Evergreen ILS Website

IRC log for #evergreen, 2014-12-31

| Channels | #evergreen index | Today | | Search | Google Search | Plain-Text | summary | Join Webchat

All times shown according to the server's local time.

Time Nick Message
01:19 RBecker joined #evergreen
01:44 bmills joined #evergreen
06:35 BigRig joined #evergreen
07:32 rjackson-isl joined #evergreen
07:34 pie_ joined #evergreen
08:11 pie_ does anyone actually use usemarcon?
08:28 mrpeters joined #evergreen
08:33 mmorgan joined #evergreen
08:33 Dyrcona joined #evergreen
08:42 abowling joined #evergreen
08:44 collum joined #evergreen
08:47 bshum pie_: I have not used that software before. Perhaps others have though. Are you sure you're asking in the right place? Or you're just wondering if other Evergreeners may have used it before?
08:48 bshum From a cursory glance, it seems to be some format conversion software for MARC formats?
08:50 pie_ i just joined the first library software related irc channel i could find
08:50 pie_ since im lacking in better ideas
08:51 pie_ yeah thats what it is
08:57 mrpeters I know this is a VERY broad question, but has anyone taken the time to document how KPAC search results are sorted?  Since the "Sort by Relevance, Title, etc" option isn't available within KPAC I'm just curious about which strategy (if any) is being used.
08:58 bshum pie_: Ah, okay. Well, maybe there'll be more activity as the day progresses.
08:58 bshum pie_: Folks in this channel tend to be most active during US east coast working hours
08:59 bshum pie_: That said, it is the day before 2015 and lots of folks may be away on holiday too.
09:00 bshum You may find more library software folks in channels like #code4lib (freenode) or #koha (on oftc network)
09:01 mrpeters Re: KPAC search sorting, I do know that there is a limit to audience(a, b, c, j) but am unsure of how those are sorted after that limiter
09:02 mrpeters which, if i recall is the 008 tag MARC code
09:02 bshum mrpeters: I'm almost certain that the order is relevance based
09:03 mrpeters OK, so the same as the default strategy for the regular OPAC?
09:03 bshum That would be my expectation.
09:04 mrpeters easy way to test, i suppose -- run the same search string in tpac and limit the audiences :)
09:04 bshum I'm sure you could try altering it by tacking on more variables to the URL.
09:04 mrpeters nope, you nailed it
09:04 pie_ bshum: heh. thanks
09:05 mrpeters NASCAR audience(a,b,c,j) depth(0) in TPAC produces identical results to a KPAC search for keyword "NASCAR"
09:05 bshum mrpeters: you should be able to take a url and change k to o in it and get the equivalent search
09:05 mrpeters right, that makes sense
09:07 pie_ MARC stuff is complex :C
09:07 mrpeters it gets easier pie!
09:08 mrpeters ive learned a lot over the last 7 years
09:08 mrpeters and i had no library experience
09:08 pie_ i have untill saturday, then my winter break ends ;P
09:09 pie_ theres a lot of stuff on the internet, but its really hard to navigate and find a solution to what youre doing, at least it seems like that so fat
09:09 pie_ *far
10:02 Guest5683 joined #evergreen
10:03 pie_ joined #evergreen
10:14 RBecker_ joined #evergreen
10:15 bshum @later tell tsbere Hmm, "Config variable ${SERVER_PORT} is not defined" is ending up in my apache warnings whenever I restart on apache 2.4 with Ubuntu Trusty. Test converting our rewrite maps/rules from 2.2 to 2.4.
10:15 pinesol_green bshum: The operation succeeded.
10:21 Dyrcona joined #evergreen
10:31 RBecker joined #evergreen
10:50 dkyle joined #evergreen
11:00 jboyer-isl joined #evergreen
11:16 mmorgan Anyone around who has worked with the Merge Monograph Parts functionality in 2.7?
11:17 bshum mmorgan: I tested it
11:17 bshum And pushed those commits to master during 2.7
11:17 bshum What seems to be up?
11:21 bshum parts-- # asking about parts is dangerous
11:21 mmorgan joined #evergreen
11:25 mmorgan sorry for dropping out, kicked the power strip :-(
11:25 bshum mmorgan: I blamed it on parts, don't worry about it :)
11:26 bshum mmorgan: What do you want to know about merging parts?
11:26 bshum I was asking Mary and she's used it since we upgraded to 2.7
11:26 bshum Though she's found a quirk apparently, I'm asking for more details
11:26 mmorgan when I merged 2 parts on our training system, the item with the bad part lost it's part.
11:26 bshum Yeah, that's what she just told me
11:26 bshum That's not how it's supposed to work
11:27 mmorgan didn't think so ;-)
11:27 bshum Or at least, not according to the original spec in the bug
11:27 bshum https://bugs.launchpad.net/evergreen/+bug/1099979
11:27 pinesol_green Launchpad bug 1099979 in Evergreen "Merge Parts" (affected: 4, heat: 18) [Wishlist,Fix released]
11:28 bshum Seems like we may have a new bug then
11:28 mmorgan seems so :-(
11:28 bshum Something for DPearl to cut his teeth on in the new year maybe ;)
11:28 mmorgan I was so excited to merge my first part in the client!
11:30 mmorgan it worked beautifully - well, except for the bug.
11:30 mmorgan I'll open a launchpad bug.
11:31 bshum mmorgan++
11:33 mllewellyn joined #evergreen
11:42 nhilton joined #evergreen
11:53 mmorgan Just opened lp 1406786 if anyone is interested in commenting/confirming
11:53 pinesol_green Launchpad bug 1406786 in Evergreen "Merge parts functionality fails to preserve copy_part mapping" (affected: 1, heat: 6) [Undecided,New] https://launchpad.net/bugs/1406786
12:02 jeff mmorgan++
12:02 jihpringle joined #evergreen
12:03 mmorgan That's it for me for the year. Happy New Year, everyone!
12:04 mmorgan left #evergreen
12:26 * bshum grumbles at bug 1406788
12:26 pinesol_green Launchpad bug 1406788 in Evergreen "KPAC Login Redirect Issue" (affected: 1, heat: 6) [Undecided,New] https://launchpad.net/bugs/1406788
12:27 bshum abowling++ # I'll test that one cause I probably broke it when I merged redirect changes a couple releases back
12:31 bshum abowling: On the other hand, initial testing doesn't seem to redirect it where I think it should.  I'll poke at it some more :)
12:32 jeff heh. i was just looking at that. :-)
12:32 bshum jeff: Heh
12:52 jboyer-isl joined #evergreen
12:55 nhilton joined #evergreen
13:02 RBecker joined #evergreen
13:56 abowling thanks, bshum
13:59 Dyrcona Whee!
13:59 * Dyrcona waits.
14:04 * Dyrcona thinks maybe zgrep would be faster.
14:11 * Dyrcona gives up and tries zgrep.
14:46 * jeff finds himself wondering "faster than what?"
14:50 Dyrcona Faster than zless honkin_file.gz, and then /search term
14:50 Dyrcona zgrep 'search term' honkin_file.gz | less is faster
14:53 * bshum wanders off, see y'all in 2015 :)
14:57 gsams joined #evergreen
14:57 Dyrcona Happy New Year!
15:25 jboyer-isl left #evergreen
15:52 Dyrcona Well, I found 1 thing today.
15:53 Dyrcona Just retrieving a patron by barcode in the staff client apparently calls open-ils.actor.user.checke​d_out.count.authoritative 6 times.
15:53 Dyrcona Which in turn calls open-ils.cstore.direct.acti​on.open_circ_count.retrieve 6 times.
15:53 berick hey, i noticed that yesterday..
15:54 Dyrcona Each one appears to go to a different cstore drone, too.
15:54 berick didn't notice the exact number, only that it was repeating a lot..
15:54 Dyrcona Well I truncated the log on my dev server and looked myself up after I noticed it in production.
15:54 Dyrcona So I counted in the truncated log.
15:55 Dyrcona fines summary is retrieved twice.
15:55 Dyrcona I was gonna wait until Friday to open a Launchpad bug.
16:01 jeff reminds me of bug 1138684
16:01 pinesol_green Launchpad bug 1138684 in Evergreen "patron.util.set_penalty_css could be called less often" (affected: 1, heat: 8) [Medium,Triaged] https://launchpad.net/bugs/1138684
16:01 Dyrcona Might do it today, since I've got not much else to do for an hour.
16:01 jeff (though not necessarily related)
16:02 Dyrcona No probably not really related.
16:02 Dyrcona But it rhymes.
16:03 Dyrcona hmmm.
16:03 Dyrcona I could check if it is the staff or the actor.user retrieve backend method doing this.
16:03 Dyrcona I could always look myself up via script.
16:04 berick if you see the API call in the activity log (or logged w/ ACT), it's coming from the client.
16:05 Dyrcona Yeah.
16:05 Dyrcona I was gonna add that it looks actor user retrieve by barcode doesn't call for the checked out count.
16:06 Dyrcona It's nice having scripts just lying around. :)
16:07 berick indeed
16:19 Dyrcona jeff: Mind if I plagiarize your bug title?
16:19 jeff feel free!
16:21 jeff staring at the launchpad search box, i tried to suss out a keyword that would find that old bug -- "needless" did the trick.
16:21 jeff speaking of the launchpad search box, i could have sworn it asked me if i was feeling lucky -- not sure if it was quoting Google or Dirty Harry.
16:28 berick jeff: you're still alive, so I think we know the answer
16:36 Dyrcona heh
16:36 Dyrcona berick: lp 1406812
16:36 pinesol_green Launchpad bug 1406812 in Evergreen "open-ils.actor.user.check​ed_out.count.authoritative could be called less" (affected: 1, heat: 6) [Undecided,New] https://launchpad.net/bugs/1406812
16:38 berick Dyrcona++
16:38 Dyrcona Hopefully, it's the last bug of 2014.
16:45 _bott_1 It's an entertaining bug.  "If it makes the call 6 times and receives back the number 6, 6 times, is it 6 or a half-dozen?"
16:47 jeff berick: i'm only still alive because Google something something cyberwar [mumble] yet...
16:48 Dyrcona Hopefully, it isn't exponential....
16:48 Dyrcona 6^6 is a rather large number.
16:54 nhilton_ joined #evergreen
17:00 Dyrcona Well, happy New Year and good night. I'm takin' off.
17:28 dkyle1 joined #evergreen
17:33 mtj_- joined #evergreen
17:43 jeff four of the six calls to open-ils.actor.user.checke​d_out.count.authoritative are from set_penalty_css
17:43 jeff set_penalty_css makes two of those requests, and retrieving a patron calls set_penalty_css two times.
17:48 kmlussier joined #evergreen
17:48 mceraso joined #evergreen
17:48 bshum joined #evergreen
18:39 makohund joined #evergreen
18:53 makohund howdy... I need to get evergreen to resend email hold pickup notices for all holds captured within a given timespan.  (I've fixed the cause of the problem... /var had filled up.  Notices are going out properly again, but I have a bunch that didn't.)  I'm familiar with action.hold_request and action_trigger.event to some extent, but not sure how to make this happen.  anybody available for a quick hand?
19:00 makohund Or is everybody out partyin' already? :)
19:35 jeffdavis makohund: I think most folks have probably headed out. Let me take a quick look.
19:36 makohund Cool, thanks.  I figured that, timing is kinda bad.  For me too, lol.  I'm in the db trying to feel the whole thing out.
19:39 jeffdavis So you have a bunch of generated events in action_trigger.event, right?  Is action_trigger.event.state = 'complete' for most of them?
19:41 jeffdavis I suspect the state for most of those events is actually 'error' or something along those lines, which makes things harder.
19:41 makohund that's what where I'm looking, almost done putting together a decent query, gimme a sec
19:43 makohund did this:  select trigger.id,trigger.target,trigger.event_def as def,to_char(trigger.add_time,'YYYY-MM-DD') as add,to_char(trigger.run_time,'MM-DD') as run,to_char(trigger.start_time,'MM-DD') as start,to_char(trigger.update_time,'MM-DD') as update,trigger.update_process as process,trigger.state,to_cha​r(notice.create_time,'MM-DD HH:MI') as noticetime, REPLACE(LEFT(notice.data,26), E'\n','') AS noticetxt from action_trigger.event as trig
19:44 makohund lol, never saw sql displayed as smiley's before
19:46 jeffdavis heh
19:46 jeffdavis sorry, the end of your query got cut off - would you mind pasting it at pastebin.com and sharing the resulting pastebin link here?
19:46 jeffdavis actually I need to run in a minute too... :(
19:48 makohund sure... but basically when I limit by status, the majority are complete.  only 39 are invalid.  462 are complete.  no errors.  meanwhile far more than 39 show in the client as notices = 0
19:49 jeffdavis For the ones that are complete, you can check the output in action_trigger.event_output
19:49 makohund http://pastebin.com/SY86AAGt
19:50 jeffdavis something like: SELECT a.data FROM action_trigger.event_output a JOIN action_trigger.event b ON a.id = b.template_output WHERE b.id IN (...)
19:50 jeffdavis It may be the case that action_trigger.event_output contains the actual emails that you need to resend, so you could just resend those.
19:51 jeffdavis If it doesn't contain the emails you want, you'd need to re-generate the events, which isn't directly possible as far as I know.
19:51 makohund yes, the query joins that in, show create & a snippet of data from there (enough to see most of an email address).  they all have emails, they really did complete.  but I need them to go again...
19:52 jeffdavis Yeah, I don't think there's a way to resend them within Evergreen. However, you can extract the email content and send them via some other system (you could do that with Perl for example).
19:53 makohund I've had something like this happen once before, and equinox was able to clear something in some fields that made them go again.  Just need to know how to do the same myself.
19:54 jeffdavis ah
19:55 jeffdavis that would be much easier wouldn't it
19:56 makohund yeah, was hoping I could figure it out rather than wait, but not having much luck.  of course it happens on a holiday!
19:57 makohund what I could use is a definition of what all of those fields really mean/do... add time, run time, start time, update, etc.
19:57 jeffdavis Ugh, sorry, I need to run. You are right, it should be doable by setting one or more of the fields in action_trigger.event to null, but I'd need to review the action trigger code to recall how exactly to do that.
19:58 makohund no prob.  pretty sure that's it, gotta set one of those to null, but no clue which one.  I suppose I could go look at the code & try to suss it out?
19:59 makohund cheers, and have a good new years!  :)
19:59 jeffdavis Could do. Or hopefully someone else will stop by. Good luck!
19:59 makohund Thanks!
22:10 makohund left #evergreen
23:52 RBecker joined #evergreen

| Channels | #evergreen index | Today | | Search | Google Search | Plain-Text | summary | Join Webchat