45:14kivilahtiotsbere: admin has a defined branch
#08:45:44kivilahtioas such (consortio -> system -> branch)
#08:47:26tsberekivilahtio: Did you do that via the permission editor and selecting a work ou, or are you looking at the home ou?
#08:47:32tspindler has joined #evergreen
#08:47:44kivilahtiotsbere: havent touched the permission editor
#08:47:49kivilahtiowill look into it
#08:49:14tsberekivilahtio: You will need to load the global admin user, edit their permissions, and add a work ou in that editor
#08:49:55dbs has joined #evergreen
#08:50:23kivilahtiook got that
#08:50:36tsbereOnce you have that done, see if the global admin user works ;)
#08:50:50kivilahtiotsbere: works like a charm
#08:56:28LBA has joined #evergreen
#08:58:06fortin has quit IRC
#09:00:09Dyrconaeeevil++
#09:01:05tater-laptop has joined #evergreen
#09:04:42akilsdonk has joined #evergreen
#09:06:12Meliss has joined #evergreen
#09:18:35collsk12 has joined #evergreen
#09:19:37jenny1 has joined #evergreen
#09:21:18collsk12Does anyone know why I'm getting "Attempt to update DB while not in a transaction" errors when attempting to import MARC records via the staff client?
#09:24:05kivilahtioTo create a provider I need a EDI account, and to create a EDI account I need a provider?
#09:24:30kivilahtioHow is this possible? Is this Sparta?
#09:24:44eeevilDyrcona: what did I do now? ;)
#09:25:11kmlussierkivilahtio: I don't think you need an EDI account to create a provider.
#09:25:46kivilahtiokmlussier: it asks for EDI default
#09:25:56kivilahtioI can't leave it empty
#09:27:12berickkivilahtio: the edi default does not have to have a value, though it may look that way in the UI
#09:27:52StephenGWills has joined #evergreen
#09:37:01sfortin has quit IRC
#09:37:15sfortin has joined #evergreen
#09:55:50csharpcollsk12: do you have access to the DB logs? that would mean that some command within a transaction failed
#09:56:40csharpand by "transaction" I mean postgresql database transaction (BEGIN; <set of commands>; COMMIT;)
#09:58:52collsk12csharp: I do.
#10:05:26collsk12csharp: http://pastebin.ca/2093128
#10:06:22collsk12That's the pg log. I don't see any BEGIN; statements. Maybe that's the problem?
#10:18:40tater-laptop has quit IRC
#10:22:27tater-laptop has joined #evergreen
#10:37:52csharpcollsk12: this is a key line: '2011-11-08 09:11:04 EST ERROR: current transaction is aborted, commands ignored until end of transaction block'
#10:38:27csharpif you can keep going backwards in the logs until before that message first appears, it should give you a clue as to what failed
#10:38:55csharp(probably a problem with the MARC/MARCXML of a particular record, methinks)
#10:40:47csharp learns chinese from the [OpenILS-feedback] spam
#10:40:59collsk12csharp: That's actually the entire log.
#10:41:15csharpcollsk12: are there no other postgres logs available?
#10:41:39collsk12csharp: Nope.
#10:41:46csharpblurgh
#10:42:34collsk12csharp: True. I can post the osrfsys.log or the open-ils.vandelay_stderr.log if you think that will help.
#10:42:41csharpwell, you could 'tail -f /var/log/<path-to-pg-logfile>' to watch what happens live and try again
#10:42:48csharpcollsk12: yes -it might
#10:43:01collsk12csharp: OK...brb.
#10:48:49collsk12csharp: http://pastebin.ca/2093133 - open-ils.vandelay_stderr.log
#10:49:29collsk12csharp: osrfsys.log will take a minute. I started to do something else and will have to rerun the import.
#10:49:37csharpok
#10:52:38collsk12csharp: http://pastebin.ca/2093136 - osrfsys.log
#10:58:51csharpcollsk12: I still can't see where the problem begins. It looks like these errors are caused by the fact that something earlier in the transaction failed
#10:59:12csharpcollsk12: can you chop up the file you're trying to import into smaller pieces for import?
#10:59:35collsk12csharp: I had tried that too.
#10:59:43csharpcollsk12: I think you might start with 3-4 records at a time, possibly even 1 at a time
#11:00:11csharpwell, can you do a single record and post the log for it?
#11:00:21collsk12csharp: I can.
#11:00:22bshumHuh, so receipt templates aren't really modified in the staff client side, but really only on the server side?
#11:00:50csharpbshum: in 1.6 and before, those were all done client side
#11:01:07csharp shudders at the thought that that has possibly changed
#11:01:08bshumcsharp: Figures, working off "old" assumptions :(
#11:01:25bshumI was just testing applying "new" defaults for receipt templates
#11:01:32bshumWrote them into our latest staff client build
#11:01:35bshumBut nothing changed
#11:01:38csharphmm
#11:01:44bshumSo I went to server side and made the same change
#11:01:47bshumAnd now everything shows up
#11:01:51csharphuh
#11:02:05bshumThis is 2.0 of course
#11:02:16bshumNot sure if it might have changed yet again in 2.1+
#11:02:17csharpphasefx: thoughts?^^
#11:02:55tsbereDefaults would be based on the server-side scripts in most cases. <_<
#11:03:20tsbereBecause most interfaces that deal with the templates (including the editor, I think) are based on the server side code, being loaded off of the server rather than locally
#11:03:26csharptsbere: but client-side modifications would override defaults?
#11:03:42tsberecsharp: Only those stored as part of the saved config
#11:04:12csharpso it's possible that the client is ignoring bshum's changes and pulling the defaults from the server
#11:04:14phasefxincidentally, some folks are using custom.js to enforce server-delivered templates
#11:04:38phasefxoverwriting whatever gets loaded from the local filesystem
#11:04:41tsbereIf you save templates they get written locally (and stored in the data stash, I think) and used client or server side. But if you, say, hit the "reset to defaults" button I think it uses the server side copy.
#11:04:56bshumThat seems to be the case, which is fine.
#11:05:03bshumJust needed to know where to edit "the defaults"
#11:05:25csharpok - that makes sense
#11:05:32tsberebshum: Take note, "the defaults" is usually server side decided, but sometimes client side. Like, say, the offline client, which is loaded locally.
#11:05:42pmplett has joined #evergreen
#11:05:43phasefxedit data.js to edit the defaults, or setup a custom.js to load an arbitrary templates file all the time
#11:05:50csharp was only alarmed by the idea that custom client-side templates might not work in 2.1
#11:06:05bshumtsbere: Right, we made a bunch of alterations to the offline components based on what we needed. Like making it say CT instead of GA for offline registration :)
#11:06:23phasefxwe should really make that dynamic :)
#11:06:24csharpbshum: there's other states out there? ;-)
#11:06:35pmplett is now known as pmpcat
#11:06:37phasefxa pref maybe
#11:06:39tsbereWe have some other changes to the offline client ourselves, like default checkout duration
#11:06:57bshumphasefx: The data.js file that I was poking at was /openils/var/web/xul/server/OpenILS/data.js (sound about right?)
#11:07:02csharpwe don't take to no yankee state defaults
#11:07:49phasefxbshum: right. Those are the defaults, but if a given workstation has ever used Save Locally even once, then they don't get used without the Restore to Defaults action (but I'm not sure if that's a temporary restore or a permanent restore)
#11:08:12tsberephasefx: I thought that if a *new* template was added it would show up, though.
#11:08:24phasefxtsbere: that sounds right
#11:09:09tsberephasefx: On a different note, you upgrade your ancient xulrunner yet? :P
#11:09:42phasefxgosh no, but I'm on a different machine today. But I thought doing an automatic upgrade would pull across new xulrunners
#11:10:27phasefxwe should probably give data.js an option for pulling in defaults from a remote file instead of hardcoding them in the javascript
#11:10:35tsbereIt should pull over new xulrunners....if your automatic upgrade isn't pulling across the "generic" client that has no bundled xulrunner. :P
#11:10:52phasefx has no idea how that works
#11:11:48tsberephasefx: If you have an autochannel.js with, say, "win" as the preference value in your default preferences folder you should be getting xulrunner updates. If you don't have that file at all you are getting generic updates.
#11:12:21phasefxI've never touched such a file, so I'm doing whatever stock does
#11:12:35tsberestock does things differently depending on how you installed.
#11:12:36tsbere<_<
#11:12:49tsbereautochannel could be lin instead of win (I think it was) for a linux with bundled stuff
#11:12:54phasefxstill haven't touched an autochannel.js
#11:13:04kivilahtioOk, I have ordered 4 items via the acquisitons workflow. I have just received the order and I see it has been charged from the fund. Why I cannot find those items via OPAC or catalog search? What is the missing link?
#11:13:30tsberephasefx: Not saying you should have. But depending on how your client came into existing with updates it may or may not be there.
#11:13:31kivilahtioDo I need to set the items shelved or ?
#11:13:32phasefxtsbere: I see that file in my client/ folder, says win
#11:13:56tsberephasefx: Then you should (when doing the make updates-client step) be getting updated xulrunners too.
#11:15:02phasefxI do this when I want to update the local client, make STAFF_CLIENT_VERSION=$@ WEBDIR=/openils/var/web/ rigbeta devbuild server-xul win-updates-client && cp lang.dtd /openils/var/web/opac/locale/en-US/lang.dtd && rm /openils/var/web/xul/current && ln -s /openils/var/web/xul/$@ /openils/var/web/xul/current && echo Finis-Done-End
#11:15:41phasefxI currently have xulrunner-3.6.22.en-US.win32.zip sitting in my staff_client directory
#11:18:12collsk12csharp: http://pastebin.ca/2093139 - osrfsys.log; http://pastebin.ca/2093140 - gateway.log -> These are the only two logs that are generated, when I try to import 1 record.
#11:20:22tsberephasefx: So long as you aren't forcing use of a different xulrunner (via the application.ini passing and such) you should be getting updated versions, then.
#11:21:41phasefxdon't recall doing any such mucking
#11:22:03tsberePre-my changes installs did that automatically, though.
#11:22:04phasefx checks to see what his local machine is running (just updated it, so it should be 3.6.22)
#11:22:44phasefx1.9.2.22
#11:23:07tsbereBetter than the .12 you reported yesterday
#11:23:10phasefx tries a pristine install into another folder from manualupdate.html
#11:24:01tsbere has 1.9.2.22 himself
#11:24:05kmlussierkivilahtio: Did you load bibs and items?
#11:24:31tsberephasefx: They changed the package version, but not the internal version, when they went to 3.x.....so that application.ini files wouldn't break
#11:25:17phasefxcool deal
#11:25:34phasefxnot sure why the other machine had an old version, but I can play with it again tomorrow
#11:27:01kivilahtiokmlussier: it says (item status - In process)
#11:27:01tsberephasefx: I suppose one option for dealing with the whitespace issue and older xulrunners would be to wrap it in an "if(XML)" block.....
#11:27:14csharpcollsk12: "ERROR: invalid XPath expression" is the error you'd need to start with
#11:27:15kivilahtioi can search for bib's and it finds the correct bib
#11:27:33csharpcollsk12: I don't know how to fix it - perhaps someone else in the channel can help
#11:27:36phasefxtsbere: I can't even wrap it in a try/catch
#11:28:06kivilahtiokmlussier: trying to find from where to mark it as shelved, or how it is marked as searcheable.
#11:28:13kmlussierkivilahtio: If it's inprocess, it won't show up in a public catalog search. If you check it in, it should display.
#11:28:37phasefxand E4X does exist for the older xulrunners, and the ignoreWhitespace thing even runs without error in some contexts.. but in my_init, syntax error :(
#11:31:21kivilahtiokmlussier: I checked in the items to a different branch than where they were ordered to, now they are in transit. So I need to check them in while the staff user is in the correct receiving branch. Can I add multiple branches to my admin user and change the active branch on the fly?
#11:32:58kmlussierkivilahtio: No, I think your workstation needs to be logged into the correct branch to check it into that branch.
#11:33:19tsberephasefx: I put it outside of my_init, found it worked as good/better there?
#11:33:35tsberephasefx: But of course you have my branch to look at what I did ;)
#11:42:28dbs has quit IRC
#11:53:38b_bonnerkivilahtio: you can register multiple workstations on the same PC by editing the hosts file : http://open-ils.org/dokuwiki/doku.php?id=scratchpad:multiple_registrations_hack_with_hosts_file
#11:54:15tsbereb_bonner: Ewwww
#11:54:34tsbereFor more recent versions of master you can just add /something to the end of the hostname ;)
#11:54:50b_bonnerah, good to know.
#11:55:06tsbere forgets if phasefx pushed that somewhere else.....assuming phasefx was the one who reviewed it
#11:57:53sfortin has quit IRC
#11:59:21phasefxpretty sure that got in
#11:59:42phasefxthough maybe it'll be 2.2
#12:00:10bshumMeeting time! (maybe)
#12:00:16b_bonnerI can check my system, how exactly would enter it?
#12:02:17bshumHere's the agenda - http://evergreen-ils.org/dokuwiki/doku.php?id=dev:meetings:2011-11-08
#12:02:22dbs_f16 has joined #evergreen
#12:02:29bshumCan wait till more folks show up
#12:02:57dbs_f16 is Dan Scott, running on a secondary Fedora 16 workstation while his laptop upgrades (or doesn't....)
#12:03:03tsbereb_bonner: host.example.ext/some_id instead of host.exmaple.ext
#12:03:04csharpdbs_f16: release day!
#12:03:11phasefxdbs_f16: man, I thought you were in the cockpit :)
#12:03:20csharpphasefx++
#12:03:30slipscomb has joined #evergreen
#12:03:44phasefxb_bonner: if that doesn't work, you can always add -profileManager to the end of the target in your staff client's shortcut
#12:03:56phasefxcreate a different profile for each workstation
#12:04:40b_bonnerit didn't work on our 2.1-ish build. host file works fine for my troubleshooting, but the profile manager might be useful for some other cases. thanks.
#12:04:45dbs_f16 changes topic to "Dev meeting agenda: http://ur1.ca/5owmy | Welcome to the #evergreen library system channel! | We are publicly logged. | Large pastes at http://pastebin.ca or http://paste.lisp.org/new/evergreen or something like that"
#12:04:51tsberelooks like that didn't go into 2.1 (based on looking for things in rel_2_1 in git)
#12:06:37dbs_f16 can take minutes
#12:07:02jlamoscsharp
#12:07:21csharpjlamos: yo
#12:07:45jlamoscheck im por favor
#12:08:29jlamos resumes lurking in his cave
#12:08:53bshumdbs_f16: I can run us through what I put on the agenda, if we have folks for the meeting.
#12:09:11bshumWho's around? Should we hold a meeting now?
#12:09:16phasefx is here
#12:09:25tsbereI am around. Ish. May go nuke a pizza.
#12:09:45dbwells is here, and also nuking pizza
#12:09:58dbs_f16 has cold pizza in his backpack that he forgot about - yay
#12:10:50eeevil appears, having gathered coffee
#12:11:50berick is present
#12:11:55bshumOkay
#12:12:16bshumDid we have any past action items to review? (it's been awhile since our last meetings)
#12:12:27bshumIf not, we can go straight to the new business
#12:12:42tsbereMaybe we need a new action item: Skip less meetings? ;)
#12:13:06bshumHeh, tsbere++
#12:13:13bshumOkay, new business then.
#12:13:22bshum4.I. Evergreen 2.1.0 official announcement
#12:13:43bshumI saw that the release notes got merged in. Should we get a volunteer to write this up even though it's been out for awhile?
#12:13:57bshumOr wait to announce with 2.1.1?
#12:15:49DyrconaWhy not wait if 2.1.1 is better?
#12:16:24dbs_f16Because we don't know when 2.1.1 will be out?
#12:16:30dbs_f16And of course 2.1.1 will be better :)
#12:17:29dbs_f16An announce of 2.1.0a should probably be accompanied by some "Known issues", like OpenLibrary cover art not working, and umpteen other issues, that fortunately can get rolled right into the 2.1.1 release notes :)
#12:18:25DyrconaThat's why I think waiting might be better.
#12:18:33dbs_f16(release notes are in RELEASE_NOTES.txt in the root directory of rel_2_1 if you missed the merge, or at http://evergreen-ils.org/documentation/release/RELEASE_NOTES_2_1_0.html)
#12:18:54dbs_f16So let's jump ahead to "When the heck is 2.1.1 going to be released?"?
#12:19:31bshumSounds good.
#12:19:58eeevil looks at the bug lists
#12:20:24eeevillooks like no blockers on the milestone-targeted bugs...
#12:20:50eeevilsame for series
#12:20:54eeevilso ... now?
#12:21:08Dyrconasure. now works for me. :)
#12:21:24eeeviltsbere: since I failed so spectacularly at cutting 2.1.0 with your new tool, would you like to do the honors for 2.1.1?
#12:21:53dbs_f16series = https://bugs.launchpad.net/evergreen/2.1/+bugs yeah? if so, I concur that now sounds good
#12:22:21tsbereeeevil: Apparently *I* missed something not going right in the readme when I cut 2.1.0 myself. But if you want, I can do so.
#12:22:42dbs_f16milestone = https://launchpad.net/evergreen/+milestone/2.1.1
#12:23:20dbs_f16tsbere: oh yeah, speaking of README, now we're including links to the download location for the release - another tweak to your uber-script?
#12:23:39tsbere can look into that. AFTER his pizza ;)
#12:24:14dbs_f16 has finished his pizza without severe spillage on self
#12:24:17dbs_f16small victories
#12:27:18eeevilok ... so ... this week-ish? today-ish?
#12:27:43dbs_f16 writes up: *ACTION*: tsbere to tweak release-cutting script to accommodate README changes (pointer to download location) / *ACTION*: tsbere to cut 2.1.1 release for testing and _severe_ criticism real soon now
#12:28:06tsbereI am willing to go with packaging today or later this week (if I do the packaging). We may want to do a last "any pullrequests we want to throw in quick?" pass though ;)
#12:29:41Dyrcona has bad feelings about "quick" pullrequests.
#12:29:52dbs_f16Serves as a good reminder to get bugs targeted to the right series / milestone ...
#12:30:11eeevilindeed
#12:30:12bshumSpeaking of that, we're supposed to continue nominating bugs for series targeting right?
#12:30:22bshum(for those of us who can't target but still tinker with bugs)
#12:30:31Dyrconabshum: yep.
#12:30:36eeevilon that, https://bugs.launchpad.net/evergreen/2.1/+bug/849447 looks possible for 2.1.1
#12:30:43Dyrcona has been remiss in bug wrangling.
#12:31:07eeevilhttps://bugs.launchpad.net/evergreen/2.1/+bug/621448 isn't a pullrequest, but would be good to do
#12:31:47dbs_f16bshum: I think we had discussed on and off defining a release team separate from the project leads - so Bug Wranglers is an open group, but Release Team folks with ability to target series/milestones would need to be approved
#12:32:29Dyrconadbs_f16: bug wranglers can nominate for a series, but release team has to approve it.
#12:32:29dbs_f16gmcharlt seems most familiar with 849447
#12:32:44dbs_f16Dyrcona: right. and right now release team = project leads
#12:33:43dbs_f16So the idea would be that release team would be broader than project leads
#12:34:33eeevilfwiw, I'm currently poking at 621448
#12:34:44Dyrconadbs_f16: I didn't get that from bshum's question. I just got that he wanted to know if he should bother with nominating bugs to a series or not.
#12:35:21Dyrconaunless I missed something else.
#12:35:24dbs_f16Dyrcona: Right, I jumped ahead to "We probably should give more people the power to actually target bugs to series, not just nominate them"
#12:35:31bshumdbs_f16: Dyrcona: Well it's nice to know that you guys might be thinking about expanding the release team members. Adding series nominations hasn't always led to new series targets in my past experience.
#12:36:24Dyrconabshum: I have to admit that I am sometimes erratic in that when I accept or decline nominations.
#12:36:47DyrconaDon't know if others are more consistent.
#12:38:48_bott_ has quit IRC
#12:38:56bshumThat's fair enough, I have been remiss myself in adding series nominations.
#12:39:58_bott_ has joined #evergreen
#12:40:45Dyrconaso, where are we on the agenda?
#12:40:58berick marks 881451 w/ 2.1.1 milestone (though it's not a showstopper)
#12:41:04eeevilfwiw, agreed on more people targetting ... we don't get notified, IIRC, when a nomination happens (and nothing else changes)
#12:41:48bshumErm, next on the list was 2.2 alpha
#12:45:23kmlussierSince nobody else is speaking up, I'll jump in with a comment that we would love to see a soon-ish release of 2.2 alpha.
#12:45:32kmlussier is Kathy Lussier, MassLNC
#12:46:42DyrconaI've been telling people the goal is this month, but Kathy is most of those people. :)
#12:47:04berick+1 -- if nothing else, it helps to keep up the momentum
#12:47:05dbs_f16Might dovetail with the Conifer TPAC hack session in early December: good to have a common artifact to rally around
#12:47:17dbs_f162.1.1 this week, 2.2 alpha next week?
#12:47:21gmcharlt+1 # momentum
#12:47:23bshum+1
#12:47:31Dyrcona+1
#12:48:06kmlussier+1
#12:48:18tspindler+1
#12:48:50tspindler*tspindler is Tim Spindler (C/W MARS) and voting on this for purely self interest
#12:48:57bshumOkay, with that, the only things left is 1.6 end of life (which I think moodaepo will be writing something about based on the community meeting we had last Friday) and the GSoC wrapup by dbs.
#12:49:58moodaepo will run it by here before posting
#12:50:35dbs_f16moodaepo++
#12:50:50bshummoodaepo++
#12:50:57dbs_f16GSoC _Mentorship Summit_ report, not a wrapup on the GSoC :)
#12:51:19bshumRight, report :)
#12:52:10dbs_f16The linked email in the agenda contains the high-level summary, I'm happy to provide further details if anyone's interested.
#12:53:13dbs_f16I do think prioritizing review & committing of patches from new contributors would be a good direction to go...
#12:55:15bshumCool, well, next up was the announcement (which dbs has kindly shared already) that Conifer tpac dev day for Dec 8 and 9. Poke dbs for more info on that.
#12:55:27bshumAnything else before we adjourn the meeting?
#12:55:44eeevil(+1 on 2.2 alpha asap!)
#12:55:50dbs_f16 hopes to poke at dbwells' LDAP work real soon
#12:55:54bshumNote, the next dev meeting date is wrong on the agenda, that ought to be two weeks from now, so Nov 22?
#12:56:17eeevilI'll just throw in that the more pullrequest pulling everyone can do, the better! </captain-obvious>
#12:56:22dbs_f16 glares at F16 preupgrade which appears to be unwell
#12:56:45bshumAlright, thanks everyone. We'll try to have more regular meetings again. ;)
#12:57:03dbs_f16Should we formalize the discussion we had in channel last week about sign-off by non-committers being very, very valuable?
#12:57:29bshumOo
#12:57:35eeevildbs_f16++
#12:57:57dbs_f16That is, if bshum pushes a branch out for review and Dyrcona signs off on it, that should be a strong pullrequest request...
#12:59:01phasefx+1
#12:59:04Dyrcona+1
#12:59:14moodaepo+1
#12:59:16moodaepobshum++
#13:03:59dbs_f16bshum++
#13:04:23bshum+1 to dbs' point (very belated, gah)
#13:04:25dbs_f16 should use "not-yet-committer" or "soon-to-be-committer"
#13:05:41tsbere"trusted tester"? "trusted community member"?
#13:05:44moodaepodbs_f16++
#13:05:57Dyrcona"happy-not-to-be-committer"
#13:06:00tsbereheh
#13:06:03bshum dislikes "trusted"
#13:07:46dbs_f16Dyrcona++
#13:09:20moodaepoDyrcona++
#13:09:51moodaepo spreading karma, staying away from actual conversation
#13:11:55eeevilgmcharlt: shall I steal LP#849447 from you?
#13:14:07eeevil(silence equals assent :) )
#13:19:28pmpcat has quit IRC
#13:19:44dbs_f16 is now known as dbs
#13:19:52dbs has joined #evergreen
#13:19:57DyrconaIf we're looking for new features to roll into 2.2alpha, I wonder if anyone would be interested in looking at my PhoneList branch?
#13:20:22tsbereWhole pile o branches that should be looked at before we push 2.2 out ;)
#13:20:25DyrconaWe've been using the holds portion of it since our migration in May.
#13:20:57Dyrconatrue dat. I should get to work on dbs' suggestions for the no can haz copiez branch.
#13:21:34tsbereOn a different note, I don't see obvious "here are download URLs" chunks in the README. Am I being blind today, or are they hiding in a branch somewhere waiting for merge?
#13:21:41tsbere pokes dbs about that ^^^
#13:21:55eeevilnote, alpha is /not/ feature locked... it's intended to get non-devs (or devs that don't focus on this IRC channel ;) ) to test what's coming
#13:22:08dbs pokes tsbere back
#13:22:09eeevilwe can and will continue pushing stuff in until beta
#13:22:45tsberedbs: Am I *adding* an entire block, or do we plan on having basic language there ahead of time for me to update? :P
#13:22:46dbshttp://evergreen-ils.org/documentation/install/README_2_1_0.html#_preamble_getting_an_evergreen_official_release_tarball
#13:23:48dbssee also http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=648e7186d81cc9d658d6459bdff404e68c38c422
#13:24:09dbsif you're looking in master, it's not there (because it's nonsensical in master)
#13:24:36tsbereSee, that is a problem. How will we remember to put it into, say, 2.2? Or 2.3? etc. :P
#13:24:45tsbere can put a "add this in when we change something away from master" though
#13:24:51eeevil(sorry, folks, for the merge-commit noise from me in 2.1 ... didn't pull before picking, and then rebase -i was failing when I tried to reorder mine to the end)
#13:25:13pmpcat has joined #evergreen
#13:25:28dbs sends eeevil to remedial git class
#13:26:00Meliss has quit IRC
#13:27:07dbstsbere: oh, I like the idea of the release packaging script just injecting the block of text into the bundled README; avoids the weirdness of saying "you can download the source from..." when you already have the source checked out
#13:27:52Dyrcona can commiserate with eeevil. had some weird rebase behavior recently.
#13:28:18tsberedbs: Perhaps master should have a different block... "Preamble: Introduction to Git" ;)
#13:38:09dbsgmcharlt++ # using perlcritic
#13:45:19_bott_ has quit IRC
#13:45:21kbeswick has quit IRC
#13:46:19_bott_ has joined #evergreen
#13:46:19kbeswick has joined #evergreen
#13:55:39eeevildbs: if you can find a git-rebase-unbreaking-for-dummies, point me at it
#13:58:31dbseeevil: a massive encyclopedia like that would need some details to find appropriate entry points
#14:00:33phasefxneed a git undo
#14:02:16tsberegit reset --hard ;)
#14:02:23gmcharltyep
#14:02:46gmcharltand if you've done something more complicated, git reflog can you and indication of where to reset time to
#14:06:30eeevilI needed git reflog, it seems, but, bah
#14:20:05sfortin has joined #evergreen
#14:22:57tspindler has quit IRC
#14:39:23tsbere tests an updated make_release script that adds or updates the preamble block there.....currently on "update" test, will do "add" test when locales finish building on the "update" test
#14:39:23collsk12Can anyone walk me through downgrading Evergreen from 2.1.0a to 2.0?
#14:41:13tsberecollsk12: Step 1: Restore from your backup. Step 2: Start services.
#14:41:32phasefxcollsk12: you could maybe use apg_diff against your 2.1.0.a schema and a 2.0 schema, and make a downgrade script from that, but it would not be fun
#14:43:36collsk12tsbere: Thanks, but I don't have a backup. I started fresh at 2.1.0a, but I would like to move back to 2.0.x.
#14:44:32tsberecollsk12: Blow out the system and start fresh with the 2.0 instructions?
#14:45:50dbswow. downgrading? you're in no-person's-land, you get to write the walkthrough guide!
#14:46:04collsk12tsbere: Is there an easy way to do this?
#14:48:03Dyrconacollsk12: If it is on a machine dedicated to testing Evergreen, I would start over by reinstalling the operating system. This way, you know eveything is gone.
#14:48:26tsberecollsk12: If you don't want to blow out the OS you could try: Stop all services, drop the evergreen database, delete /openils
#14:49:13phasefxcollsk12: it may be easier to fix whatever it is about 2.1.0a you don't like
#14:49:36Dyrconacollsk12: or wait for 2.1.1 later this week.
#14:50:31collsk12 has quit IRC
#14:54:16enhancin has joined #evergreen
#14:55:08enhancinI'm having trouble with electronic records...I'm not sure how to make them show up in the public OPAC area. I'm also not really a librarian....
#14:56:11tsbereenhancin: For troubleshooting evergreen I think not being a librarian is a plus, not a downside ;)
#14:56:27enhancinha. yeah, i guess that's why I'm the one doing it and not the librarian =P
#14:56:54dbsenhancin: can you point at an example record?
#14:56:56tsbereThere are two options. Transcendant bib source and some kind of "add these tags to the MARC" that I can never remember the number for and have no understanding of.
#14:57:37dbs856, indicator 1=4, indicator 2=0, add one or more subfield $9 with the shortname of the library where the record is supposed to be visible
#14:58:17enhancinI'm not sure if I could point you to an example record, I can only get them to show up when I'm in thr staff client.
#14:58:46tsbere defers to dbs and gets back to testing his changes to his make release script ;)
#14:59:03dbsenhancin: you can give us the record ID for the example record and the URL for the public site
#15:00:27enhancin21916 is the record ID and the public site is at http://209.236.66.49/opac/en-US/skin/default/xml/index.xml (no domain pointed to it yet)
#15:00:33dbsDyrcona: while you're poking at broken fm_IDL.xml, you might want to check out / sign-off (hint!) the fix-idlval branch I bugged a week or so back and get a comprehensive list of what the idlval util thinks is broken
#15:01:52dbshttp://209.236.66.49/opac/en-US/skin/default/xml/rdetail.xml?r=21916&d=0&hc=1&adv=21916&rt=tcn
#15:02:32dbsneeds a $9 subfield in that 856
#15:03:09dbsprobably with a value of library shortname "LIB"
#15:03:25dbsadd that and Bob should become your uncle
#15:04:06pmpcat has quit IRC
#15:04:41enhancinhm. alright. would I be able to run a script to edit all those MARC records in the database and then restart and autogen.sh and it should work out?
#15:05:31dbsenhancin: no need to restart
#15:05:39enhancinalright, cool
#15:07:03tsbereeeevil: If I am going to be generating a release or two, should my key get installed somewhere so I can upload them more directly?
#15:09:28phasefxtsbere: your marc space branch works with the latest xulrunner. I'm not sure how to handle older xulrunners. We could check the version and eval the statement as a string maybe if conditions are met
#15:10:03tsberephasefx: Or insist people start using more recent xulrunners ;)
#15:10:09phasefxor just let it break horribly and do that, yeah :)
#15:10:28phasefxbetter to break than to allow folks to mangle more marc records
#15:10:41phasefxsound sane to other folks?
#15:10:45Dyrcona+1
#15:10:55eeeviltsbere: csharp should be able (IIRC) to do the server dance for you, if you don't have access to the main web server
#15:11:04DyrconaWhich reminds me, I should add a task to my list to write something to fix some of these.
#15:11:21tsbereeeevil: To my knowledge I do not have access to it.
#15:11:23DyrconaWe see it mainly in 008 because of our templates.
#15:11:36Dyrconadbs: I'm looking at fix-idlval right now.
#15:11:38dbsenhancin: also, you could just run a regex against the contents of biblio.record_entry
#15:11:42tsberephasefx: Probably not a bad idea. "We can't fix this in older xulrunners, so you will need a newer one..."
#15:11:46dbsDyrcona++
#15:11:53enhancinthat's what I planned on doing : )
#15:12:00enhancinI want to make sure I get this one to work first though
#15:13:08dbsenhancin: http://209.236.66.49/opac/en-US/skin/default/xml/rresult.xml?rt=keyword&tp=keyword&t=social%20voice&ft=&l=2&d=1&f= works for me now!
#15:13:21phasefx commits the marc fix
#15:13:38enhancinyep i just changed it to LIB. I put it to the overall library but I guess it had to be down in one of the branch libraries
#15:16:28bshum likes the purple opac colors in enhancin's catalog
#15:18:49Dyrconaidlval works for me in master, so I'll sign off on that branch.
#15:19:11enhancinin the MARC in xml format (in the database) is there anything else that would use subfield code="y" besides that 856 entry?
#15:21:21enhancins/(<subfield code="y">.+?<\/subfield>)(<\/datafield>)/\1<subfield code="9">LIB<\/subfield>\2/
#15:24:20jamesrf650 uses subfield y
#15:25:33enhancinso this would work better I guess? is it possible that 650 can come after or does it organize them numerically? s/(<datafield tag="856".+?<subfield code="y">.+?</subfield>)(</datafield>)/\1<subfield code="9">LIB</subfield>\2/
#15:26:12dbsenhancin: fields are generally supposed to be ordered numerically, but practice trumps theory / specs
#15:26:33Dyrconaoften finds fields in MARC records out of order.
#15:26:51enhancinright...hm. okay.
#15:26:58jamesrfmaybe s/<subfield code="y">LIB</subfield>/<subfield code="9">LIB</subfield>/g
#15:27:10enhancinthe y has different values
#15:27:17enhancinsuch as Primary Search, Full Text, etc
#15:29:00tsberedbs: :D My make release now adds (or updates, if it is already there) the download preamble when building releases (but only when building releases)
#15:29:43dbstsbere++
#15:57:27edoceo has joined #evergreen
#15:58:06pmpcat has joined #evergreen
#15:58:56edoceo has quit IRC
#16:13:48enhancin has quit IRC
#16:24:55sfortin has quit IRC
#16:39:53kmlussier has quit IRC
#17:00:57akilsdonk has quit IRC
#17:06:40LBA has quit IRC
#17:07:58DyrconaI am almost sorry that I ran this: http://pastebin.com/r69fYkJw
#17:09:11bshumHuh... how does the same item end up on the pull list more than once? I didn't think you could have holds target the same current_copy
#17:10:58tsberebshum: Hold targeter running while the pull list is generating
#17:11:09bshumHmm
#17:11:10slipscomb has left #evergreen
#17:11:19tsberebshum: Thus when hold A was grabbed it targeted the copy. Then the targeter moved stuff around and by the time it hit hold B it was also targeting said copy...
#17:11:45bshumWe ended up with 3 holds for the same target
#17:11:57bshumI don't know when the hold targeter ran though
#17:12:08bshumThey appear to be really old holds
#17:12:27bshumThat are now targeting brand new items that were just added to a holdable status last night
#17:12:58tsbereok
#17:14:15bshumEh, we'll move them along and see what happens in the next couple days
#17:14:23bshumWeird flukes I guess
#17:15:18DyrconaHold targeter fun.
#17:15:43DyrconaFun little factoid about that script I posted.
#17:16:04DyrconaI ran it at ~ 5:00 pm and got 6,320 results.
#17:16:30DyrconaI modified it to add the bre.edit_date to the output and ran it at 5:05 pm.
#17:16:37DyrconaI got 6,321 results.
#17:17:02DyrconaIn 5 minutes, someone added or modified a record to have a bad 008.
#17:17:40dbsDyrcona: luckily you guys haven't been live that long
#17:17:54dbs:)
#17:18:30Dyrcona;)
#17:18:48Dyrconagot one record migrated from the legacy ILS with an 008 that is 347 characters long.
#17:19:23Dyrcona looks around for youdonotexist.
#17:19:41Dyrconadang. never here when you want him to be.... ;)
#17:20:19bshumHeh
#17:21:39dbsDyrcona++
#17:25:17tecoripa has joined #evergreen
#17:26:26tecoripahello, I have a few questions about the evergreen actor schema, and the stat_cat tables
#17:29:54bshumtecoripa: Might be catching folks at the end of the work day here, but feel free to ask and someone might read the backlog later to answer you. Alternatively the mailing lists are always a good option.
#17:30:05bshum is on his way out of the door too
#17:30:19tecoripaok, thanks, bshum. I'll post here, and send to the dev list
#17:30:22tecoripahave a good evening
#17:30:25Dyrconabye, y'all!
#17:30:28Dyrcona has quit IRC
#17:31:49StephenGWills has quit IRC
#17:33:25tecoripaso: I see that the org_unit table has a pointer to a parent org_unit. This tells me that hierarchies of org_untis can be arbitrarily deep, and that children point to their parents, but parents do not point to their children. Is this in fact the case
#17:33:42tecoripa?
#17:34:42tecoripaand is there anywhere in the db schema and/or the code where consortia are distinguished from a library system, which is in turn distinguished from a branch?
#17:35:21tecoripaare all they just org_units, consortia being a parent of system libraries, and a system library being a parent of branches?
#17:35:56tecoripathanks, bye
#17:36:02tecoripa has left #evergreen
#17:47:26dbs has quit IRC
#17:56:36tater-laptop has quit IRC
#17:59:44jenny1 has left #evergreen
#18:08:33pmpcat has quit IRC
#19:35:10akilsdonk has joined #evergreen
#20:42:39wlayton has joined #evergreen
#20:43:16wlayton changes topic to "Welcome to the #evergreen library system channel! | We are publicly logged. | Large pastes at http://pastebin.ca or http://paste.lisp.org/new/evergreen or something like that"
#21:07:35wlayton has quit IRC
#21:19:25wlayton has joined #evergreen
#22:06:37akilsdonk has quit IRC
#22:27:08tfaile has quit IRC
#22:27:13tfaile has joined #evergreen
#22:27:29jamesrf has quit IRC
#22:28:05jamesrf has joined #evergreen
#22:33:14eeevilbshum: re multiple holds targetting one copy -- C-type holds? they just point at the copy, period, IIRC
#22:36:12wlayton has quit IRC
#23:13:18jlamos has quit IRC
#23:15:07jlamos has joined #evergreen
#23:16:06jlamos has quit IRC
#23:18:46dbs has joined #evergreen
< Monday, November 7th, 2011Raw Log FileWednesday, November 9th, 2011 >