Evergreen ILS Website

IRC log for #evergreen, 2018-02-22

| 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
00:29 Akum joined #evergreen
00:30 Akum Hi
05:54 yar joined #evergreen
06:01 dbs joined #evergreen
06:01 jeff joined #evergreen
06:02 serflog joined #evergreen
06:02 Topic for #evergreen is now Welcome to the #evergreen library system channel! | We are publicly logged: http://irc.evergreen-ils.org/evergreen | Large pastes at http://paste.evergreen-ils.org
06:25 pastebot joined #evergreen
06:25 dbs_ joined #evergreen
06:31 pinesol_green News from qatests: Failed Running Evergreen browser client build/test - Expected 6 errors but encountered 2. <http://testing.evergreen-ils.org/~live>
06:31 pinesol_green News from qatests: Failed Running perl live tests <http://testing.evergreen-ils.org/~live>
06:31 jeff___ joined #evergreen
06:31 RBecker_ joined #evergreen
06:38 serflog joined #evergreen
06:38 Topic for #evergreen is now Welcome to the #evergreen library system channel! | We are publicly logged: http://irc.evergreen-ils.org/evergreen | Large pastes at http://paste.evergreen-ils.org
07:02 agoben joined #evergreen
07:11 rjackson_isl joined #evergreen
07:31 rlefaive joined #evergreen
07:43 rlefaive joined #evergreen
07:53 akilsdonk joined #evergreen
08:04 rlefaive joined #evergreen
08:14 rlefaive joined #evergreen
08:18 collum joined #evergreen
08:34 ngf42 joined #evergreen
08:36 mmorgan joined #evergreen
09:01 dbs joined #evergreen
09:05 csharp shared in #code4lib: https://github.com/npm/npm/issues/19883
09:05 csharp yowza
09:13 kmlussier joined #evergreen
09:17 dbwells whoa
09:17 ngf42 O_O https://media.giphy.com/med​ia/DGiZfWmc0HWms/giphy.gif
09:25 yboston joined #evergreen
09:27 Bmagic yeah, that is a doozy
09:27 Bmagic csharp++
09:30 dkyle joined #evergreen
09:31 JBoyer csharp++
09:32 miker that's so bad that I'm afraid to click on the link to the commit...
09:32 JBoyer miker, it's worth it to see what they called the function call.
09:33 JBoyer named, rather.
09:34 miker HA.. first, the old name is GOLDEN. "make derp". yes, yes you did
09:34 dbwells ah, to laugh or to cry
09:35 miker second, "you keep using this word 'correct' ..."
09:35 dbs Do we recommend using "sudo npm" in our install docs at any point?
09:35 dbs (or running npm commands as root, perhaps in some broader context)?
09:36 jeff___ joined #evergreen
09:36 JBoyer I don't believe we do, it appears to be run as opensrf normally.
09:36 dbs http://docs.evergreen-ils.org/reorg/dev/command_l​ine_admin/_extra_steps_for_web_staff_client.html doesn't say what user to run those steps as
09:37 jeff do we document steps where we use npm to update/upgrade npm itself?
09:37 dbs but the previous step at http://docs.evergreen-ils.org/reorg/dev/comma​nd_line_admin/_installing_prerequisites.html are to be done as root; and normally we say "Perform the following steps as the XXX user", so it's likely some people are running the npm steps as root
09:37 jeff it looks like that's the only way you end up with the (unclearly tagged) pre-release version with this (impressive) bug.
09:38 miker dbs: grep says we have, but it doesn't look like that's the case in recent master
09:38 jeff and running as root (without sudo) doesn't trigger the bug, at least per what portion i've read so far.
09:38 miker have recommended 'sudo npm' in the past, I mean
09:39 dbwells Yeah, it seems our npm instructions have gotten more fuzzy recently, but I wasn't sure of the intention.
09:40 JBoyer jeff, well, most of the things touched should already be owned by root; without poring over it in depth I would guess that appears to work as a lucky accident
09:40 Dyrcona joined #evergreen
09:42 dbs http://docs.evergreen-ils.org/2.12/_optional_e​xtra_steps_for_browser_based_staff_client.html says "sudo npm update"
09:43 dbs but that's after installing node from source, which...
09:43 dbs v0.10.28 at that
09:45 bshum Hmm
09:45 bshum The main readme for 2.12 doesn't say to install that source anymore
09:45 bshum Sounds like something slipped through the cracks in the docs
09:48 Dyrcona Yeah. You should be installing 6.11 something.
09:48 bshum dbs: http://docs.evergreen-ils.org/2.12/_optio​nal_extra_steps_for_web_staff_client.html ; when I navigate through the doc book, it changes the page to this one
09:48 bshum Which just says to install the nodejs LTS
09:48 bshum I don't know how that extra page ended up in the middle that you found :)
09:48 bshum But it's an old page
09:48 Dyrcona Did 2.12 get updated to install the LTS?
09:48 bshum And probably shouldn't be there
09:49 bshum Dyrcona: Yeah kmlussier pushed that fix into 2.12 series cause of the problems with old Node
09:49 Dyrcona Ah, cool. I haven't checked.
09:50 littlet joined #evergreen
09:51 dbs bshum: seems likely that the doc build process doesn't clean up the directories after they're built; I found that page through Google
09:52 dbs "site:docs.evergreen-ils.org npm"
09:53 Dyrcona Documentation lies.... :)
09:53 * Dyrcona just had to say it.
09:53 Dyrcona So, who's "in charge" of the doc process these days?
09:54 * dbs also wonders where "the previously described Optional: Developer Addition" can be found (as referenced in current docs)... heh
09:54 Dyrcona That's the distro-release-developer target.
09:54 dbs grep says it can't be foudn in the docs
09:55 Dyrcona My advice is never use the online installation docs. Always use the README.
09:55 Dyrcona It's 5. in the README.
09:57 dbs README is a symlink to docs/installation/server_installation.adoc
09:57 Dyrcona Yes, I'm aware of that.
09:57 dbs "Developer Additions" appears there only in the sentence "the previously described Optional: Developer Additions"
09:58 terran joined #evergreen
09:58 dbs The online docs are built from the README, which is docs/installation/server_installation.adoc
09:58 Dyrcona I know that, too, but as you've discovered errors happen.
09:59 dbs So what I'm trying to do is say a) there is a dangling reference in the README aka docs/installation/server_installation.adoc that either needs to be corrected or removed and b) we need to revisit the doc build process to do something like mv the old docs out of the way if the build is successful
10:00 dbs Saying "always use the README" doesn't get us closer to solving either of those problems
10:02 kmlussier dbs: I think the previously described Optional: Developer Additions are from step 5 of Installing prerequisites.
10:03 kmlussier It makes more sense if you follow the link from the Downloads page since it's all on one page. In the official doscs, you need to click the 'Prev" link to see it.
10:04 Dyrcona dbs: I agree, but all too often I find new people coming in channel trying to install using out of date docs from the web site.
10:05 Dyrcona So, I recommend use the README from the tar ball until the docs are fixed.
10:05 Dyrcona That's why I asked "who's in charge."
10:05 dbs kmlussier: ah i get it, the mismatch in case ("Developer additions" vs. "Developer Additions") was breaking my search
10:05 dbs kmlussier: so we need to add a proper link target to that
10:06 kmlussier dbs: Yes, that would be a useful addition, I think.
10:07 * dbs will give that a shot as a way of procrastinating writing
10:07 kmlussier dbs++
10:08 remingtron dbs: Robert Soulliere is still hosting the live docs server, so he's "in charge" of implementing build changes in a visible way. But we have a docs dev server too, where things can be tested.
10:08 remingtron docs-testing.evergreen-ils.org
10:11 mmorgan1 joined #evergreen
10:12 dbs ah installing 200mb of packages for latex support :)
10:13 Dyrcona Yeah, there is that...Just so that asciidoc can generate PDFs.
10:14 dbs The five minutes are a worthwhile one-time cost :)
10:14 Dyrcona Yes, unless you promised to reboot the z39.50 server at 10:00 am and apt says it's going to take 7 more minutes to download the rest of the packages at 9:59. :)
10:15 Dyrcona And, yes, that happened this morning.
10:15 dbs ouch
10:15 Dyrcona Ctrl-c is your friend. :)
10:15 Dyrcona We had some networking fallout after switching out the load balancers this week.
10:16 Dyrcona Seems odd, but the machines (real and virtual) with two ip address on different networks had to have the gateway changed to the internet-routable network before they would completely funtion.
10:17 Dyrcona Dunno what necessitated that because I used the same ip-tables config.
10:18 Dyrcona But, now I know when I set up the others later this month.
10:18 Dyrcona Or next month.
10:18 mmorgan joined #evergreen
10:20 beanjammin joined #evergreen
10:32 dbs hmm, pdf transform is dying during dblatex step because "*** Error: orderedlist too deeply nested (> 4)"
10:32 dbs happens with stock master for me *sigh*
10:33 csharp @who is too deeply nested?
10:33 pinesol_green Glen is too deeply nested.
10:34 dbs oh i guess we're using fop now?
10:45 dbs successful build (using 'a2x --format pdf --fop docs/root.adoc' for posterity)
10:46 dbs branch is at user/dbs/link_to_optional_developer_additions if anyone wants it
10:46 * dbs out
10:49 dbs well not quite out, jamesrf asked to join https://gitlab.com/evergreen-libra​ry-system/evergreen-library-system which had automatically been migrated over from gitorious years ago, so I added him and set up automatic syncing of the repo from git.evergreen-ils.org
10:49 Christineb joined #evergreen
10:50 beanjammin joined #evergreen
11:07 remingtron anyone know if the password changed for web client on demo.evergreencatalog.com?
11:07 remingtron We've been using it for taking screenshots, but today we can't login.
11:08 Dyrcona The password for which user? There are several in concerto.
11:08 remingtron admin
11:08 Dyrcona Have you tried another user?
11:08 remingtron I haven't tried others, no
11:09 remingtron other concerto logins seem to work
11:10 Dyrcona Then, it's likely the password for admin changed.
11:10 Dyrcona But, I don't know.
11:10 Dyrcona Can you use the other accounts to get your screen shots, or is there something about admin that you need?
11:10 remingtron some concerto's don't work, some do
11:11 remingtron yes, I think we can use one of the logins that has worked. Thanks for the suggestion.
11:11 remingtron Dyrcona++
11:11 Dyrcona Sounds like someone has been messing around. I don't think I have access to that server to reset anything.
11:13 remingtron Look what happens when you give people a demo server!
11:13 remingtron :)
11:16 Dyrcona :)
11:43 yboston joined #evergreen
12:03 csharp @blame whoever broke the demo server
12:03 pinesol_green csharp: Your failure is now complete, whoever broke the demo server.
12:04 beanjammin joined #evergreen
12:06 jihpringle joined #evergreen
12:07 eby speaking of the demo server, would anyone have a second to try duplicating something for me
12:07 eby in webby
12:15 kmlussier sure
12:15 * mmorgan can have a go, too.
12:15 eby if you go to /eg/staff/circ/holds/shelf/XXX with XXX being a hold id that has already been picked up by a patron
12:15 eby i'm curious what the status shows as
12:19 berick BTW, that's select hold in row -> choose Detail View
12:20 mmorgan 9
12:21 mmorgan eby: what do you see?
12:22 eby 9
12:22 * mmorgan has no idea what "9" means.
12:22 eby we've had some lag we are looking into so we've seen checked out holds still show up on the shelf list for a little bit
12:23 eby but wanted to make sure wasn't something else with that 9
12:23 eby thanks for testing
12:23 berick 9 == hold fulfilled
12:23 berick sounds like a bug in the UI
12:23 mmorgan np!
12:24 berick in the sense that it should show a friendly status label, not 9, obviously
12:24 mmorgan Ideally you would not get there for a fulfilled hold, since it would no longer show on your holds shelf.
12:24 berick oh yeah
12:25 mmorgan Unless you had the detail page open already, and realoaded. Or if there's a lag with filled holds coming off the shelf as eby reports.
12:26 eby yeah the hold_strings.tt2 only goes up to 8. I think us seeing it is performance related. I just added a 9 = checked out to our tt for now until we get to the bottom of our performance
12:27 alynn26 joined #evergreen
12:28 eby thanks again
12:36 alynn26 I am working on our Custom Receipts for the web client, and was working on the holds pull list.  Currently it sorts by the hold ID.  I trying to get it to sort by shelving location, I've used ng-repeat="hold_data in holds | orderBy:hold_data.copy.location" but it still only sorting by the ID.  Any suggestions
12:46 eby mmorgan / berick would you prefer i open a bug on the missing translation?
12:47 berick eby: well, like mmorgan said, you should not be seeing captured holds in that interface
12:47 berick unless the hold was already selected, and the page was later refreshed
12:47 berick but I suppose having the translation won't hurt
12:47 berick and could help in these unexpected cases
12:49 berick eby: so, yeah, i guess a bug would be good, but maybe clarify in the description it's an edge case
12:49 * mmorgan would agree it wouldn't be a bad thing.
12:49 eby thanks
12:56 berick alynn26: couple things to consider..  the argument for orderBy needs to be in qotes.  also, if you want to order by location name, you'll probably need ...copy.location.name.
12:57 berick alynn26: so orderBy:'hold_data.copy.location.name' -- but I'm not 100% sure ordeBy supports nested values like that.  not seeing any good examples of that in our code.
12:57 berick alynn26: and finally, if you use Copy Location Order configs, it should sort by that for you automatically
12:58 mmorgan joined #evergreen
12:58 berick that's the default sort, followed by request time (same as id sorting in most cases)
13:03 alynn26 I went back and reread the orderBy information, and found the quote issue.  We do not user Copy Location Order, so it was only using the request time.  But not my receipt will print in Shelving Location order.  Thanks berick.
13:04 hbrennan joined #evergreen
13:04 alynn26 ng-repeat="hold_data in holds | orderBy:'copy.location.name'" worked like a charm.
13:05 berick alynn26: nice!
13:06 alynn26 now if i can figure out how to sort it by location then by call number my staff would be happy.
13:08 kmlussier alynn26: bug 1749502
13:08 pinesol_green Launchpad bug 1749502 in Evergreen "Web Client Holds Pull List Printing" [Undecided,New] https://launchpad.net/bugs/1749502
13:08 kmlussier I think Joan mentions in that bug how to get the call number sorting in there.
13:09 kmlussier Which reminds me of a question Joan asked me about call number sorting in that print template...
13:10 alynn26 I did a bug search but did not see that one. Thanks for pointing it out to me.  I will make my changes now.  Thanks.
13:10 kmlussier In the holds pull list, there is a call number label column, which I believe is a combination of the cn prefix, label, and suffix.
13:10 kmlussier We wanted to sort by that full call number in the print template, but we weren't sure if it could be done.
13:11 * kmlussier hasn't had much time to look into it, but since we were chatting about pull list templates, I'll throw the question out there.
13:18 alynn26 There is a hold_data.call_number.label vs the hold_data.volume.label.  I cannot tell if there is a difference as we do not use the prefixes or suffixes so, they both print the same thing.
13:22 miker fwiw, there should also be a label_sortkey which normalizes the CN value.  however, it does /not/ include affixes
13:28 khuckins joined #evergreen
13:28 JBoyer miker, re: label sortkeys not incluing affixes, is there a specific reason for that? One of the reasons some of our libraries don't use them is that they don't sort as expected when using things like OPAC browse and etc.
13:30 JBoyer There hasn't been enough interest in them to have much of a discussion about it (Is it a bug? add an OUS? etc.) I'm just curious.
13:30 kmlussier We have many people who would love to see them included in the sort order.
13:30 miker JBoyer: well, affixes are "local" additions to what, say, LoC would consider a call number.  for instance, a JUV prefix is important to humans because it effectively removes the need to look at the shelving location of an item (it's in the juvenile section) but it doesn't have anything to do with what the book is about.
13:31 miker IOW, it's not part of, say, a Dewey number
13:32 miker the "answer" is, include the shelving location (under the assumption that that's the purpose of the prefix.  does that assumption generally hold?)
13:33 miker note: that's just the logic behind the decision. I'm not arguing that it's the best UX for librarians that want to shortcut the shelving location and use the prefix instead
13:34 miker (or the best UX for any given workflow other than "be technically correct about CN normalization")
13:34 kmlussier miker: I've been told there are times when multiple prefixes can be used within a shelving location. I'm not sure how often that happens.
13:35 JBoyer That's a difficult assumption with fiction. Are the SCI FIC items shelved with regular FIC? how about MYSTERY, WESTERN, etc.
13:35 miker it also makes normalization monumentally simpler, because the main CN label can reasonably be expected to follow the rules of the assigned label_class
13:35 JBoyer We have both kinds of libraries in our group (it's all fic, or each genre has an acpl) I'm sure BISAC has other warts that I'm not aware of.
13:36 kmlussier Moving off topic. When is the beta deadline again?
13:36 dbwells Next Wednesday, 2/28
13:36 miker heh... I used to advocate for bisac. and then I went to a book store last week and spent 45mins hunting for something :)
13:36 kmlussier dbwells: Thanks!
13:37 * kmlussier is taking the day off tomorrow and was wondering how much she needed to stress about getting things in before beta.
14:00 rlefaive joined #evergreen
14:49 mmorgan Regarding the call number prefix/shelving location discussion - one example where prefix is important is a shelving location "Adult Video", and prefixes "DVD", "Blu-Ray"
14:52 * miker wishes that would just come out of the MARC
14:53 collum_ joined #evergreen
14:55 jeff or two shelving locations.
14:56 jeff but I suspect the number of call numbers with Blu-Ray or DVD in the label which don't 1:1 match with what's in the MARC is also a bit of a problem there.
14:58 kmlussier miker++
14:58 kmlussier No for the current discussion, but for ongoing bug fixing for our search highlighting.
15:00 mmorgan Also, combopacks have one of each format.
15:00 jeff sure, but those have a challenge in the call number label/prefix are also :-)
15:01 jeff s/are/area/
15:02 mmorgan1 joined #evergreen
15:17 kmlussier @marc 242
15:17 pinesol_green kmlussier: A translation of the title proper that is made by the cataloger when the translated title does not appear as a parallel title on the item. For a note, the introductory phrase Title translated: may be generated based on the field tag for display. (Repeatable) [a,b,c,h,n,p,y,6,8]
16:49 mmorgan joined #evergreen
17:06 mmorgan left #evergreen
17:37 miker thanks for being down, launchpad...
18:30 jwoodard joined #evergreen
18:31 pinesol_green News from qatests: Failed Running perl live tests <http://testing.evergreen-ils.org/~live>

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