Evergreen ILS Website

IRC log for #evergreen, 2015-04-29

| 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
02:02 StomproJ joined #evergreen
02:09 BigRig joined #evergreen
07:06 mrpeters joined #evergreen
07:33 jboyer-isl joined #evergreen
07:37 ericar joined #evergreen
07:44 akilsdonk joined #evergreen
07:47 Newziky joined #evergreen
07:48 Newziky left #evergreen
07:53 graced joined #evergreen
08:29 ericar_ joined #evergreen
08:38 mmorgan joined #evergreen
08:44 Dyrcona joined #evergreen
08:44 * csharp decides that asking libraries to use MarcEdit to remove the {dollar} characters manually is a pretty viable workaround until bug 1449724 is addressed
08:44 pinesol_green Launchpad bug 1449724 in Evergreen "Acq Record Import Fails because of dollar sign in price field" (affected: 2, heat: 12) [Undecided,New] https://launchpad.net/bugs/1449724
08:45 * csharp wonders how other EG-using agencies ordering from Ingram handle the issue
08:52 kmlussier csharp: I sent a question to my acq people but I haven't heard back.
08:53 kmlussier I suspect we may just omit that subfield altogether. But I don't know for sure. I do know our libraries are not using MarcEdit before importing acq records.
09:01 buzzy joined #evergreen
09:02 mmorgan csharp: I'm not directly involved in the acquisitions piece, but it's my impression that the $ has never been an issue for us, and many of our libraries have been importing Ingram records for some time.
09:07 csharp kmlussier: that's interesting... if they omit the subfield, do you know how they estimate price?
09:08 kmlussier csharp: They use a subfield in the holdings tag for price information.
09:09 csharp kmlussier: well, the order records I'm looking at don't have holdings information... do you know if the order records they're using do?
09:09 kmlussier Ah, I see now. Maybe it would be best if I send a response to the list, then. :) I just assumed you were using a holdings tag for other info.
09:10 csharp yeah, it's the 020 or 022 $c
09:10 csharp @marc 020
09:10 pinesol_green csharp: The ISBN assigned to a monographic publication by designated agencies in each country participating in the program. The field may include terms of availability and canceled or invalid ISBNs. It may be repeated for multiple numbers associated with the item (e.g., ISBNs for the hard bound and paperback manifestations; ISBNs for a set as a whole and for the individual parts in the set). (1 more message)
09:10 kmlussier I'll do it right now.
09:10 csharp @more
09:10 pinesol_green csharp: (Repeatable) [a,c,z,6,8]
09:10 kmlussier We generally use the 962 field for holdings info here, but you can use anything.
09:11 csharp pretty sure we use 952, but I don't remember off the top of my head - but that's with exports
09:11 csharp @marc 022
09:11 pinesol_green csharp: The ISSN, a unique identification number assigned to a continuing resource. (Repeatable) [a,y,z,2,6,8]
09:11 csharp hmm
09:13 csharp so the 022 doesn't have a $c
09:14 * csharp was just looking at the marc display attributes table where I believe our staff have been trying to fix this themselves (before berick clarified that those only affect display)
09:15 csharp I don't know what's stock and what's custom :-/
09:16 csharp it's the consequence of the fact that I'm really only involved with Acq when something's not working as expected - I usually have to backtrack over everything they've done without really knowing for sure what they've changed :-/
09:18 ningalls_ joined #evergreen
09:19 gmcharlt csharp: in the line item attribute definitions, there's a "remove" column that lets you specify a regexp
09:19 gmcharlt stuff that matches the regexp then gets removed when the acq MARC record is ingested
09:21 ericar_ joined #evergreen
09:28 Dyrcona csharp: When it comes to holding information in acq, it is all pretty much custom. :)
09:29 Dyrcona That is, local configuration.
09:29 * bshum thinks we don't use Ingram anymore for acq ordering in Evergreen.
09:30 bshum Mostly I hear about weird B&T stuff.
09:30 bshum So either we don't use Ingram or it works really well... probably the first one.
09:30 kmlussier We use Ingram a lot here. And we find it works pretty well.
09:31 bshum I think how kmlussier described holdings subfield for cost is how we deal with it too.
09:32 kmlussier By adding it to the holdings subfield, it automatically applies that price to the lineitem. I don't think that can happen if you're using the 020.
09:36 mmorgan csharp: gmcharlt: FWIW, In acq.lineitem_marc_attr_definition we have, in the "remove" field for ISSN and ISBN and UPC this string:  (?:-|\s.+$)
09:37 mmorgan We have that same "remove" string in every attr table associated with ISBN, ISSN, UPC.
09:37 csharp gmcharlt: berick told me yesterday that's just for display - is that true?
09:37 gmcharlt indeed - removing hyphens and anything after the first space
09:38 gmcharlt "12345679X (pbk.)" ... you will not defeat us!!! ;)
09:38 kmlussier csharp: I think that's a differnt thing. From what you said above, berick was referring to marc display attributes.
09:38 csharp in that case $ signifies "end of line", right?
09:39 gmcharlt csharp: yes
09:39 kmlussier Or maybe it's the same thing.
09:39 csharp ohh
09:39 csharp lemme look there
09:39 * kmlussier stops talking now because she is focusing on something else.
09:39 csharp god - so many moving parts :-/
09:42 Dyrcona csharp: It's what you asked for. :)
09:42 csharp Dyrcona: :-)
09:43 csharp argh - I wish I needed regexes more often so I would actually have to learn them :-/
09:44 Dyrcona csharp: Which kind? :)
09:44 Dyrcona jwz underestimated the number of problems. :)
09:44 Dyrcona To match a $ in Perl, you'll want \$
09:45 Dyrcona Otherwise, $ is treated as end of string.
09:45 csharp okay, I'm tranlating (?:|\s.+$) to mean "match either "?:"  or "\s.+" until the end of the line ($)
09:46 csharp but I don't know what ?: means here :-/
09:46 gmcharlt csharp: try this - [^0-9.]
09:46 gmcharlt i.e., moving anything that isn't a digit or a period
09:46 gmcharlt *removing
09:46 csharp gmcharlt: excellent - I'll try that
09:46 Dyrcona csharp; No, "(?:" starts a non-capturing group.
09:47 csharp Dyrcona: ah - thanks
09:47 csharp that was throwing me
09:47 Dyrcona csharp: It's like "(" but without back references.
09:47 Dyrcona It's a smidge faster if you don't need \1, etc.
09:48 csharp makes sense now
09:48 csharp I was reading ? as "the preceding character is optional" but that didn't make sense when it was first
09:48 gmcharlt (also, for the room - the relevant regexp_replace call in the extract_marc_field() stored function is invoking the replace with 'g' -- i.e., apply it to all matches)
09:49 ericar_ joined #evergreen
09:49 csharp hooray! that did it
09:49 csharp gmcharlt++ # thanks
09:49 csharp Dyrcona++ # helping translate ;-)
09:50 Dyrcona csharp: the original was (?:-|\s.+$) that means hyphen (-) or the first space (\s) followed by one or more of anything (.+) to the end of string ($).
09:50 csharp okay, so | still means "or" there - makes sense
09:51 Dyrcona Yep. The parens with ?: just make a non-capturing group out of it.
09:59 csharp kmlussier++ # responding on the list
10:00 csharp acq questions tend to result in radio silence in our experience, so I appreciate everyone who took time to respond! :-)
10:02 kmlussier Really? I usually try to answer whenever I see an acq question posted to the list.
10:03 bshum It wasn't asked that many hours ago.
10:08 jwoodard joined #evergreen
10:10 Dyrcona Well, more sites are using acq these days, so that experience is changing.
10:15 csharp Dyrcona: have you done any more experimentation with EG on 14.04?  looks to me like most of the bigger issues are fixed...
10:16 csharp I'm working on building our debs on 14.04 and jessie right now, but I'm deep in the weeds of packaging best practices right now
10:16 Dyrcona You mean for acq? If so, no. I don't really have much of a way to test, since I don't actually have any vendor accounts.
10:16 csharp nah, just in general
10:16 Dyrcona In general, I have been running Evergreen on 14.04 with success for about a year.
10:16 csharp I think I have the ruby problem solved for now
10:16 csharp oh good to know
10:16 Dyrcona yep. I saw that.
10:17 Dyrcona I thought about faking an EDI transaction for testing.
10:17 Dyrcona That is, setting up a bogus vendor with a canned order and response to send.
10:17 berick we've deployed 2 ubuntu 14.04 servers to productoin.
10:17 berick soon, soon we'll be free of centos
10:17 csharp berick: oh - awesome (on more than one level)
10:17 dbs centos! omg.
10:18 berick yeah
10:18 * csharp theoretically wants us to support CentOS/RHEL, but understands why it doesn't have traction
10:19 csharp I have an aging OpenSRF branch that installed dependencies from EPEL and got it up and running on CentOS
10:19 csharp 6, I think
10:19 csharp but I didn't get as far as Evergreen
10:22 bshum I think for front-end app servers, I'm not too nervous about Ubuntu 14.04.  But for backend utilities, I get a little anxious to test further.
10:22 bshum Acq EDI stuff, sure, but also have to test things like print notice generation, other scripts.
10:22 bshum z39.50, who knows?  :)
10:23 berick bshum: we should be fully migrated by the end of May.  i'll let you know how it goes
10:23 bshum berick++
10:24 csharp oh yeah - z39.50 is broken in 12.04 too, right?
10:24 csharp (as a server)
10:24 * bshum hasn't found that to be the case.
10:24 * csharp just remembers a bug about it
10:24 bshum Since he's using that on 12.04.  But anywho.
10:25 berick we're setting up a parallel ubuntu utility server and carefully migrating one thing at a time until there's nothing left to migrate :)
10:26 bshum csharp: You're probably talking about the errant https://bugs.launchpad.net/evergreen/+bug/1172332
10:26 pinesol_green Launchpad bug 1172332 in Evergreen "simple2zoom has segmentation faults" (affected: 1, heat: 8) [Undecided,Confirmed]
10:26 dbwells Not necessarily CentOS/RHEL specific, but this conversation reminded me of my brief experience with SELinux, which this fits to a tee: http://devopsreactions.tumblr​.com/post/82078906895/selinux
10:26 berick dbwells++
10:28 Dyrcona I believe z39.50 works on 14.04. It works on 12.04 with a more up to date libyaz from indexdata's PPA/repo.
10:29 dbs https://coffeecode.net/archives/25​9-Leaving-SELinux-in-enforcing-mod​e-with-Evergreen-on-Fedora-17.html
10:29 dbs SO OLD. And pretty hamfisted.
10:29 Dyrcona I've successfully searched without segfaults on 14.04.
10:30 csharp dbwells++
10:38 Dyrcona SELinux has its place, just not on my computers. :)
10:38 jboyer-isl dbs++ for sharing that. If I had known the only reason apache is run as opensrf is for logging I'd have stopped doing that a long time ago. I still want nothing to do with SELinux at the moment, but I did learn something!
10:43 rjackson_isl joined #evergreen
10:44 kmlussier gmcharlt: I need to post a job to evergreen-ils.org. It looks like we still don't have any direct links to job listings. Is that something I could add to About Us?
10:44 gmcharlt kmlussier: sure, that sounds good
10:44 kmlussier Or maybe to the Get Involved menu. No better way to get involved than to come work for MassLNC. :)
10:48 berick gmcharlt++ csharp++ # acq data cleanup.  sorry for misleading you, csharp
10:54 bshum I'm stepping away for a lunch meeting, but will be back this afternoon berick/dbwells/etc. to work on cutting releases.
10:55 kmlussier Speaking of cutting releases, I took a stab at point release notes last night for 2.8.1. last night.
10:55 jboyer-isl gmcharlt, kmlussier: Get involved does sound pretty good for job links. People may head there naturally, but I don't think About Us would immediately come to mind for that.
10:56 kmlussier My hope was to do a bulletted list, but there were already two lengthier release notes entry there, so it didn't work out that way.
10:57 kmlussier http://git.evergreen-ils.org/?p=workin​g/Evergreen.git;a=shortlog;h=refs/head​s/user/kmlussier/2-8-1-release-notes
10:57 kmlussier I'm open to any feedback
10:58 kmlussier jboyer-isl: Yeah, I can see that. I'm on the fence
10:58 kmlussier @coin
10:58 pinesol_green kmlussier: heads
10:58 * Dyrcona agrees with jboyer-isl.
10:59 jboyer-isl I think something like @coin "option1" "option2" could be a nice addition. Why make me map my random decisions in my head, just tell me what to do! :)
10:59 Dyrcona Does it make sense to add a Jobs category with only one posting at the moment?
11:00 Dyrcona jboyer-isl: That kind of thinking can get you replaced by a single line of Perl code. :)
11:00 jboyer-isl Dyrcona: Without a place to track them, how will people know when there's more than 1? (unless they're all from the same place, I suppose)
11:00 kmlussier Dyrcona: Well, the thing is, if you don't add the direct link to jobs, nobody remembers it is there and then don't think to post it to the Evergreen site.
11:01 Dyrcona See, that's where my thinking was headed about the Jobs category.
11:02 Bmagic when you make a org_unit opac invisible and run autogen, it should disappear from the opac dropdown right?
11:02 Dyrcona +1 on the release notes
11:02 berick Bmagic: may have to reload apache
11:02 Bmagic oh, just a reload? not a restart?
11:02 berick right
11:02 Bmagic I'll try that
11:03 Dyrcona Bmagic: If a restart didn't work, a reload won't either.
11:03 Bmagic I didnt restart apache
11:03 Bmagic I just ran autogen
11:03 Dyrcona Bmagic: OK, then, reload away.
11:03 Dyrcona ;)
11:03 Dyrcona I took your comment to mean that you had and it failed.
11:04 * Dyrcona reads too much between the pixels.
11:04 Bmagic berick++  #that did it
11:04 berick yay
11:04 Bmagic I dont understand why. Did autogen alter the apache config?
11:04 Dyrcona General rule of thumb, do an apache reload after autogen.
11:05 Bmagic Dyrcona: good to know
11:05 Dyrcona Nope. But, it alters data that is read by perl mods at startup.
11:06 Bmagic and a reload reinitializes all of the perl stuff?
11:06 Dyrcona A reload tells running Apache processes to finish the request that they are handling, then shut down.
11:06 Dyrcona When the next request comes in, it should spawn a new Apache child.
11:06 Dyrcona Thus rereading the data.
11:07 Dyrcona A restart says, shut down immediately, I don't care what you're doing.
11:07 Bmagic so, it really is a restart but with a graceful first. apache2 reload = apache2 graceful-stop && apache2 start   ?
11:08 Dyrcona That's one way to think of it, but I've never looked at the code or scripts to see what is actually done.
11:08 Bmagic thanks guys!
11:08 Dyrcona yw
11:08 berick with a reload, the master apache process is not restarted.  it stays alive to queue requests while the children (gracefully) die off and start back up
11:09 berick in case anyone is curious
11:20 pinesol_green [evergreen|Jason Boyer] LP#1427309 Fix Crash in Collections User Balance Summary - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=804210a>
11:23 kmlussier Nice! The job automatically posts to all the social networking spots.
11:24 bmills joined #evergreen
11:27 Newziky joined #evergreen
11:28 kmlussier That is, everything but Twitter. hmmm
11:30 kmlussier Probably because we don't use JetPack to post to Twitter.
11:35 Dyrcona @@++
11:35 pinesol_green Dyrcona: Zoia knows how to make fusilli.
11:35 Dyrcona Good for zoia!
11:35 Dyrcona pinsesol_greeen: I make a killer London Broil.
11:36 Dyrcona Doh! That's what I get for being cheeky.
11:38 kmlussier Sounds like pinesol_green sometimes sneaks over to code4lib for dinner with zoia.
11:41 Dyrcona The increment above was intended for the @@ operator in Postgress. I figured it would elicit nonsense from pinseol_green.
11:42 Dyrcona And time for some left over, killer london broil. ;)
11:45 berick pinesol_green and zoia...  dining philosophers
11:45 pinesol_green berick: Yeah, well, you know, that's just like uh, your opinion, man.
11:45 pinesol_green berick: I am only a bot, please don't think I'm intelligent :)
11:52 grharry1 joined #evergreen
12:05 csharp berick: no problem - just glad there was an easy front-end solution :-)
12:05 berick csharp: it takes a village
12:06 csharp @who is one of the Village People?
12:06 pinesol_green hopkinsju is one of the Village People.
12:10 Bmagic lol
12:18 buzzy joined #evergreen
12:49 Newziky left #evergreen
12:57 yboston joined #evergreen
13:58 jihpringle joined #evergreen
14:05 Dyrcona Meh, Launchpad....
14:09 buzzy joined #evergreen
14:40 jihpringle joined #evergreen
15:20 maryj joined #evergreen
15:47 abowling joined #evergreen
15:57 kmlussier @karma
15:57 pinesol_green kmlussier: Highest karma: "dbs" (992), "bshum" (966), and "Dyrcona" (663).  Lowest karma: "ie" (-62), "^" (-30), and "marc" (-28).  You (kmlussier) are ranked 5 out of 2470.
15:58 kmlussier I've heard that code4lib clears out its karma every year around conference time. Is there any interest in doing that here?
15:59 kmlussier It would give new community members a chance to rank highly along with those of use who have been around for a while.
16:00 vlewis joined #evergreen
16:00 Dyrcona Sounds fine to me.
16:01 kmlussier @karma
16:01 pinesol_green kmlussier: Highest karma: "dbs" (992), "bshum" (966), and "Dyrcona" (663).  Lowest karma: "ie" (-62), "^" (-30), and "marc" (-28).  You (kmlussier) are ranked 5 out of 2470.
16:01 kmlussier Sigh...that's not what I meant to do.
16:01 kmlussier @karma kmlussier
16:01 pinesol_green kmlussier: Karma for "kmlussier" has been increased 622 times and decreased 0 times for a total karma of 622.
16:02 kmlussier On the other hand, if I try really, really hard, I do have a chance of sneaking past Dyrcona in the next year. ;)
16:02 bshum kmlussier: We'll just volunteer you to be our next RM and you'll definitely get past him ;)
16:02 Dyrcona Yes, you do.
16:03 Dyrcona kmlussier++
16:03 bshum Or you know, just general awesomeness.
16:03 bshum kmlussier++
16:03 kmlussier bshum: Somehow, I don't think the payoff is worth the trouble of being RM.
16:03 bshum But, but... karma!
16:03 bshum :D
16:04 Dyrcona hah
16:09 jihpringle_ joined #evergreen
16:27 grharry1 left #evergreen
16:28 Bmagic @karma Bmagic
16:28 pinesol_green Bmagic: Karma for "Bmagic" has been increased 27 times and decreased 0 times for a total karma of 27.
16:28 Bmagic Watch out! I'm coming!
16:29 bshum Bmagic++ # for future good deeds (and some occasional past good ones, including epic magic tricks)
16:29 Bmagic lol
16:30 kmlussier Bmagic++
16:30 Bmagic Awwww dang! Watch out Dyrcona, Im gonna getcha
16:30 kmlussier Bmagic: I've been thinking it's time to schedule our next Bug Squashing Day.
16:31 Dyrcona Bmagic++
16:31 Bmagic I love bugs, I am excited to be apart of them!
16:31 kmlussier It's supposed to be in May, but since the conference is in May and since the last one was a little late, I'm thinking of pushing it into June.
16:31 Bmagic Just don't make me eat them
16:31 kmlussier Ewwwww!
16:32 bshum "This job is not worth eleven five a year"
16:32 kmlussier Anyway, that's for another day.
16:32 * kmlussier needs to disappear
16:32 * Bmagic waves
16:32 mmorgan kmlussier: Have a good evening!
17:15 bshum Hmm
17:16 bshum We can't seem to add any bibs with URIs to lists.
17:16 bshum That's special.
17:16 pinesol_green Incoming from qatests: Test Success - http://testing.evergreen-ils.org/~live/test.html <http://testing.evergreen-ils.org/~live/test.html>
17:18 mmorgan bshum: lp 1246385 ?
17:18 pinesol_green Launchpad bug 1246385 in Evergreen "internal server error when adding an item to a list from within staff client" (affected: 1, heat: 6) [Undecided,Confirmed] https://launchpad.net/bugs/1246385
17:18 bshum No, this is in public side
17:19 bshum If I go to try adding a bib that's a URI (aka, $9 trick), it just stays off the list and I get a refreshed page that still says "add to list"
17:19 mmorgan Oh, ok. sounded familiar. But it's too late in the day...
17:19 bshum I'm going to have to get a test system going with 2.8 stock to see if it's a template issue in our systems, or if it's a 2.8 regression.
17:20 bshum It's not broken on MVLC's or PINES systems, but they're on 2.7ish
17:20 bshum mmorgan: We should probably close that bug.  Since no one's confirmed anything since it was reported "fixed" or at least "not broken" for 2.5.
17:22 mmorgan Oh, probably so.
17:24 mmorgan left #evergreen
17:37 kmlussier bshum: There's a bug for that. And old one
17:38 kmlussier bug 979038
17:38 pinesol_green Launchpad bug 979038 in Evergreen "tpac: Records with a located URI cannot be added to a list" (affected: 3, heat: 18) [Low,Confirmed] https://launchpad.net/bugs/979038 - Assigned to Jeff Godin (jgodin)
17:40 kmlussier Filed that one while C/W MARS and NOBLE were getting ready to go live on Evergreen.
17:40 bshum kmlussier: That sounds more like my problem.
17:40 bshum I didn't notice it on MVLC/PINES cause I was only testing temp lists on there (no account to test real with)
17:40 bshum But temp lists don't even take ours on our systems.
17:40 bshum But maybe that's cause we don't use the temp list warning or whatnots.  Who knows.
17:41 bshum Sigh, oh well.
17:41 * bshum clicks "affects me too" and contemplates dinner.
17:43 kmlussier jeff: Do you still want to be the assignee on that bug?
18:16 bshum csharp: Should we be deprecating https://bugs.launchpad.net/evergreen/+bug/1050384 in favor of your newly filed https://bugs.launchpad.net/evergreen/+bug/1450218
18:16 pinesol_green Launchpad bug 1050384 in Evergreen "Views referenced in reporter UI are not created by default" (affected: 1, heat: 6) [Low,Confirmed]
18:16 pinesol_green Launchpad bug 1450218 in Evergreen "PINES example reports available in fm_IDL.xml by default, but not installed in DB by default" (affected: 1, heat: 6) [Undecided,New]
18:17 bshum Or vice versa.  One of those should be marked duplicate of the other I think.
18:19 bshum And as for stat cats, that can do strange things when you either a) have something else as "1" or "2" than what's expected, or b) not have id 1 or 2 at all, cause our stat cats happen to start at 100+
18:19 bshum So some of us manually edit and remove those from the view definitions prior to creation.  To avoid weirdness.
18:28 mrpeters left #evergreen
20:56 bmills joined #evergreen
21:20 maryj joined #evergreen
21:38 bmills joined #evergreen
22:35 * dbs hates marking older bugs as duplicates of newer bugs because it masks how freaking old the underlying problem is
22:38 bmills joined #evergreen
22:59 bmills joined #evergreen

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