Evergreen ILS Website

IRC log for #evergreen, 2013-09-09

| 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
00:09 bshum And thus the big change begins.
00:15 bshum We have much work to do.
00:17 jeff hmm?
00:36 paxed morning
01:05 jeff bshum++ website work
01:06 bshum Still a work in progress
01:06 bshum but I like to think we're making some progress
01:06 bshum Hard part is going to be tracking down and fixing broken links
01:06 bshum Or resurfacing content that looks too buried.
01:25 Mark__T joined #evergreen
01:45 bshum @later tell dbs Looks like the transition to wordpress has done some funky things to the planet CSS and images.  Short term, we may just put a simple logo up and link back to the main site till we get the menus stabilized (I expect more change)
01:45 pinesol_green bshum: The operation succeeded.
01:55 phasefx bshum++
01:56 bshum :D
02:51 bshum Okay, added in some custom redirects from old static PHP pages to their new Wordpress equivalents.
02:51 bshum That's enough for now.  Off to bed for a nap!
02:51 bshum @later tell moodaepo So much to do, so little time!  :P
02:51 pinesol_green bshum: The operation succeeded.
02:52 bshum @later tell kmlussier See new conference menu link and initial 2014 page. Whee!!
02:52 pinesol_green bshum: The operation succeeded.
02:52 bshum kmlussier++
03:17 pinesol_green` joined #evergreen
05:20 remingtron_ joined #evergreen
05:22 dconnor joined #evergreen
05:59 csharp bshum++
06:01 csharp webteam++
06:57 kmlussier joined #evergreen
06:57 kmlussier bshum++
07:00 timf joined #evergreen
07:06 akilsdonk_ joined #evergreen
07:08 kmlussier @later tell bshum 246 should be part of the alternative title index, unless the 2nd indicator is set to 1, in which case it should be part of the translated title index.
07:08 pinesol_green kmlussier: The operation succeeded.
07:09 jboyer-isl joined #evergreen
07:10 kmlussier @later tell bshum In either case, I would expect it to show up in a browse search if the browse flag is set to true. But, no, I didn't test it specifically.
07:10 pinesol_green kmlussier: The operation succeeded.
07:53 bshum kmlussier: Ah, cool thanks. I will ask Mary to check her examples for indicator flags.
07:54 bshum We'll have to look in production cause we're live on new master! :D
07:55 kmlussier bshum++
07:55 kmlussier You did an upate at the same time you were wrangling the web site. I'm impressed!
07:58 bshum kmlussier: Well, not at the same time. ;)  Same weekend, sure.
07:58 collum joined #evergreen
08:02 kmlussier bshum: Somewhere in my notes, it says that the 246 with 0 in the 2nd indicator are not indexed as an alternative title. However, I'm not sure where I got that information. It doesn't line up with what I see here: http://www.loc.gov/standards/​mods/v3/mods-mapping-3-2.html.
08:09 Shae joined #evergreen
08:12 bshum kmlussier: Hmm, thanks. I'll note that in my testing later today.
08:22 mrpeters joined #evergreen
08:26 bshum csharp++ GPLS++ awitter++
08:26 bshum For keeping the servers happy. :)
08:27 bshum (and flexible)
08:27 csharp ;-)
08:28 csharp also, after I solve a networking kink, we can get mundungus up and moving which will allow for community admins to administer the VM host
08:29 bshum Cool!
08:36 kbeswick joined #evergreen
08:42 csharp @blame java applets
08:42 pinesol_green csharp: It's all java applets's fault!
09:07 mmorgan joined #evergreen
09:16 mllewellyn joined #evergreen
09:18 ericar joined #evergreen
09:20 finnx joined #evergreen
09:28 remingtron_ joined #evergreen
09:45 RoganH joined #evergreen
09:53 bshum kmlussier: Looking at the mods32.sql
09:53 moodaepo bshum++
09:53 bshum It seems like the metabib_field draws from "alternative-nfi"
09:53 rjackson-isl joined #evergreen
09:53 bshum And that in turn does a test for ind1>0 to grab the right substring or not.
09:54 bshum Other than that, I don't see anything in the mods32 for 246 or alternative title info
09:54 bshum but I'm definitely not showing any 246 information for a particular bib record
09:54 bshum Kind of weirding me out
09:55 bshum For giggles, the ind2 is 0.  But even changing that to 1 or whatever doesn't do anything new.
09:55 bshum So I don't know why it doesn't work yet.
09:59 eeevil bshum: first thing I'd check is to be super-double-extra sure that config.xml_transform.xslt contains what you believe it to contain ...
10:01 bshum eeevil: I took a look at that and it seemed right.
10:01 bshum Or at least no different than what I saw in the .sql version
10:02 bshum Oh actually, hmm
10:03 bshum just grabbed the xslt and I don't see "alternative-nfi" in there
10:03 bshum That's odd.
10:03 eeevil by "grabbed the xslt" you mean...
10:03 eeevil because alternative-nfi is evergreen-special
10:03 eeevil the official MODS doesn't know about NFI for anything other than 245
10:03 eeevil (though it should)
10:04 eeevil so, senator added it
10:04 eeevil in support of proper sorting for browse
10:04 bshum Right, just tracked it back to e710ecbe
10:04 pinesol_green [evergreen|Lebbeous Fogle-Weekley] Bib record browser with 'see also', etc from linked authority headings - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=e710ecb>
10:04 bshum I went to check the corresponding entry for mods32 in config.xml_transform.xslt
10:04 bshum And it doesn't contain any reference to alternative-nfi
10:05 * bshum checks the upgrade script...
10:06 krvmga joined #evergreen
10:07 eeevil bshum: YYYY contains an update to the expected row...
10:07 bshum eeevil: Indeed it should.  And the final 0816 has it in there.
10:07 bshum I wonder why it didn't take.
10:09 jboyer-isl Does anyone running 2.3+ know off hand what versions of FF are supported for the xpi extension client, or has that gone away?
10:10 bshum Yep, pager shortcut lib setting at end of script is in production
10:10 bshum And the leading changes too
10:10 bshum Just not the xslt changes
10:10 bshum Weird.
10:11 bshum And also, doh, more reingesting required after we update this all I imagine :(
10:12 bshum jboyer-isl: I... probably wouldn't trust that to work properly given how much Firefox has changed lately.  I know it worked once upon a time with Firefox 14 or so :)
10:12 yboston joined #evergreen
10:12 jboyer-isl I don't trust it at all, but a school wants options for their Macs. :)
10:13 bshum Not ready to build your own mac clients?
10:13 bshum (not that I fully trust that either)
10:14 jboyer-isl At least the xulrunner in the app bundle doesn't change on it's own. I think they were interested in the extension because we don't have the OS X client available on our manualupdate.html page. (Don't really want to commit to supporting it, because there aren't many Macs around here to try to test things)
10:15 jboyer-isl bshum: I also thought you were having some issues with newer clients on Macs? Was that a master thing, or is it all newer xulrunners?
10:16 bshum jboyer-isl: I don't have enough points of reference to test things.  By that I mean I haven't had time to spin up more versions of Evergreen like 2.4, etc. to see if the problems persist.
10:16 bshum I'll get to it.  Sometime.
10:16 bshum And yes, xulrunner on mac had some weird display bugs.
10:16 jboyer-isl Ah, ok. Maybe I'll fight with that at home sometime and see what happens.
10:17 bshum eeevil: I'm going to test that particular portion of the upgrade script again (the mods32 stuff) and see if it changes the xslt or generates some sort of better error telling me something
10:18 bshum Though I may wait till after I can get a snapshot of the production DB to try again
10:18 bshum So sometime tomorrow.
10:18 bshum Though I suppose I should see if this is the case on our previous test server.... presumably it'd be broken there too.
10:18 * bshum will look at that.
10:19 eeevil bshum: I will await your report
10:22 bshum Well that's just annoying
10:22 bshum The mods32 entry on our test server has the right xslt entry for alternative-nfi
10:24 bshum I don't know why this is weird.
10:27 bshum And for the life of me I can't seem to find a metabib.title_field_entry for the bib record anyways.
10:28 bshum On the test server, even with the proper xslt
10:28 bshum Well, there are entries, just not for the alternative title
10:30 * bshum goes back to do more digging
10:31 bshum Oh, what the heck
10:31 bshum Okay, I must be seeing things
10:31 bshum Cause now I'm seeing the alternative-nfi in the xslt entry for mods32
10:32 bshum I must have done a bad copy/paste eeevil, sorry for scurrying down the wrong path.
10:33 rfrasur joined #evergreen
10:33 bshum Doesn't solve my missing entry problem, but at least it wasn't the upgrade script gone awry
10:40 dbs jboyer-isl: MARC Editor won't work with recent versions of Firefox/XULRunner due to the removal of E4X XML support, so that much we know would be broken.
10:41 dbs bshum: all in all, the CSS change doesn't hurt the planet too much; still functional. Lemme know when things settle down.
10:44 bshum dbs: For me or for the site?  :P
10:45 rfrasur bshum++
10:45 bshum dbs: But really, I'm juggling some post-upgrade madness presently.  I'm taking breaks here and there to look at other stuff as I go along.
10:46 dbs bshum++ # no rush!
10:48 bshum It looks like in our last upgrade, we might have released a client that didn't have auto-updates enabled.
10:48 bshum So folks lost the chain of updating
10:48 bshum And had to download fresh client this morning.
10:48 bshum But it's not a consistent issue across the board.
10:48 bshum Some had it, some didn't.
10:48 bshum So I'm trying to track back through my work to see where I might have screwed up last time, but it was months ago :D
10:48 dbs bshum: our last auto-update was a mixed success, most people ended up having to download a fresh client.
10:49 bshum The little things in life.
10:49 dbs Seemed that some things were/are very strongly cached. Or something.
10:50 bshum For fun, brick_ctl.sh probably needs an update to handle the change to osrf_control
10:51 gmcharlt bshum++
10:51 gmcharlt mighty purty
10:51 bshum Or at least it really wanted to complain about that as I turned on bricks yesterday
10:53 * bshum was glad they still turned on at least.
10:57 rfrasur_ joined #evergreen
11:02 tony_ joined #evergreen
11:04 tony_ Good Morning everyone.. I got something crazy going on-- when using: osrf_ctl.sh -l -a start_all and service apache2 restart-- I get NO errors within putty but on the opac I get a 500 error and staff client can't connect--  Any Suggestions
11:05 jeff tony_: your apache error logs should contain some clues. i'd start there, then check the opensrf logs (usually at /openils/var/log/osrfsys.log )
11:07 tony_ Thank you for the suggestion..  didn't even think about checking the logs-- this has been a long day already
11:09 bshum B&T--
11:10 paxed kmlussier: we'll join in on the icon project, but i don't know with how big a contribution. also, kivilahtio wanted some graphs and/or more technical blueprints. (he's big on those ;)
11:11 kmlussier paxed: I should be getting something with more detail this week that I can forward along. But the full tech specs don't typically come until later in the project.
11:12 * paxed nods
11:13 kmlussier Oh, and before I forget...
11:13 kmlussier paxed++
11:13 kmlussier kivilahtio++
11:13 mllewellyn bshum, no cussing.  :)
11:14 kmlussier bshum cusses? I'm shocked!
11:14 paxed it's something we'll need, and even though my quick hack would've worked for us, this is better for everyone.
11:14 mllewellyn He said "B&T" which is beginning to be a cuss word around here
11:15 rfrasur_ mllewellyn++ #that's kinda really funny
11:16 * bshum smacks head on table
11:17 dbs tony_: 1) you're sure you need the "-l" in osrf_ctl.sh and 2) try a stop_all first and see if any opensrf / open-ils processes are still running; then start_router and ensure that Router is running, then start_perl and ensure that open-ils.storage is running; then start_c and ensure that open-ils.cstore is running
11:27 Dyrcona joined #evergreen
11:37 bshum Oh that's awkward
11:37 bshum The dojo fields for MARC import/export render fine in Linux
11:37 bshum But in Windows, you click on them and nothing shows :S
11:39 bshum Looking to see if maybe this is a weird glitch resulting from the new vandelay default stuff
11:39 bshum There is an undefined error in the javascript console
11:40 bshum TypeError: vlbibSources[0] is undefined
11:40 bshum I don't see that in my Linux client.
11:43 bshum Oh that's happening cause I clicked on bib source and it didn't generate any dropdown entries
11:43 bshum That's lovely
11:44 Dyrcona bshum: Switch everyone to GNU/Linux. Problem solved.
11:44 bshum Dyrcona: I definitely suggested that ;)
11:45 * tsbere is attempting to figure out why a/t event defs won't load in his staff client in production
11:45 rfrasur Let's do it.
11:47 kbeswick_ joined #evergreen
11:49 jdouma joined #evergreen
11:50 bshum Wow, it's a cache issue
11:50 _zerick_ joined #evergreen
11:50 bshum Using "Clear Cache" works to resolve the problem on our Windows boxes
11:50 bshum Super weird.
11:50 _zerick_ joined #evergreen
11:50 zerick joined #evergreen
11:53 GeoffSams joined #evergreen
11:54 GeoffSams joined #evergreen
12:06 mllewellyn left #evergreen
12:11 bshum PG 9.3.0.  Oy, another version behind :(
12:18 mllewellyn joined #evergreen
12:21 mrpeters tater: moodaepo: any chance http://git.evergreen-ils.org/?p=contrib/equi​nox.git;a=blob;f=eg-stats/parse-eg-stats.pl;​h=403175bf4a2a78cda319920d51fb811cb5a82b2f;h​b=d01c046e9c79e7337e58c9032d7e8a6e9d7d54c1 is busted?   It reports the lost drones properly from command line, but nagios always reports it as OK :(
12:22 * mrpeters doesn't rule out the possibility that it could be an icinga vs nagios thing....but usually the plugins are interchangeable
12:22 mrpeters will try modifying the perl for Icinga if you guys know for sure it works on nagios
12:23 mrpeters I'm leaning towards that icinga interprets $status differently...
12:25 bshum Okay, clear cache worked for some, but not everybody.
12:25 bshum And apparently, they still lose the dojo fields eventually
12:25 bshum Maybe there's some timeout occurring
12:46 dboyle joined #evergreen
12:47 moodaepo mrpeters: Yea I think it's worked fine with Nagios for us.
12:48 mrpeters yeah, i think it may just be a translation thing, though icinga accepts the same 0,1,2,3 status codes as nagios
12:48 mrpeters its puzzling...its just the web interface where the check fails
12:48 mrpeters just always reports a 0 status
12:49 mrpeters but ill figure it out, and share back in case it is an icinga thing
12:50 kmlussier1 joined #evergreen
12:55 krvmga joined #evergreen
13:16 stevenyvr2 joined #evergreen
13:44 ericar joined #evergreen
14:35 RoganH joined #evergreen
14:50 ericar joined #evergreen
14:54 bshum Hmm, so after a little while, the MARC import/export stops working, even for me on Linux.
14:54 bshum Like a few hours later, I log in and it's broken
14:54 bshum Till I clear cache again
14:54 bshum And then log back in and things seem fine again
14:54 bshum Very weird.
14:55 Dyrcona Vandelay? I never use it. Our catalogers curse it.
14:56 bshum yeah it's vandelay
14:56 bshum I'm pondering if this is either a side effect of the stuff we changed with dojo filters
14:56 bshum Or if this is a problem with the new vandelay defaults
14:56 bshum Slowly testing each thing
15:07 bshum Well it's not the autogrid stuff, not that I expected it to be.
15:07 bshum Alright, what else changed...
15:08 Dyrcona @blame Evergreen
15:08 pinesol_green Dyrcona: Evergreen musta been an Apple employee.
15:13 jeff_ bshum: vandelay can be run in a browser, where you might have more ready access to debugging. is it possible that you have an issue with a single backend system or brick?
15:13 bshum jeff_: I tried logging into each app server individually
15:14 bshum And they work and don't work sporadically as well
15:14 bshum I could try using vandelay in a browser i guess
15:38 bshum Yeah
15:38 bshum Even in Firefox browser, it sometimes shows values and sometimes it doesn't
15:39 bshum Even on the same app server, just hitting refresh
15:39 bshum It'll be there one load, gone the next
15:47 bshum Interesting
15:47 bshum If I click on the record type
15:47 bshum And change it from Bibliographic Record to something else
15:48 bshum Like Authority or Acquisitions
15:48 bshum The other fields turn on or off
15:48 bshum And then when I flip it back to Bibliographic Record, the fields behave normally again
15:50 bshum Maybe there's some weird fight going on between the lookups on that page
15:50 bshum Sigh
15:55 bshum Yeah, that workaround seems to work right now for our catalogers.  Guess we'll play with that for a bit till I get more time to look more closely.
15:55 bshum (old) dojo--
16:13 stevenyvr2 joined #evergreen
16:23 jdouma joined #evergreen
16:43 ldwhalen In Evergreen's SQL JSON.  Is there a way to join on they primary keys and two other fields from the same tables?
16:43 Dyrcona ldwhalen: yes. if the keys have a foreign key relationship defined in the fieldmapper, you get that for free.
16:44 ldwhalen Ah, so update fm_IDL is my solution?
16:44 Dyrcona Not necessarily.
16:44 Dyrcona But maybe.
16:44 ldwhalen Dyrcona: ok, thanks
16:44 eeevil ldwhalen: you mean, like: a JOIN b ON (a.x = b.x AND a.y = b.y AND a.z = b.z)?
16:45 Dyrcona What are you trying to do? Could you paste your JSON somewhere?
16:45 eeevil if so, yes.  see the filter clause of the join
16:45 ldwhalen eeevil and Dyrcona: I willpaste
16:46 eeevil ldwhalen: http://wiki.evergreen-ils.org/doku.php?id=doc​umentation:tutorials:json_query#join_filters
16:46 * Dyrcona Was just about to paste a link for those interested in more information.
16:47 ldwhalen http://pastebin.com/crYNd9sY
16:47 eeevil Dyrcona: if it's a different link, please do!
16:47 ldwhalen eeevil: I am loking at that doc, but I cannot find the case I am lookign for.
16:47 ldwhalen I need ON (a.pkey = b.pkey AND a.text_field = b.same_text_field_different_anme)
16:49 Dyrcona My link is just a different (older) version of the same doc.
16:49 eeevil ldwhalen: meh... I'd just put that in the WHERE ... there's no left join there
16:50 eeevil well, I guess there is, nm
16:50 eeevil but or-is-null takes care of that in the where
16:50 ldwhalen eeevil: ahh thank you.  That was what I was missing.
16:52 eeevil np
16:55 Dyrcona Glad eeevil could help. I was going to say that I'm not all that familiar with those tables.
16:55 Dyrcona And, LibreOffice Writer has messed up the formatting of my bookmarks file.
17:01 * Dyrcona wonders where the option to view the raw HTML went in LibreOffice 4.
17:01 ldwhalen Dyrcona: thank you for you help too.  I hope your bookmark file is recoverable.
17:02 Dyrcona Oh, it's just a matter of formatting. The content is still there.
17:04 Dyrcona Ah ha!
17:05 Dyrcona If anyone cares, in LibreOffice Writer 4, go to Options, and under LibreOffice Writer View options turn on Field Codes under Display.
17:05 Dyrcona Then the view option for HTML Source comes back on the View menu.
17:07 Dyrcona hm.. But that doesn't seem to stick....
17:08 Dyrcona Opened the document again and the option was gone....
17:08 Dyrcona software.....
17:09 yboston joined #evergreen
17:17 mllewellyn left #evergreen
17:17 mmorgan left #evergreen
19:08 GeoffSams joined #evergreen
19:51 mrpeters joined #evergreen
21:10 ldwhalen eeevil: You were right, those columns do not join.  Figured it out in the end.
21:33 mrpeters joined #evergreen
21:53 mrpeters joined #evergreen
21:53 mrpeters left #evergreen
21:55 RoganH joined #evergreen
22:54 rfrasur joined #evergreen
22:55 rfrasur Does anyone know if anyone has tried using the staff client on a linux tablet?
22:56 rfrasur o0(because I'm sure everyone else is thinking about mobile circulation at 11 p.m.)
23:10 jcamins rfrasur: this doesn't answer your question, but I think there was a class working on a mobile interface for EG at some point.
23:10 jcamins At IU, maybe?
23:10 * jcamins calls it a night.
23:45 zerick joined #evergreen

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