Evergreen ILS Website

IRC log for #evergreen, 2013-10-22

| 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
01:21 DPearl joined #evergreen
01:43 moodaepo_nb joined #evergreen
03:27 tsbere joined #evergreen
03:57 zxiiro joined #evergreen
07:01 timf joined #evergreen
07:44 jboyer-isl joined #evergreen
08:13 tfaile joined #evergreen
08:14 rjackson-isl joined #evergreen
08:15 sylvar joined #evergreen
08:17 sylvar Out of Context Theater presents... deleted_mrs. Sounded like a divorce, at best, until I realized the context was metarecords.
08:26 akilsdonk joined #evergreen
08:39 kbeswick joined #evergreen
08:59 RobertL_ joined #evergreen
09:00 Dyrcona joined #evergreen
09:09 dbs huh. Ubuntu Saucy has Encode.pm 2.51 (http://packages.ubuntu.com/saucy/libencode-perl), so unless someone is running Jessie or Fedora or has manually upgraded their version of Encode, they're unlikely to hit the changed decode_utf8() behaviour
09:12 ericar joined #evergreen
09:13 mmorgan joined #evergreen
09:15 jbfink joined #evergreen
09:16 mrpeters joined #evergreen
09:24 Dyrcona dbs: Is there a solution that works whether one has upgraded or not?
09:27 dbs Dyrcona: It would be great if someone not on one of those distros would test my branch and tell me how much breaks :)
09:30 Dyrcona dbs: I only look at LTS releases from Ubuntu, so 14.04 would be the next that I experiment with.
09:31 dbs Dyrcona: right, but if you have a 12.04 box at the ready it would be instructive to just check out the branch and run 'make check' on it
09:32 Dyrcona dbs: Will do.
09:33 * dbs starts up ye olde Precise virtualbox. this 4-year-old laptop is starting to feel long in the tooth
09:35 paxed i've been having "fun" with 13.10 since i updated on friday. twice now lightdm has switched from my user to "guest login" while i've been typing the password. once the kernel crashed, and once it just wouldn't boot at all.
09:38 * berick eyes the Saucy upgrade window that popped up this morning warily
09:41 * dbs waits for 200+mb of updates to apply to his precise vbox
09:44 Dyrcona working/user/dbs/encode-changed-behaviour <- that branch?
09:50 Dyrcona This is not good.
09:51 Dyrcona I told osrf_control to stop-all, and cstore would not stop right away.
09:51 Dyrcona I check the database afterward and there are three queries running from the 12th.
09:51 Dyrcona I think this is related to upgrade scripts.
09:53 Dyrcona Yeeup, select * from metabib.browse(....)
09:55 * Dyrcona thinks someone else needs to have a good look at those upgrade scripts with "real" data, to make sure it isn't just my dev. environment.
09:59 dbs Dyrcona: yessir
10:00 Dyrcona dbs: How would I know if things are broken? That is, what should I look for after applying the branch?
10:01 dbs Dyrcona: does make check
10:01 dbs Dyrcona: does "make check" pass for starters?
10:01 dbs And then if that passes, "eg_db_config <yada yada> --load-all-sample"?
10:02 Dyrcona I'm testing this on a system that already has data loaded.
10:02 dbs three queries from the 12th? Wow.
10:02 paxed dbs: not sure. i know i haven't done such a post. i could try to compile one...
10:03 dbs Dyrcona: ah, okay, my branch doesn't have any upgrade scripts for the database functions yet, so just "make check" or running "prove -l lib t" in the Open-ILS/src/perlmods directory would be great
10:03 Dyrcona I'm about to paste the output from make.check, looks some tests fail.
10:04 dbs paxed: yeah. I mean, I know you have a long list of bugs but some level of abstraction & prioritization would probably help us understand our failings
10:04 dbs Dyrcona: thanks
10:05 pastebot "Dyrcona" at 64.57.241.14 pasted "make check > ../make.out 2>&1" (131 lines) at http://paste.evergreen-ils.org/28
10:05 Dyrcona One of the tests that fails is entityize diacritics.
10:05 dbs Dyrcona: perfect. that's what I was worried about.
10:05 paxed dbs: for example, today i found out the "Serials" as done by both Eg and Koha does not really serve our needs - our expert on it said the Serials is more for scientific papers and such, rather than "periodicals" (as he said), or magazines.
10:06 Dyrcona paxed: You'll just have to implement your own ILS, then. :)
10:07 paxed Dyrcona: in java, perhaps? :)
10:07 jcamins paxed: node.js is more fun. :P
10:07 dbs paxed: yeah, given how much Evergreen's serials does (in several different ways) for serials, some insight into that would be interesting. Your expert wants to track at the article level or something?
10:10 jcamins Koha has rudimentary support for analytics, but there's not a separate module. Just a few tools you can use to simplify creation of article-level records.
10:10 paxed dbs: i'm not sure, it could be his workflow, too. one of the things that is missing is the ability to place a hold on specific magazine number? say, "Pet Care, 2012, 10"
10:10 Dyrcona paxed: Of course, in Java. It would only cost 1.4 billion euros!
10:10 Dyrcona left #evergreen
10:10 paxed dbs: i did talk with him today a bit, and i'm going to talk with him more, and hopefully i'm able to translate his (our) needs for the community at large :)
10:10 Dyrcona joined #evergreen
10:11 dbs paxed: uh. I thought the new Evergreen serials stuff that lebbeous / dbwells were working on for that past few years _could_ do that
10:11 * Dyrcona curses his fingers.....Ctrl+w in the wrong window!
10:11 dbs paxed++ # that would be awesome
10:12 paxed dbs: well, then it could be something that's missing from our database, and clicking on one of the issues under "issues held" does nothing.
10:12 paxed s/and/as/
10:12 paxed (in opac)
10:12 dbs Dyrcona: oh, fascinating. "prove -l lib t/14-OpenILS-Utils.t" works fine on its own, which is where I was focused on fixing the problem that had popped up, but last night / this morning I failed to run all the tests
10:12 jboyer-isl paxed, Dyrcona: Somebody beat you to it: http://en.wikipedia.org/wiki/NewGenLib
10:13 dbs jboyer-isl: oh yeah, that's been around a long time
10:13 jboyer-isl Even has Acq. I tried to set it up in 2008-ish. Daunting is the polite term.
10:13 dbs jboyer-isl++
10:14 paxed dbs: http://62.148.106.92/eg/opac/record/404413?locg=​1;qtype=keyword;query=anna;expand=issues#issues
10:15 dbs Dyrcona: adding 'use utf8;' to the top of t/01-OpenILS-Application.t fixes that for me. I'll push that in.
10:15 Dyrcona heh, no mention of ESB, EIP, POJO, REST, JMS, etc. I can't take it seriously as a Java project.
10:20 Dyrcona See, this all reinforces my conclusion from yesterday: "Software sucks!"
10:21 Dyrcona Well, I guess most bridges collapsed for the first hundred or so years that people started building them, so we've got a few years before we know what we're doing.
10:21 jboyer-isl Eh, there are types of data where sql works fine, and some where it doesn't. I don't think there's going to be a good point to say "ok, throw away those SQL books, we won't ever need that old thing again!"
10:21 jcamins jboyer-isl: of course.
10:22 Dyrcona Damn! Why do all graphic novels have to go in Young Adult?
10:22 jboyer-isl jcamins: I may be misunderstanding what you wrote then, were you referring to really large sized databases, or individual units of data that are really large?
10:23 * Dyrcona mumbles something derogatory about bun heads.
10:24 jboyer-isl Dyrcona: We had a few donated at my last lib that frazzeled some buns, I think they've got some in Adult now. :D
10:24 Dyrcona jboyer-isl: PostgreSQL has this feature where you can throw in a blob of data and get an OID back that can be listed in a relational table for later retrieval.
10:24 Dyrcona jboyer-isl: PostgreSQL is officially an object/relational database.
10:25 jcamins jboyer-isl: I was thinking of BLOB-ish data, but databases with staggering volumes of non-relational data also probably doesn't require an RDBMS.
10:25 dbwells paxed: looking at the "Anna" record you posted, something is not working as designed.  When you click on a summary (like "2013:1, 2013:3 - 2013:5, 2013:11"), you should get a list of issuances, and that is where you can place issuance-level holds.
10:25 jboyer-isl Dyrcona: Sure, I just wasn't following jcamins earlier remark about NoSQL.
10:25 gsams joined #evergreen
10:26 paxed dbwells: as i figured. any idea where i should start looking for what's wrong?
10:26 Dyrcona dbs: Nothing seems broken in TPAC on the development server where I loaded your banch.
10:27 senator paxed: http://docs.evergreen-ils.org/2.4/_​displaying_issues_in_the_opac.html
10:27 senator that has a screenshot of the kind of thing you should see
10:27 Dyrcona And, seriously, no one in our consortium has A La Recherche du temps perdu in French?
10:27 senator to set it up, follow the "prev" link from that documentation page to seoe the section on administration
10:28 dbwells paxed: My only guess would be that your distributions don't have a "Display Grouping", but I think you can set a library-level default for that as well.
10:30 paxed dbwells: we'll take a look tomorrow.
10:32 dbs Proust, in his first book, wrote about (he wrote about)
10:33 dbs Dyrcona: yeah, the affected database functions are the usual suspects: naco_normalize and search_normalize, maintain_901 and maintain_control_numbers
10:37 _bott_ joined #evergreen
10:52 bshum jcamins: http://evergreen-ils.org/conference/eg14/ has details about EG 14 next March.
10:53 jcamins bshum: thanks! I guess you woke up this morning after all.
10:53 bshum jcamins: I'll probably hate myself more later in the week.
10:53 bshum Or on the weekend.
10:54 jcamins Heh. What time did you get home?
10:54 bshum Around 1:40 or so
10:55 jcamins Eww. I took a cab and was back before 12:30.
10:56 jcamins Usually I am frugal and take the bus/subway, but I just couldn't bring myself to test the late night schedule.
10:56 bshum Indeed, it was late.
10:57 bshum Good to be home though.
10:57 ktomita joined #evergreen
10:57 sseng joined #evergreen
10:57 fparks joined #evergreen
10:58 Dyrcona joined #evergreen
10:59 dconnor_ joined #evergreen
11:04 frank____ joined #evergreen
11:08 frank____ hello everybody, Is it posible change the display of issues held, now it displays in one row, but we´d like to show it in vertical listing
11:10 artunit joined #evergreen
11:21 Dyrcona joined #evergreen
11:22 rjackson-isl joined #evergreen
11:23 Dyrcona Fun with networking!
11:24 senator frank____: what timing! someone else was asking about that sort of thing just a little while ago. you can do this: http://docs.evergreen-ils.org/2.4/_​displaying_issues_in_the_opac.html
11:29 frank____ senator: Doesn´t it work with http://docs.evergreen-ils.o​rg/2.4/_mfhd_records.html?
11:33 senator frank____: ah, no. only holdings received through an evergreen receiving interface (or migrated as if they were received in evergreen) can be displayed in the way described in the documentation that i linked. mfhd holdings cannot.
11:34 frank____ ok, senator thanks
11:38 smyers_ joined #evergreen
11:52 dbwells frank____: you should be able to edit the issues-mfhd.tt2 template to do what you are after.
11:54 dbwells frank____: we added some extra breaks in our display by changing the line starting 'thing.join' to:  thing.join(", "); %]<br />[%
11:55 dbwells frank____: you might also consider changing that join, but that level of lined-ness didn't work for our data.
11:56 frank____ excellent dbwells, I added it and it works!
11:56 frank____ thanks
11:56 dbwells frank____: yay :)
11:58 smyers_ joined #evergreen
12:00 pinesol_green [evergreen|Ben Ostrowsky] LP#1240207: Spellchecked the docs - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=b4e7ab2>
12:05 ericar_ joined #evergreen
12:05 timhome joined #evergreen
12:06 Callender_ joined #evergreen
12:06 tfaile_ joined #evergreen
12:06 BigRig_ joined #evergreen
12:06 phasefx joined #evergreen
12:06 akilsdonk joined #evergreen
12:07 tater joined #evergreen
12:08 Sato`kun joined #evergreen
12:34 csharp sylvar++ # spellchecking
12:35 bshum csharp: I remember that z3950 bug so I'll mark it confirmed.  I never got to the bottom of it either, but it seemed to only appear as of the recent 2.5/master builds
12:35 bshum I'll try it again once I get a new fresh system going later this week.
12:37 csharp yeah - I was going to experiment with reverting 3335c54 just to see if that makes a difference - nothing about that commit touches 'obj.active_services' though, so dunno ;-)
12:37 pinesol_green [evergreen|Bill Erickson] Import bib trash fields : XUL Z39.50 UI - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=3335c54>
12:41 jeff ah yes... that lovely moment when you're testing your user editor changes and the username is replaced with: tfunctionanonymousnifargume​ntslengththisanreturnthisa
12:49 gmcharlt heh
12:54 jeff also fun: those moments where you spend some much time trying to write a non-awkward comment that you also end up re-factoring the one if statement that you were commenting
12:54 jeff ...not in a significant way, but just so that the ordering matches the ordering of the comment, because you have no interest in re-writing the comment Yet Again.
12:59 jboyer-isl jeff: beats the other way around: "If I swing this comment hammer hard enough, maybe I can bang this code into shape."
13:01 Dyrcona Comments? We don't need no steenkin' comments...
13:01 csharp @praise comments
13:01 * pinesol_green the upgrade came off brilliantly, and it's all because of comments
13:01 Dyrcona @blame comments
13:01 pinesol_green Dyrcona: comments caused the white screen of death!
13:04 csharp @praise [blame]
13:04 * pinesol_green (blame [<channel>] [<id>] <who|what> [for <reason>]) -- Blames <who|what> (for <reason>, if given). If <id> is given, uses that specific blame. <channel> is only necessary if the message isn't sent in the channel itself. is the very model of a modern major hacker
13:05 artunit joined #evergreen
13:11 Dyrcona @praise comments for great justice!
13:11 * pinesol_green comments is the very model of a modern major hacker for great justice!
13:18 smyers__ joined #evergreen
13:22 csharp okay - I can confirm that reverting 3335c54 did not fix bug 1243280
13:22 pinesol_green [evergreen|Bill Erickson] Import bib trash fields : XUL Z39.50 UI - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=3335c54>
13:22 pinesol_green Launchpad bug 1243280 in Evergreen "'TypeError: obj.active_services is undefined' error when entering z39.50 MARC import UI" (affected: 1, heat: 6) [Medium,Confirmed] https://launchpad.net/bugs/1243280
13:22 csharp so it's somethin' else
13:28 Meilyn joined #evergreen
13:28 Meilyn left #evergreen
13:37 smyers_ joined #evergreen
13:58 hopkinsju joined #evergreen
14:01 tater joined #evergreen
14:01 akilsdonk joined #evergreen
14:01 ericar joined #evergreen
14:02 timf joined #evergreen
14:03 phasefx joined #evergreen
14:03 BigRig joined #evergreen
14:03 Callender joined #evergreen
14:03 tfaile joined #evergreen
14:11 Sha`Bren joined #evergreen
14:17 JennB joined #evergreen
14:18 akilsdonk_ joined #evergreen
14:18 tfaile_ joined #evergreen
14:18 Callender_ joined #evergreen
14:18 timhome joined #evergreen
14:18 phasefx_ joined #evergreen
14:19 BigRig_ joined #evergreen
14:19 ericar_ joined #evergreen
14:20 BigRig__ joined #evergreen
14:21 mtate joined #evergreen
14:22 BigRig joined #evergreen
14:24 BigRig_ joined #evergreen
14:26 BigRig__ joined #evergreen
14:30 BigRig joined #evergreen
14:30 JennB hello Evergreen Community, can you tell me if EG can be in Arabic? I saw this listserv message http://list.georgialibraries.org/pipermai​l/open-ils-general/2011-March/004343.html but as you see its from 2011
14:31 * bshum imagines it's possible, but doubts we've tested RTL extensively
14:31 bshum (or at all)
14:31 JennB ok, thanks Ben!
14:32 bshum In theory, if we had the translations for arabic, we could try reviewing various parts of the GUI for right to left compatibility.
14:32 Dyrcona Well, I've processed some records recently that had Arabic in them.
14:32 hopkinsju joined #evergreen
14:32 * bshum imagines we'd have similar quirks like Koha does with RTL CSS problems.
14:32 JennB and the records were OK?
14:33 JennB I have a library in Jordan asking so Id imagine all would need to be in Arabic
14:33 Dyrcona JennB: Some had issues, but I shared them with gmcharlt, and IIRC he updated MARC::Charset and I believe they work now.
14:33 JennB ah, good to know Dyrcona
14:33 Dyrcona You will need the latest MARC::Charset from CPAN.
14:34 JennB ok, thanks
14:34 Dyrcona Of course, these were only a handful, and only parts of the records were in Arabic, like proper names.
14:35 Dyrcona As for the user interface, like bshum said, that's a bigger job.
14:37 dagreen joined #evergreen
14:39 stevenyvr2 joined #evergreen
14:39 bshum One of the sessions at Koha described i18n issues, especially with RTL languages and now I've got the i18n bug in me to find out more in Evergreen too.
14:40 bshum When I have some more free time to poke.
14:41 dkyle left #evergreen
14:45 paxed bshum: unfortunately my knowledge of RTL languages is very close to nil.
14:49 bshum paxed: Do you have any strong opinions about translation tools that might work better?  (I'm still contemplating stuff like Transifex, Weblate, etc.)
14:53 roses joined #evergreen
14:53 roses Are there plans to have the 655 fields be validated in the future so we can keep what comes with the bib record and they will be useful in a controlled VEL vocabulary context?    As opposed to keep them and not have a way of checking them against the controlled VEL vocabulary.
14:53 roses And yes I asked this question on the listserv but haven't gotten a response and I have a library waiting for a response.  Thanks.
14:54 paxed bshum: i haven't tried either of those. i just know i don't like Pootle
14:55 Dyrcona <beavis_and_butthead>Heh. Heh...He said, "pootle."</beavis_and_butthead>
14:56 Dyrcona roses: I don't know of any such plans.
14:57 roses Dyrcona: Is that something that we could ask for (if enough libraries wanted it) to be considered or is it because it's not doable?
14:58 Dyrcona roses: It probably can happen. It most likely won't happen unless someone pays for it.
14:59 roses Dyrcona:  Thanks - that's what I needed to tell them.  I appreciate your help.
14:59 Dyrcona BTW, what does VEL stand for? My Google-Fu is weak today.
15:02 csharp roses: that looks like a good candidate for a wishlist bug
15:02 roses Dyrcona: Virginia Evergreen Libraries :-)
15:02 csharp getting it out there lets people know there's a need/interest
15:02 csharp and I'm sure if your library wants it, at least one other out there does too
15:02 Dyrcona So, you're asking about a locally controlled vocabulary?
15:03 hopkinsju joined #evergreen
15:03 roses csharp: Thanks.
15:03 * Dyrcona wasn't sure if it was something like that or some standard he'd never heard of. :)
15:03 roses Dyrcona: Ummm - I'm not sure - does it make a difference?
15:03 csharp @marc 655
15:03 pinesol_green csharp: Terms indicating the genre, form, and/or physical characteristics of the materials being described. A genre term designates the style or technique of the intellectual content of textual materials or, for graphic materials, aspects such as vantage point, intended purpose, or method of representation. A form term designates historically and functionally specific kinds of materials distinguished (1 more message)
15:04 csharp @marc 650
15:04 pinesol_green csharp: A subject added entry in which the entry element is a topical term. (Repeatable) [a,b,c,d,e,v,x,y,z,2,3,4,6,8]
15:05 csharp roses: validation (iiuc) is validating against your authority file - do you have those genre headings entered?
15:12 roses csharp:  Had to call the librarian/cataloger who asked the question and we are not sure.  If they have been downloaded - they (655) should validate?
15:13 roses csharp: 650 validates.
15:18 smyers_ joined #evergreen
15:21 dbs looks like stock Evergreen uses '155','180','181','182','185' from loaded auth records to validate 655
15:21 dbs based on authority.control_set_authority_field anyway
15:22 dbs Also from the control map for the staff client -- roses ^^
15:24 roses dbs: That helps a whole bunch.
15:25 roses dbs: That 655 can be validated against loaded authority records - right?
15:25 csharp dbs++
15:25 * csharp was going to dig into the code but got sidetracked ;-)
15:26 roses csharp: The librarian/cataloger that was asking is from Georgia and used Georgia Pines.
15:26 dbs roses: that's the idea. so if it isn't, it's either a bug in our code, or in the auth records, or in my understanding :)
15:26 dbs Also note that that is Evergreen master, I haven't dug into older code
15:43 csharp roses: is it Sara?  If so, tell her hello ;-)
15:50 smyers__ joined #evergreen
16:03 roses csharp: Yes it was Sara.  She's my troublemaker :-)
16:06 roses csharp: She is also our advanced cataloger trainer and I think she is super - even though she asks difficult questions (to me a non-cataloger)
16:06 dagreen joined #evergreen
16:23 csharp roses: I believe she and I are related.  I have ancestors with the same last name from the area of Georgia where she's from
16:32 roses csharp: That's even better to have that connection.  Thanks for all of your help today.
16:34 dagreen left #evergreen
16:37 * Dyrcona starts to wonder if the world really did blow up.
16:55 eeevil it did afaict
16:55 Dyrcona Netcraft confirms it!
16:58 tater out
17:00 smyers_ joined #evergreen
17:03 hopkinsju joined #evergreen
17:04 GeoffSams joined #evergreen
17:07 hopkinsju Alrighty… and now for the question I just inadvertently posted to #code4lib...
17:07 hopkinsju I cannot for the life of me figure out this Org Unit Hiding Depth setting. Anyone using this that can advise?
17:08 hopkinsju I'm trying to get a basic handle on how it works and I can't seem to get it to have any effect at all.
17:09 hopkinsju In the description of the setting it says it bases the users OU on the ol parameter. Is that still used? It seems like TPAC just throws it away most of the time.
17:14 mmorgan left #evergreen
17:16 hopkinsju I'm really wondering if it's working at all.
17:17 tsbere hopkinsju: I think it works with physical_loc only
17:17 tsbere so if you don't have one you see no effect
17:17 hopkinsju tsbere: What's a physical_loc?
17:18 tsbere hopkinsju: An attempt to say "the user is physically at...." though we use it with subdomains to say "this subdomain represents..."
17:18 tsbere I believe it can be set with a url param in addition to IP based and apache-level configs
17:19 hopkinsju tsbere: Any chance of documentation on this? Particularly the apache config
17:20 tsbere hopkinsju: At the apache level you set an environment variable named "physical_loc"...for that matter for your testing just add physical_loc=id to the query string (add a ? or & or whatever before it as needed)
17:20 hopkinsju Gotcha
17:21 tsbere we do some fancy rewrite tricks to set all of that stuff for our subdomains, but you can also just hardcode it if you have widely varying configs
17:22 tsbere hopkinsju: Once you have physical_loc set then the hiding depth (if >1) says "start the tree at that depth". 1 would be the "System" in a standard setup, I believe (with 0 being "everything" and 2 being "branch within the system")
17:22 tsbere err, >=1
17:24 hopkinsju tsbere: Thanks a lot. That really helps.
17:46 smyers_ joined #evergreen
17:59 jeff hrm.
18:01 smyers__ joined #evergreen
18:28 smyers_ joined #evergreen
18:43 hopkinsju joined #evergreen
18:48 smyers__ joined #evergreen
20:10 kbeswick joined #evergreen
20:24 stevenyvr2 left #evergreen
22:00 rjackson_isl joined #evergreen
22:05 rjackson-isl joined #evergreen
23:51 hopkinsju left #evergreen

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