Evergreen ILS Website

IRC log for #evergreen, 2017-04-21

| 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
04:31 pinesol_green News from qatests: Test Success <http://testing.evergreen-ils.org/~live>
05:58 Callender joined #evergreen
06:40 rlefaive joined #evergreen
07:10 rjackson_isl joined #evergreen
08:21 Dyrcona joined #evergreen
08:27 dteston joined #evergreen
08:33 mmorgan joined #evergreen
08:34 Dyrcona So, with the release of 2.12.1, shouldn't working/user/blake/tags/rel_2_12_1 be copied to tags/rel_2_12_1?
08:49 bos20k joined #evergreen
08:49 Dyrcona I forgot about setting the pager to none when running the upgrade script....
08:49 Dyrcona So my time call will be irrelevant.
08:49 Dyrcona invalid, whatever.
08:56 gmcharlt Dyrcona: yes, it should be pushed
08:56 gmcharlt I'll do it now
09:00 Dyrcona OK.
09:06 dbwells gmcharlt: I am about to push the 2.12.1 upgrade to master/rel_2_12, FYI
09:06 gmcharlt dbwells: whoops, I think I just beat you to it
09:07 dbwells gmcharlt: doh!  That's okay :)
09:07 gmcharlt one question for dbwells and bshum: should we cherry-pick the PO and POT updates from the tag branch into rel_2_12, or reserve that for 2.12.2?
09:07 dbwells Just stuff I told Bmagic I would take care of.
09:08 pinesol_green [evergreen|Galen Charlton] forward-port 2.12.0-2.12.1 database update - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=4d636e5>
09:09 collum joined #evergreen
09:15 dbwells As long as Bmagic is handling the 2.12 builds, cherry-pick back to rel_2_12 seems like the easiest workflow to me, and I don't think it will make a difference.
09:35 dbwells bshum: before we add the new pot files back to rel_2_12, I have a burning question I've always wondered about.
09:36 bshum dbwells: gmcharlt: Yeah, I'd like to double check the contents before we merge it.
09:36 bshum Just to make sure we didn't have any oddities
09:37 bshum It looks okay from a glance
09:37 dbwells bshum: Our standard practice is to exclude from the new pot commit "non-trivial changes".  Is that solely to make the new pot commit cleaner, or does, for example, updating the "POT-Creation-Date" have some other undesirable side-effect on the LP side?
09:38 bshum dbwells: My understanding is that it was a practice adopted to avoid seeing unnecessary churn in the git history for the files
09:38 bshum To my knowledge it should have no impact to LP side of things
09:39 Callender joined #evergreen
09:39 dbwells So, perhaps nice to think about, but nothing to fret about.
09:40 bshum That's my opinion.
09:40 dbwells Sounds good to me.
09:41 bshum gmcharlt: Eyeballing the POT and PO changes, nothing looks alarming to me from Bmagic's branch.  Looks like building it on 14.04 with the older translate toolkit was a safe call
09:42 Dyrcona Yeah, my understanding is the same as bshum on the pot exclusion.
09:42 Bmagic yeah, I looked through it, looked ok
09:42 bshum Bmagic++
09:42 dbwells I'm not sure the git history of an autogenerated file is ever going to be particularly useful :)
09:47 dbwells bshum: Are you going to to ahead and push the pot/po bits to rel_2_12?  If not, I can take care of it whenever you're done looking them over.
09:50 rlefaive joined #evergreen
09:54 bshum dbwells: Feel free to push away
09:59 dbwells bshum: will do, thanks
10:06 jvwoolf joined #evergreen
10:15 dbwells bshum: So, I noticed in this po push the final deletion of string like "Example Branch 2".  It seems like we would still want those translated for anyone trying out Evergreen.  Perhaps we should add Open-ILS/tests/datasets/sql/libraries.sql to the translation process?
10:16 bshum dbwells: I noticed that too for other things I was testing
10:16 bshum dbwells: And yeah, adding the sample dataset as a translation option sounds like a good idea
10:17 bshum But building that out will require a little more thought I think
10:17 bshum Probably similar to how we do db.seed-data-values.sql now I guess.
10:18 gmcharlt maybe? sample data (as opposed to test case data) strikes me as something where we don't necessarily want _mechanical_ localization
10:18 gmcharlt or, rather, where such a thing would be less ideal than getting somebody from the relevant locale to put together realistic data
10:19 gmcharlt anyway, here's an easy-peasy webstaff pullrequest for someobdy to test and merge: bug 1685232
10:19 pinesol_green Launchpad bug 1685232 in Evergreen "webstaff: pcrud.apply() does not work" [Medium,New] https://launchpad.net/bugs/1685232
10:19 gmcharlt also pointing out that pcrud.apply(), when it works, can be quite handy
10:21 berick gmcharlt: hm, i thought I fixed that already.
10:21 * berick scratches head
10:21 gmcharlt berick: heh. I think miker or I also fixed that already in a much larger feature branch
10:21 gmcharlt which is why I decided to make a micro bug so that WE CAN ALL STOP FIXING THE SAME BUG!!! ;)
10:21 berick i bet it's lingering in one of my pullrequests
10:21 berick exactly
10:22 miker berick: yeah, I seem to remember you doing that, but launchpad search is so terrible...
10:23 gmcharlt berick: also, I'm thinking of writing a fancier version of apply() that would do things like deflesh has_a linked objects and dive into hash_many arrays
10:23 berick miker: yeah, and this is going to (*ahem*) bug me.  i must find it!
10:23 berick in the meantime, I'll look at the bug
10:24 berick gmcharlt: cool
10:24 gmcharlt that way, one could use a fully flesh results of a pcrud search, possibly passed to and from a "typedHash" (and I need a better name for that), set isnew() etc., and have it jsut do the Right Thing
10:27 rlefaive joined #evergreen
10:34 rlefaive joined #evergreen
10:39 Dyrcona rhamby++ # I was going to point out the low traffic on the catalogers' list and about the development discussion.
10:39 * Dyrcona sometimes thinks the catalogers' list is redundant. :)
10:39 rhamby Dyrcona: yeah, I actually had to check to see if I was still subscribed since I didn't remember any regular traffic
10:40 Dyrcona I should probably subscribe to the documentation list.
10:41 rlefaive joined #evergreen
10:42 pinesol_green [evergreen|Galen Charlton] LP#1685232: fix egCore.pcrud.apply() - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=8817b75>
10:46 csharp ALL THE DEVS FIX ALL THE BUGS!
10:47 Dyrcona heh.
10:47 Dyrcona I was thinking: SQUASH ALL THE SILOS!
10:47 Dyrcona berick++ That was fast.
10:48 berick if we had one mail list for then entire EG community, it would still be a slow list
10:48 Dyrcona It would.
10:48 berick my point being, of course, that more lists seems odd to me
10:48 * csharp , scarred from the "admin" email list discussions of c. 2011-12, hides in a bunker until all email list discussions are done
10:49 bshum We, few, we happy few...
10:49 dbs my stance concerning new mailing lists as expressed during the "admin" days stands
10:49 Dyrcona Well, the circulation list discussion got me to question why does the catalogers' list exist.
10:50 dbs evil idea: make the url for the new list be an alias for an existing list
10:51 * berick chuckles
10:51 dbs mozilla is moving all of their mailing lists to Discourse; it seems like a decent platform tbh
10:51 Christineb joined #evergreen
10:52 Dyrcona The sysadmin list is still mentioned here: https://evergreen-ils.org/​communicate/mailing-lists/
10:52 bshum Maybe we should change the name when we migrate general to the lists server (general is still hosted on GPLS lists, right?) and we just called it "evergreen" and make it one list for everything then :D
10:52 dbs Dyrcona: probably so the archives can be found?
10:52 bshum Well, okay maybe not evergreen.
10:52 bshum But uh, "discussion" haha
10:53 Dyrcona dbs: Sounds legit. :)
10:53 Dyrcona I should read the fine print. :)
10:54 csharp we should just move to Slack!
10:54 * csharp ducks
10:55 dbs gitter, heh
10:55 Dyrcona IRC for all discussion. :)
10:55 Dyrcona heh. Since we're already installing ejabberd.... :)
10:55 * csharp weeps for the glory days of #code4lib's IRC channel
10:56 * berick too
10:56 * dbs too
11:04 miker Dyrcona: re bug 1684984, it looks like ingest.disable_metabib_field_entry was made obsolete with the flags for skipping specific uses of metabib.field_entry stuff; facet, browse, search.  It was likely the early version of ingest.skip_search_indexing
11:04 pinesol_green Launchpad bug 1684984 in Evergreen "Internal Flag ingest.disable_metabib_field_entry not used" [Undecided,New] https://launchpad.net/bugs/1684984
11:05 Dyrcona miker: That was my thought immediately after filing the bug. So, should the flag be removed from the table?
11:06 miker +1 I think.  I'd say "set the three if you want that".  one flag could be slightly cheaper, but it's just asking for confusion
11:06 Dyrcona I'll update the bug and prepare a branch.
11:12 * csharp decides to leave Ubuntu desktop in the dust, moves back to Fedora, prolly for good :-/
11:12 Dyrcona hmm...
11:12 berick csharp: do tell
11:12 * Dyrcona is considering switching to something else, also. Since 16.04, Ubuntu has been in decline.
11:12 Dyrcona @blame systemd
11:12 pinesol_green Dyrcona: systemd stole Dyrcona's ice cream!
11:12 Dyrcona :)
11:13 csharp berick: oh, I'm... what Dyrcona said - they're dropping Unity, etc. and more or less throwing in the towel
11:13 Dyrcona But, my hmm was about targeting bugs on launchpad.
11:13 berick i've been quite happy w/ xubuntu
11:13 Dyrcona Unity 8 stinks-- UI by Crayola.
11:14 Dyrcona Unity not *8 is OK.
11:14 berick xfce++ # what gnome users really want
11:14 Dyrcona BlackBox, baby!
11:14 Dyrcona :)
11:14 Dyrcona @karma stinks
11:14 pinesol_green Dyrcona: Karma for "stinks" has been increased 0 times and decreased 1 time for a total karma of -1.
11:14 berick yeah, blackbox  / fluxbox is great
11:15 csharp I'm on Ubuntu GNOME 17.04 and it's fine, but if I'm living in GNOME, Fedora is better
11:15 csharp but I'm open to other DEs too
11:15 Dyrcona BlackBox is not a DE, it's a WM.
11:15 * csharp always had a hard time remembering the difference :-)
11:15 Dyrcona Just to clarify things.
11:16 * miker considers going back to E
11:16 gmcharlt regarding the lists discussions... while I share the general lack of optimism that a new one would get used, I also think that it's the sort of thing that is better to either let die after disuse (or succeed!) rather than have a bunch of devs squash each time
11:16 Dyrcona I never used E much, but tried it for a bit.
11:17 berick gmcharlt: agreed.
11:18 * Dyrcona sort of agrees. But disagrees strongly if acq. ui discussion is going to happen in a silo.
11:18 Dyrcona It should happen in a different silo. :)
11:18 berick heh
11:19 mmorgan gmcharlt++
11:19 berick if it's discussion that otherwise would not have happened, then by all means, add a list.
11:19 * mmorgan feels like there must be a lot of questions out there that aren't getting asked.
11:20 Dyrcona Why? Because there is not a specific list for them?
11:20 Dyrcona Now, I like the idea of one list for everything even more. :)
11:20 jeff Dyrcona: can you explain the "disagrees strongly if acq. ui discussion is going to happen in a silo. / It should happen in a different silo." comment above -- or just let me know if I missed the joke? :-)
11:21 Dyrcona jeff: I think the discussion of acq. ui development should happen on the development list and not an acq. list.
11:21 jeff aha
11:22 mmorgan Dyrcona: Because folks that have questions about using the system on a daily basis might not feel comfortable joining and posting to a list where questions like "I'm installing evergreen and..." come up
11:23 Dyrcona Well, the general list is for any question.
11:24 Dyrcona I thought the idea was that installation questions should go to the development list, but we can't control what lists people send to.
11:24 berick do we just need a staff list?
11:26 jeff i don't think so.
11:26 dbs csharp++ # welcome back to Fedora
11:26 csharp seems like the factors involved on the "pro new list" side are 1) not wanting to ask something in the "wrong" place 2) not wanting to "junk up" a general list with topic-specific discussions and 3) librarians naturally like neat categories for things
11:26 jeff i think we can create an acq and a circ list and see where they go from there.
11:27 dbs kind of what we did with sysadmin
11:27 jeff and maybe it'll turn out the same way that sysadmin did, and maybe not.
11:27 * Dyrcona has seen this movie before, and I was on the other side of the discussion the last time.
11:28 dbs as I was walking back from the meeting I was in, I was reflecting that people communicating primarily via IRC are not necessarily in the same circle in the Venn diagram of communications as those who prefer mailing lists
11:28 csharp dbs: thanks - I've been running F25 on my laptop since the fall with no problems - using Fedora with nvidia drivers is frustrating (rpmfusion lag, can't use Wayland), so my desktop stations have been happier on Ubuntu
11:28 mmorgan Seems like a good time to try what jeff suggests since two interest groups at the conference both saw a need.
11:29 dbs csharp: ah, I've only used nouveau
11:30 csharp I do just enough gaming that nouveau doesn't do the job for me :-)
11:41 Bmagic Anyone heard of using a selfcheck machine where certain items are checked out and then immediately checked back in on the server (tracking in-house purposes) ?
11:44 jvwoolf left #evergreen
11:58 khuckins joined #evergreen
12:08 miker Bmagic: never ... and, of course, we have a separate in-house-use function, so the numbers would be wonky (compared to the designed workflow)
12:09 Bmagic miker: yeah, I thought it was a strange concept. Apparently the SIRSI system did this for the library
12:10 Bmagic Any solutions for SIP and In-House use?
12:10 dbs more like a self-in-house system - I could see that as being a potentially useful thing, kind of like my hacked-together simple web UI for moving items to a different location or deletion
12:13 jihpringle joined #evergreen
12:13 berick Bmagic: this is for patrons?
12:14 Bmagic yep
12:16 Bmagic Things like paperbacks and magazines and even seed packets where the library doesn't care that they bring them back. So a traditional circulation will end up getting marked lost at some point
12:16 Bmagic They do however, need to know that they were checked out, for inventory reasons
12:18 bshum Are these items actually inventoried?  meaning attached barcodes on records, etc.
12:19 Dyrcona Wonder if you can set a loan duration to infinity?
12:19 bshum Cause hey, if you had the workstation registered for the selfcheck, you could trace all circs applied there or the selfcheck user account too, and then setup a script to periodically check back in the copies (using some flags to disable things like transit, hold triggering, etc.) ?
12:20 bshum Dyrcona: 999 years, obviously?  :D
12:21 Bmagic bshum: yes, they are cataloged with barcodes
12:22 Bmagic I would like to avoid more one-off software for this one library
12:22 Bmagic I offered them a solution that would deny the checkout, forcing the patron to come to the desk, so that the staff can perform an in-house checkout
12:23 Bmagic it hurts a little bit, because they were used to it working just fine
12:23 Bmagic It was called a "ephemeral" checkout
12:24 Dyrcona bshum: Interval can apparently go to 178000000 years
12:24 Dyrcona I was hoping for an actual inifinity, though.
12:25 Dyrcona Though infinity is a special value for date/time fileds.
12:26 Dyrcona I suppose with some modification infinite check outs could be supported.
12:26 jeff Bmagic: non-cat circ via sip2 was something we considered doing work to support. it looked reasonable.
12:27 Dyrcona Anyway, I should grab something to eat.
12:27 Bmagic jeff: yeah, I think that is the real answer
12:27 Bmagic however, does SIP have this type of thing in the spec?
12:27 Dyrcona jeff: I ran into some issues with that and NCIPServer, though I'm fuzzy on the details.
12:27 jeff Bmagic: our use case was paperbacks, and we were going to either rfid tag them with all the same thing, or with a given prefix and a meaningless serial number.
12:27 jeff Dyrcona: non-cat or pre-cat?
12:28 jeff Bmagic: not officially spelled out that i've seen, but at least one vendor has "support generic all-the-same item ids" for just this kind of thing.
12:28 Dyrcona pre-cat prolly. Whatever the option name that I'm too lazy to look up right now says. :)
12:29 Bmagic best case, we have a special in-house category and the self check would know based on the barcode which category to assign the in-house checkout
12:29 jeff Dyrcona: pre-cats are "this item has a barcode and maybe a dummy title/author/isbn/circ_mod"... non-cats are "this is a PAPERBACK, or a MAGAZINE, it has no barcode, we don't care if it comes back, there are no overdues, etc"
12:29 mmorgan Bmagic: noncat circ seems like it would save them some work, if they don't need to know the details on the items, just counts.
12:29 mmorgan They wouldn't need to catalog them.
12:29 Dyrcona Long term, I think an infinite checkout duration that can be controlled by the matchpoints is the way to go.
12:30 jeff and in-house use is something else entirely -- and doesn't seem to fit what you're describing.
12:30 jeff Dyrcona: i agree with you there. :-)
12:30 Dyrcona jeff: yeah, pre-cats. There was a branch to make it easier/better, but it still didn't do what I expected and with a deadline looming, I gave up on making it work.
12:30 Dyrcona I was going to use them for incoming items as an option.
12:31 * Dyrcona goes to get something to eat for real this time.
12:31 Bmagic I probably need more information from the library to decide.
12:32 jeff adding support to sipserver for generic item ids or item id prefixes that translate to the same thing that you get in the staff client when checking out a non-cat item is something i would probably be interested in helping with.
12:32 mmorgan So, infinite-term checkout, the items would always show checked out to the patron? That may not be what they're looking for.
12:32 jeff mmorgan: i don't think anyone has voiced a need for that.
12:32 jeff mmorgan: i think Dyrcona was pondering it as a means of accomplishing something, but then dismissed it. :-)
12:33 mmorgan Ok, gotcha
12:33 jeff especially in what Bmagic has said, i don't think he wants/needs that based on "the library doesn't care that they bring them back"
12:34 mmorgan Right. Sounds like a non-cataloged checkout
12:34 Bmagic yeah, I am leaning toward non-cat now after this discussion
12:36 jeff Bmagic: it would be good if you could determine from the library how their selfchecks supported this -- are the items in question barcoded or rfid tagged? was there an on-screen "i am taking 3 paperbacks" option? are the item ids if present all identical for a given type, or have a serial number of some kind?
12:37 Bmagic jeff: the impression I got, was the machine didn't know the difference. The server however, knew that this type of item needed to be checked back in immediately.
12:37 jeff ideally the sip clients send a checkout message, and the item identifier in the checkout message contains something that you can use to determine the "type" of item. beyond that, it doesn't much matter.
12:38 Dyrcona Yeah, inifinite checkout has issues. (My lunch is heating up.)
12:39 Dyrcona Basically, you want to give something to a patron but have it counted as a checkout for stats and other purposes, but not count as an item out for the patron?
12:39 Dyrcona I'm leaving selfchecks out of it for now.
12:39 jeff right, that's what non-cat does.
12:40 Dyrcona It does? I thought they still had due dates... I guess I'd better review that again.
12:40 jeff they aren't circs, though. you report on them separately, or you join them together in a report via one of a handful of other ways.
12:41 Dyrcona Oh, right.
12:41 Dyrcona I'm always confusing that with pre-cat. :)
12:41 jeff also, non-cats can have in-house uses as well as "circs" (that aren't circs)
12:41 jeff action.non_cataloged_circulation vs action.non_cat_in_house_use
12:42 rlefaive joined #evergreen
12:43 jeff entries in config.non_cataloged_type do have a boolean for in_house and also have a circ_duration interval
12:47 Bmagic what does the hold targeter do when it encounters an item with age protection and the settings are setup to compare to active_date but the active date is null?
12:48 berick Bmagic: looks like it defaults to NOW()
12:48 jeff non-cataloged circulations do not appear to patrons in tpac, and they don't appear by default in the staff client unless you specifically select non-cataloged circulations in the items out views in staff clients.
12:48 rlefaive joined #evergreen
12:48 Bmagic berick: oh sweet
12:49 jeff and webby at least seems to have an issue either with checking out a non-cat item or showing it.
12:49 jeff disregarding that for a moment, the non-cat "circs" go away completely in the staff client once they're past their "due date"
12:50 jeff and i think that's circ_time + interval, meaning if you change the interval it'll affect the display of existing non-cat "circs"
12:51 jeff (i don't consider that a big issue -- but i can imagine that it might surprise someone, so it should probably at least be documented if not changed)
12:51 jeff (and i'm not advocating changing it)
12:51 Dyrcona So an interval of 0 minutes makes them disappear.
12:51 Dyrcona Or a similarly short interval, like 1 minute.
13:00 khuckins_ joined #evergreen
13:27 khuckins__ joined #evergreen
13:35 rlefaive joined #evergreen
14:45 gmcharlt https://evergreen-ils.org/everg​reen-3-0-development-update-2/
14:47 bshum gmcharlt++
14:54 berick gmcharlt++
15:00 mmorgan gmcharlt++
15:02 remingtron bird_jigsaw_puzzles++
15:04 abneiman gmcharlt++ # privacy, offline, AND duck trivia!
15:11 gmcharlt my_friend_arlene++ # every birder in my life is now at risking of being asked for all their pictures of ducks
15:18 jeff heh. that moment when you return to your irc window only to think, "wait! where did the last hour or two of scrollback go," before realizing that said scroll was all privmsg.
15:19 jvwoolf joined #evergreen
15:21 jeff er, that sounded more mysterious than intended.
15:39 Jillianne joined #evergreen
15:54 khuckins_ joined #evergreen
16:29 khuckins__ joined #evergreen
16:31 pinesol_green News from qatests: Test Success <http://testing.evergreen-ils.org/~live>
17:04 mmorgan left #evergreen
17:19 jvwoolf left #evergreen
21:02 khuckins__ joined #evergreen

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