Evergreen ILS Website

IRC log for #evergreen, 2014-02-05

| 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
00:58 shadowspar joined #evergreen
01:14 shadowspar joined #evergreen
06:25 edoceo joined #evergreen
06:36 edoceo joined #evergreen
07:08 timf joined #evergreen
08:02 kmlussier We're due for a developers meeting. Doodle poll at http://doodle.com/shbpr649h6xint7r
08:02 * kmlussier most likely won't be able to attend the next meeting.
08:16 Shae joined #evergreen
08:26 akilsdonk joined #evergreen
08:31 kmlussier @wunder 02771
08:31 pinesol_green kmlussier: The current temperature in Ladd Observatory, Providence, Rhode Island is 27.1°F (8:30 AM EST on February 05, 2014). Conditions: Light Snow. Humidity: 95%. Dew Point: 26.6°F. Windchill: 21.2°F. Pressure: 30.07 in 1018 hPa (Falling).  Winter Storm Warning in effect until 6 PM EST this evening...
08:32 kmlussier A very snowy winter indeed.
08:56 rfrasur joined #evergreen
08:57 rfrasur joined #evergreen
08:57 collum joined #evergreen
08:57 rfrasur jboyer-isl: will SIP2 authentication still be available during the upgrade or will the whole thing be down?
08:59 rfrasur I see that OD will still be available.  Does that go for other stuff as well?
09:25 Dyrcona joined #evergreen
09:30 rjackson_home joined #evergreen
09:38 ericar joined #evergreen
09:49 Dyrcona Aren't snow days wonderful?
09:51 pastebot "Dyrcona" at 64.57.241.14 pasted "Better way to do this or cache it in tpac?" (8 lines) at http://paste.evergreen-ils.org/12
09:52 Dyrcona So, I wrote that code above, and that seems fine for a script that will only be run occasionally, but I want to put this into TPAC eventually.
09:53 Dyrcona The question is: Is there a better way to do that already in TPAC or is there a good way to cache the item_type hash?
09:53 dbs Dyrcona: hey, I think we could use the auditor tables to revert changes for small batches of bibliographic objects!
09:54 Dyrcona dbs: yeah, you could, but we clear ours every 6 months, so if there will be more than 6 months between changes, then it won't work for us. That's why we made the custom table.
09:54 dbs Dyrcona: I'm making a joke about me mentioning audit tables and then you saying the same thing in that thread :)
09:55 Dyrcona Ah ha! I'm making a joke of myself, apparently. :)
09:55 dbs If we ever do base functionality on audit tables I guess we should document that it might be subject to a given site's purging practices.
09:55 * dbs doesn't think we have any purging practices... probably should look into that some day
09:55 berick Dyrcona: ctx.search_ccvm('name', 'item_type')
09:56 berick will cache
09:56 Dyrcona berick: Thank you muchly! I'll give that a whirl when I get to TPACifying the code.
09:56 jeff tpacification
09:57 Dyrcona I figured there had to be something that I overlooked, since TPAC needs that info in a number of places.
09:58 jeff I'm starting a fake library content vendor. We're going to have no content, but exemplary patron authentication and user experience. :P
09:58 dbs MuchDogeVendor?
09:58 dbs GoodVendorDoge
09:59 jeff wOw! such auth! very UX!
09:59 jeff Welcome to your library's doge-ital branch!
10:05 Dyrcona jeff: I see what you did there.
10:05 Dyrcona well, dbs, really.
10:05 Dyrcona jeff: So, it's a mining operation, is it?
10:07 Dyrcona Hmm... Maybe EG needs a cryptocurrency payment module?
10:08 Dyrcona Set fine rates in Satoshis. :)
10:17 jeff heh
10:17 jeff I'm assuming that I mangled the doge-speak above.
10:25 jeff But no, the idea isn't to mine bitcoins or accept dogecoin. :-)
10:25 Dyrcona No, it sounds almost like the dogecoin website exactly: much coin! such community! much news! very fashion!
10:26 Dyrcona so, very UX is spot on. :)
10:26 jeff The idea is to have an example of "this would be a great way to auth patrons" and "this would be a great way to treat patrons", rather than [current status quo of library content vendors here].
10:26 * Dyrcona wishes library vendors actually had a clue.
10:26 jeff I get worked up about this when evaluating new services. Sometimes the interest fades, sometimes not.
10:27 jeff But part of the idea too is that it would give me some perspective on Evergreen/Koha/etc from a "content provider" point of view also.
10:28 dbs jeff: this would let us test improved core security too, like replacing unsalted MD5-hashed passwords
10:29 jeff "here's an idea, let's not use a self checkout protocol with access to holds/circ/bills/more just to verify that you are a patron, and let's not require your library card number (which one?) and library password be entered on the vendor site, and let's not have the vendor store your library password in cleartext in their systems!"
10:29 jeff dbs: yes, though i see that as a parallel, higher priority goal. :-)
10:29 dbs jeff: so teaching Evergreen to be an OAuth2 provider?
10:29 jeff yep.
10:30 jeff and/or other options.
10:30 jeff but that seems to be the most promising in terms of "go implement this" -- a web standard vs a library standard.
10:30 dbs jeff: jrochkind says its as simple as switching md5 to bcrypt. EASY!
10:31 dbs jeff++
10:32 dbs web_standards++
10:35 Dyrcona dbs: Now, you go and tell all the library vendors that its easy.
10:35 Dyrcona You know the same vendors that can't follow the existing library standards.
10:36 * Dyrcona is jaded after 10 years of dealing with this nonsense.
10:37 dbs Dyrcona: heh, and I'm being more than a little tongue in cheek when I say "EASY!" because of course the easiest part is for jrochkind to say it's easy :)
10:37 Dyrcona Heh.
10:38 Dyrcona That's the great thing about standards! There are so many ways to get it subtly wrong!
10:38 dbs Wow that's a lot of bugfixes getting pushed out :/
10:38 dbs dbwells++
10:52 Dyrcona Zawinski's Law is coming to Evergreen.
10:52 Dyrcona Although, really, it's already here.
10:54 Dyrcona For reference: http://www.catb.org/jargon​/html/Z/Zawinskis-Law.html
10:55 jeff Dyrcona: someone's trying to make Evergreen read mail? :-)
10:55 Dyrcona jeff: Yep.
10:55 jeff Dyrcona: link/pointer?
10:56 Dyrcona http://masslnc.cwmars.org/node/2927
10:56 Dyrcona Really, though, he was talking about the tendency of applications to turn into frameworks, which Evergreen already is.
10:57 * Dyrcona has read the blog post the quote comes from.
10:57 * jeff nods
11:00 jboyer-laptaupe joined #evergreen
11:04 finnx_ joined #evergreen
11:06 rfrasur Dyrcona: You wrestling with Ebsco?
11:07 Dyrcona I was, why?
11:07 rfrasur Just reading back comments.
11:07 Dyrcona Ok.
11:12 ningalls joined #evergreen
11:29 Dyrcona bshum: You getting snow today?
11:30 bshum Dyrcona: Just a lil' bit.
11:30 bshum :)
11:30 Dyrcona We haven't gotten tons, but it was coming down really hard right at the height of the morning commute.
11:31 Dyrcona I'm going to push a commit to the "real" Marque.pm branch in a moment.
11:32 Dyrcona It'll be a force push. I rebased on master since the branch had been sitting for a couple of months.
11:32 Dyrcona I'm going to try to add the missing mfhd and progress features real soon.
11:34 bshum I think we've gotten about 8 inches or so here at my house.  I've been told towards our offices it's like 12-14 inches right now.
11:34 Dyrcona I'm not sure how much we've gotten. I haven't gone out side.
11:35 Dyrcona Half of our members have emailed in that they are closed.
11:35 bshum And yay, Marque.pm whee!
11:35 Dyrcona I'm target it for the beta now. I'll see if I can't finish it by the end of the week.
11:36 Dyrcona Everything is going into working/dyrcona/lp1223903-MARC-export-Mk.2 from now on.
11:37 jboyer-laptaupe bshum: Re: commit db7b99e8ccba2ebf81a345157de5a90ba5387626, is there anything else that may need changing for rel_2_5? We have a test system built from rel_2_5 pulled just after that commit went in, but auto-print still doesn't seem to stick on the checkout page with a fresh staff client.
11:37 pinesol_green [evergreen|Ben Shum] Revert "Eliminate an annoying and useless warning in the JavaScript Console." - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=db7b99e>
11:38 bshum jboyer-laptaupe: It's a server side change.
11:38 bshum Not just a client one
11:38 jboyer-laptaupe Ah, the path led me astray. I haven't checked the server side file yet.
11:39 bshum /openils/var/web/xul/server/OpenILS/global_util.js
11:39 bshum Yeah for the staff client I usually tend to check first if there's a server equivalent.
11:40 bshum The one that gets me is the chrome directory stuff I think
11:44 Dyrcona Yeah, just changing the file on the server fixed it for me.
11:45 Dyrcona bshum: We've got six inches of snow at my house and its still coming down.
11:45 jboyer-laptaupe Well never mind that. Our test system isn't running the version of evergreen we wanted to test. :(
11:45 jeff jboyer-laptaupe: oops.
11:47 jboyer-laptaupe jeff: I'd actually feel better about it if it were my fault. :/
11:49 akilsdonk joined #evergreen
11:49 kmlussier Dyrcona: Did you say your office had a snow day today? Do you all do work-at-home snow days or is it a day off snow day?
11:50 * kmlussier is just wondering about the likelihood of getting a response on an e-mail I sent.
11:52 Dyrcona kmlussier: Officially, I believe it is a work-at-home snow day. At least, I'm doing it that way.
11:52 kmlussier OK, thanks!
11:53 Dyrcona Password reset action trigger runner went bonkers again.
11:56 Dyrcona Heh. Been that way since 8:00 am. I guess Thunderbird wasn't checking mail as often as I thought.
11:56 bshum Dyrcona: Did you see any of what we sussed out at the end of the other day about action_trigger_runner and the parallel options for trigger?
11:57 Dyrcona bshum: Not likely. What day was it and I can check the logs.
11:57 bshum It's been a few days now since I commented out parallel in production and I haven't seen any weird collected states that are broken.
11:57 Dyrcona Is this for all action trigger runners?
11:57 bshum http://irc.evergreen-ils.org/​evergreen/2014-02-03#i_66180
11:58 bshum For me, it mostly affected my hold ones it seemed.
11:58 bshum Though I do know that before I made my overdues into granular events (by type), it broke my overdues too.
11:59 bshum My gut feeling (without much extra evidence) is something weird happening since we upgraded back in December last year to the newer OpenSRF control stuff
11:59 Dyrcona We've been seeing this long before last December.
11:59 Dyrcona We just kill the hung process, remove the lock file, and let the others take it from there.
11:59 bshum Well then I got nothing new ;)
12:00 Dyrcona I'll try removing the parallel option to see if that helps.
12:09 bshum Well, periodically at varying points in our history, it seems our action_trigger_runner script has died out too.  I can see that from clumps of collected A/T from years past.
12:09 bshum But I always assumed that it was because we weren't breaking things up more granularly.
12:10 bshum @hate A/T
12:10 pinesol_green bshum: The operation succeeded.  bshum hates A/T.
12:11 jwoodard joined #evergreen
12:17 bshum Anyone have an opinion on https://bugs.launchpad.net/evergreen/+bug/1125471 ?
12:17 pinesol_green Launchpad bug 1125471 in Evergreen 2.4 "Use label instead of id in recording in-house use prompt" (affected: 2, heat: 10) [Medium,Confirmed]
12:17 bshum Seems like it'd be a straightforward change from ID to type.
12:18 bshum phasefx: Any reason we used ID over name?
12:19 * phasefx looks
12:22 phasefx bshum: ooh, I don't remember any of that stuff; I'll take a look again after lunch :)  I think that was just oversight.  No reason not to show the name/label of the noncat type there instead of a barcode
12:23 bshum phasefx: No worries, it's pretty old, but I wasn't sure if there was some sort of reason.  Thanks.
12:39 jwoodard joined #evergreen
12:42 snowball joined #evergreen
12:42 sseng joined #evergreen
12:49 ktomita joined #evergreen
12:49 sseng joined #evergreen
13:05 jihpringle joined #evergreen
13:13 jboyer_laptaupe joined #evergreen
13:19 ericar joined #evergreen
13:26 fparks I am working on a new rendition of Boolean Search https://bugs.launchpad.net/evergreen/+bug/1152863
13:26 pinesol_green Launchpad bug 1152863 in Evergreen "Support for traditional Boolean operators" (affected: 3, heat: 18) [Wishlist,Triaged]
13:26 fparks Currently I'm stuck on how to store translations for keywords.
13:28 fparks Is anyone familiar with what best practices would be in this case?
13:30 smyers_ joined #evergreen
13:42 hbrennan joined #evergreen
13:46 ktomita joined #evergreen
13:50 jboyer_laptaupe joined #evergreen
13:50 akilsdonk_ joined #evergreen
13:50 sseng_ joined #evergreen
13:51 ktomita__ joined #evergreen
13:58 smyers__ joined #evergreen
14:29 sseng joined #evergreen
14:37 smyers_ joined #evergreen
14:46 smyers__ joined #evergreen
14:51 pmurray joined #evergreen
14:57 roses joined #evergreen
14:57 roses I'm upgrading my demo server from 2.3.7 to 2.5.2 and in the upgrade instructions it has these lines: cd /home/opensrf/Evergreen-ILS​-2.5.2/Open-ILS/src/sql/Pg psql -U evergreen -h localhost -f version-upgrade/2.4.3-2.5.0-upgrade-db.sql evergreen psql -U evergreen -h localhost -f version-upgrade/2.5.0-2.5.1-upgrade-db.sql evergreen
14:58 roses How do I do this from 2.3.7 to 2.5.2
15:05 sseng joined #evergreen
15:11 dbwells roses: The instructions are assuming you are coming from 2.4.  You cannot upgrade directly from 2.3 to 2.5, you must run the upgrade scripts in between as well.
15:12 roses dbwells: I see that there are 10 scripts that I need to run.
15:12 dbwells roses: In other words, you should find a script that goes from 2.3.x to 2.4.0, then 2.4.0 -> 2.4.1, and so on until you get to 2.4.3, then follow the instructions.
15:12 stevenyvr joined #evergreen
15:13 dbwells roses: depending on when 2.4 was released, there might not be a direct route from 2.3.7 to 2.4.0, so you may need to massage the early upgrade scripts until you are back on track.
15:14 roses dbwells:  I'm running the first script (.sql) and I'm having trouble with the password for evergreen - shouldn't it be evergreen because of #10 in the instructions?  perl Open-ILS/src/support-scripts/eg_db_config --update-config --service all \ --create-offline --database evergreen --host localhost --user evergreen --password evergreen
15:15 dbwells roses: There is also at least one supplemental script for 2.4, so make sure you run that too, or at least understand what it does (sometimes scripts do something which gets done again later anyway, especially reingesting records).
15:15 dbwells roses: Is this a test server?
15:15 roses dbwells:  Yes - a test server
15:15 roses dbwells: Which I might just wipe and start with a fresh installation of 2.5
15:16 dbwells The password will be whatever you set it to when you set it up, your evergreen DB password.
15:17 dbwells That is, if you ever login to the evergreen DB using 'psql', it will be the same username/password you use for that.
15:19 dbwells It's might be nice to run through the upgrade scripts for general practice, understanding, and to test the scripts themselves, but unless you have a lot of data in your test server, installing fresh is probably easier.
15:20 roses dbwells:  Thanks.  I found my password and now am going to run through the scripts.  I might be back later.
15:24 fparks I am working on a new rendition of Boolean Search https://bugs.launchpad.net/evergreen/+bug/1152863
15:24 pinesol_green Launchpad bug 1152863 in Evergreen "Support for traditional Boolean operators" (affected: 3, heat: 18) [Wishlist,Triaged]
15:25 fparks Currently I'm stuck on how to store translations for keywords.
15:25 fparks Is anyone familiar with what best practices would be in this case?
15:30 Dyrcona joined #evergreen
15:30 Dyrcona For some reason, my laptop just shut down.
15:31 Dyrcona Fortunately, I only lost four lines of copy and paste, so I could past it again.
15:33 Dyrcona bshum: Marque.pm exports MFHDs now!
15:33 Dyrcona Unless someone really wants progress output, I think the branch is done.
15:34 Dyrcona Well, one step remains and that is making marc_export a link to Marque.pm.
15:44 Dyrcona 'course we only 56 SREs in our database, so more eyes might be good on that part. :)
15:44 Dyrcona ...only [have] 56....
15:47 roses joined #evergreen
16:10 phasefx bshum: I tried looking at the commit log for that line re: non-cat in-house, and I think it was indeed oversight + intermediate testing
16:12 phasefx for those curious (and I only just now learned how to do this), I did  git blame in_house_use.js, got a useless commit message for some whitespace cleanup, but then did this to get the commit prior to that:   git blame in_house_use.js b228af4f~1
16:12 pinesol_green [evergreen|phasefx] Change each tab to 4 spaces in the staff client javascript files. - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=b228af4>
16:35 Dyrcona learning++
16:35 Dyrcona I learned a new git trick today.
16:35 Dyrcona I learned how to split up an old commit into different ones.
16:36 dbs reset / add -p ?
16:42 sseng joined #evergreen
16:58 dcook joined #evergreen
17:50 kmlussier Dyrcona: For https://bugs.launchpad.net/evergreen/+bug/980296, can I change the status from "In Progress" and remove you from "Assigned To"? You're done with that for now, right?
17:50 pinesol_green Launchpad bug 980296 in Evergreen "Marking a lost item Claimed Returned does not void lost charges " (affected: 3, heat: 18) [Wishlist,In progress] - Assigned to Jason Stephenson (jstephenson)
18:18 Dyrcona kmlussier: I just took care of it.
18:18 kmlussier Dyrcona++
18:18 Dyrcona I was away from the keyboard having something to eat, then shoveling some snow.
18:18 kmlussier If I don't get the release notes done tongiht, I'll do it in the morning.
18:19 kmlussier Eating is good. Shoveling, not so much.
18:19 bshum kmlussier: Since you assigned this... https://bugs.launchpad.net/evergreen/+bug/1155769
18:19 pinesol_green Launchpad bug 1155769 in Evergreen "Items repeated in tpac display, others not visible" (affected: 3, heat: 16) [Undecided,Confirmed]
18:19 bshum Do you know how that bug is replicated?  I've never been able to consistently get broken displays as described.
18:19 Dyrcona dbs: git rebase -i / then edit the commit, reset HEAD^, add some files, commit, add the remaining files, commit, rebase --continue
18:20 kmlussier I've seen it happen. You've got to find the right record to make it happen.
18:20 kmlussier I was just about to test it and was first looking for a record.
18:20 bshum And I'm not sure the change ktomita makes there to order by barcode instead of copy number would necessarily be good.  Just for the folks who do use copy number in their cataloging.
18:20 bshum Not us, for sure, but someone might?
18:21 kmlussier Sounds like a good comment to make on the bug report. :)
18:21 bshum Oh yes, I'll add it.
18:21 kmlussier I was looking at that too, but then I wasn't sure what was copy number referred to there.
18:21 bshum I'm just not sure how to connect the dots since I've never been able to get a record to misbehave to verify the problem.
18:22 kmlussier I'll be happy to look on acorn to see if I can find one for you.
18:22 bshum I assumed it was referring to asset.copy.copy_number
18:23 bshum Which is normally empty if catalogers don't put a number in there during their work.
18:23 kmlussier So maybe doing copy number then barcode might work. Instead of just eliminating copy number.
18:25 bshum Hmm
18:25 bshum I'm not sure what it'd do if there was a mix of NULL and actual numbers in that case.
18:25 bshum Would it put the copy 1 ahead of the NULLs or after the NULLs
18:26 bshum Either way I'll add my question to the ticket
18:26 bshum I think after talking it out I understand the intent now.
18:28 kmlussier Well, since it was already doing the copy number sorting, I expect it would continue handling nulls the same way as it does now, wouldn't it?
18:31 bshum Probably right.
18:31 kmlussier bshum: Got one for you. http://acorn.biblio.org/eg/opac/record/2881599
18:31 kmlussier The last barcode for Brookfield is duplicated on the next page when you click 10.
18:31 kmlussier You'll probably find that there's one barcode missing from the display.
18:32 kmlussier Or I should say the first barcode for Brookfield.
18:32 bshum Gotcha, I see it there
18:32 bshum Aha
18:32 kmlussier Gotta love the Fifty Shades of Grey series for providing ready samples of records with lots of copies.
18:43 bshum Okay, adding the fourth order by of barcode seems to resolve the issue on that record.
18:43 bshum And you were right there was something hidden quietly there
18:44 kmlussier I think it really became a problem because, sometimes, the hidden one might be the only copy that is checked in at the time.
18:45 kmlussier So people aren't seeing that there's something on the shelf.
18:45 bshum No it makes sense now.  I just couldn't get enough symptoms to show, but how would you know something wasn't there unless you worked for a library affected.
18:46 bshum And the solution worked out by ktomita makes sense now.  Just might want to retain copy_number sorting for those who use it.
18:47 kmlussier Makes sense to me. Thanks for looking at it Ben!
18:47 kmlussier bshum++
20:03 afterl left #evergreen
21:17 ktomita joined #evergreen
22:20 stevenyvr joined #evergreen
22:40 pinesol_green [evergreen|Bill Erickson] LP#1164720 Prevent empty names in TPAC lists - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=237b7e7>
22:44 pinesol_green [evergreen|Ben Shum] LP1271218 - clarify how opt-in user notifications are shown in my account - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=8ca1ea0>
23:20 dcook joined #evergreen
23:21 zerick joined #evergreen

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