Evergreen ILS Website

IRC log for #evergreen, 2020-10-05

| 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
06:01 pinesol News from qatests: Testing Success <http://testing.evergreen-ils.org/~live>
06:12 calvinm joined #evergreen
07:21 rjackson_isl_hom joined #evergreen
08:06 alynn26 joined #evergreen
08:25 Dyrcona joined #evergreen
08:40 mmorgan joined #evergreen
09:17 rfrasur joined #evergreen
09:22 dbwells joined #evergreen
09:56 dbwells joined #evergreen
10:19 dbwells joined #evergreen
10:21 alynn26_away joined #evergreen
10:38 Bmagic I'm fooling with the docs site this morning, yall, don't be alarmed if it acts weird. Troubleshooting the lunr search box
10:41 * csharp disables site-fooling alarms accordingly
10:42 csharp so I may have dreamed this, but I remember seeing a bug report where JBoyer (?) had created views for Dewey, etc. to make Classic Item List and Classic Circulation View obsolete...
10:42 csharp trying to find the bug in LP but I'm either overlooking it or it's not findable with my search terms/tags
10:45 mmorgan csharp: bug 1813191 ?
10:45 pinesol Launchpad bug 1813191 in Evergreen "Move calculated dewey ranges/blocks to their own reporting view" [Wishlist,New] https://launchpad.net/bugs/1813191
10:54 csharp mmorgan: yes!  thank you
10:55 mmorgan YW!
10:58 JBoyer mmorgan++
10:59 JBoyer csharp, I'd be delighted if you could throw that on a test server with PINES data to see what's what with a real dataset. Obviously works pretty well with concerto. ;)
10:59 csharp JBoyer: will do
11:00 JBoyer (If that's what you're thinking)
11:00 JBoyer csharp++
11:00 csharp we're having fits with Classic Circulation View, which would also be rendered unnecessary with that improvement
11:00 JBoyer Nice.
11:02 JBoyer My aim was to make it easier / possible to do a git rm :q
11:02 csharp yes
11:02 JBoyer Not a command line, bad typist.
11:12 Bmagic JBoyer: I hate to say it, but it seems that removing the ui supplement stuff from Evergreen is bugging the search. Though, I can't figure out why. I agree, that you've migrated the overrides over to eg-antora. Still diffing
11:13 Bmagic right now, I can say for sure that building antora with the ui-bits in Evergreen overlayed over eg-antora makes it work
11:33 Bmagic JBoyer: we may need to put the supplement stuff back. The issue was our nightly Evergreen builds were bombing on the absence of header-content.hba? Can we just make that file exist (with the updates from eg-antora) and call it a day?
11:34 Bmagic I'm starting to think that the lunr stuff needs* itself to be in the repo with the docs during build time. Alternatively, the build script could cp eg-antora -> Evergreen/docs/ui, commit a test branch, build that branch, then revert it all
11:47 gmcharlt the former, if there's no other way, is preferable to the latter: creating and throwing away git commits wouldn't be great for anybody with uncommitted changes in the clone who simply wants to try a doc build
11:57 jihpringle joined #evergreen
12:03 Christineb joined #evergreen
12:06 Bmagic gmcharlt: I agree for sure, though the commits would only exist locally and thrown away locally. They wouldn't ever need pushed during the build process
12:08 Bmagic But programming generate_docs.pl to do that, would be pretty antora server specific, which makes it less adoptable by the community. Though, I could see a way for it to allow for that sort of behavior with switches or something along those lines. Not my favorite idea
12:13 Bmagic One thing that bufuttled me when I was first getting familiar with Antora is: all of your changes need to be commited to branch. Antora will not build/include any files/changes that are not commited to the main build branch. Including antora.yml
12:17 JBoyer Bmagic, yeah, if that file is missing we should be able to just put it in place with the edits it needs. It's possible I accidentally left it out when moving things over.
12:18 Bmagic JBoyer: header-contents.hba exists in the eg-antora repo. No problem. The issue, it seems, is that lunr doesn't like zero files being in the Evergreen branch
12:19 Bmagic I haven't tried building it with an empty folder structure and a call out to it in the yaml file though.... I wonder
12:19 JBoyer I thought the issue was something with the CSS? I was under the impression it was fine aside from that.
12:22 Bmagic the resulting site has a broken search box right now
12:22 Bmagic I've been rebuilding it with different ideas, and it's currently working with the ui-bits back in the Evergreen repo
12:22 eady joined #evergreen
12:23 Bmagic so, right this second, it's working fine
12:53 collum joined #evergreen
12:55 sandbergja joined #evergreen
13:34 JBoyer Bmagic, it looks like the issue with the built-in lunr changes is that the path for search.css is wrong. I'm getting 404s for it and seeing the weird search problem you're talking about.
13:34 Bmagic ah! There we go
13:34 Bmagic nice catch
13:35 JBoyer so it's not a matter of using the supplemental thing again, but making that happy.
13:35 Bmagic sweet, we update search.css in eg-antora then? It sounds like you already know which lines?
13:36 JBoyer search.css itself is fine, I don't think it's being included in the ui-bundle.
13:38 Bmagic I see, search.css (part of lunr's overrides) is not compiled into ui-bundle with gulp pack
13:38 * Dyrcona shakes his head....
13:38 JBoyer yeah, because it wasn't a changed file in the original antora-ui, it's as if we just dropped a new file into a project and wondered why make didn't pick it up.
13:39 Bmagic You're thinking of making gulp do that? or perhaps putting search.css in the Evergreen repo?
13:39 JBoyer gulp just needs to be told that it's part of the build target.
13:39 JBoyer wherever that's done...
13:39 * Bmagic nods at Dyrcona. The agony
13:50 Bmagic JBoyer: it sure seems like it should* be including search.css: css: `${srcDir}/css/**/*.css`,
13:51 Bmagic oh right, it compiles* them into site.css
13:51 Bmagic minifies them
13:52 Bmagic I'll bet search.css won't lint
13:52 JBoyer :q
13:52 JBoyer bah
13:53 Bmagic lol
13:54 JBoyer the only thing that shouldn't lint is lunr.js. It's such a mess according to their rules I just disabled linting that file altogether.
13:54 Bmagic I saw that
13:54 JBoyer search.css only needed a couple minor edits.
13:55 Bmagic grepping the resulting site.css, I'm not seeing any of the definitions from search.css included. Though, it seems it should be with the ** syntax from gulpfile.js
13:56 Bmagic in fact, the word "search" doesn't apear in ui-bundle.zip:css/site.css
13:56 Bmagic appear
13:56 JBoyer something needs to be changed in gulp.d/build.js
13:57 JBoyer OR, we just throw the contents of search.js at the bottom of site.js and call it good.
13:57 JBoyer I don't know enough about the layout and build process to know which would be the better way to deal with it
13:58 * Bmagic reads the files referenced
13:59 Bmagic sheesh, only 114 lines in search.css
14:00 JBoyer well, my latter suggestion is no good, when I said that I assumed there was only a single site.css file in there; it can obviously handle multiples somehow, it just doesn't want to add search.
14:01 Bmagic adding the stuff back to Evergreen would be cleaner anyway IMO
14:01 Dyrcona @decide lunr.js or luny.js
14:01 pinesol Dyrcona: go with lunr.js
14:02 Dyrcona @decide bottle in front of me or frontal lobotomy
14:02 pinesol Dyrcona: go with bottle in front of me
14:02 Bmagic lol
14:04 JBoyer If we have a custom eg-antora what would be "clean" is keeping all of the custom stuff in one place. Otherwise we have some custom antora stuff in eg-antora, more custom antora stuff in Evergreen/docs/ui-lunr, etc.
14:16 nfBurton joined #evergreen
14:16 JBoyer paying just a little more attention to the grep I ran earlier shows that all that *should* need to be done is drop another @import into site.css.
14:16 * JBoyer feels silly now
14:16 JBoyer provided it works, anyway
14:17 JBoyer And yes, that's all it needs. (and removing the line about search.css from head-meta.hbs) I can push a fix to eg-antora.
14:18 JBoyer done
14:32 Dyrcona :)
14:33 * Dyrcona deals with a mess caused by squashed commits and wishes he still had the original commit chain somewhere.....
14:34 Dyrcona Maybe ediff or emerge modes in Emacs can repair this?
14:38 Dyrcona Oh, joy of joys..... This is NOT gonna be fun.....
14:39 Dyrcona I'm gonna have to split differences, etc. Deleting blank lines in the file was a bad idea...... Maybe I have another way to deal with this?
14:41 Dyrcona Hm.... If I compare the file to the previous commit rather than the tip of master, that *might* make things easier.
14:44 Dyrcona It may also be time to stop using rebase and go with merge on the CW MARS custom branch....
14:45 Dyrcona Because of the rebase, I can't really find the "previous" commit.....
14:47 Dyrcona git log -- Open-ILS/examples/opensrf.xml.example # Does the trick? :)
14:48 Dyrcona And the merge/diff is just as hairy as going with origin/master....
14:54 Dyrcona Merge gets the same conflict markers.........
15:26 jihpringle joined #evergreen
15:28 jvwoolf joined #evergreen
15:38 Dyrcona So, I think I'll tackle that again tomorrow morning in the opposite direction. I'm going to start with master and move our changes over, and I'll update the opensrf.xml.example manually after the fact.
15:46 collum joined #evergreen
15:46 Dyrcona Hmmm... I just got a "PIN recovery" email from my library's ILS, except I didn't ask for one, and the link is not to my home library, but another one.
15:48 * Dyrcona emails tsbere to see if they're doing anything system-wide.
16:27 troy___ joined #evergreen
16:27 jeffdavis_ joined #evergreen
16:27 troy__ joined #evergreen
16:27 jeffdavis joined #evergreen
16:27 jeffdavis joined #evergreen
16:27 jeffdavis joined #evergreen
16:40 khuckins joined #evergreen
16:53 sandbergja joined #evergreen
17:05 mmorgan left #evergreen
17:29 Dyrcona joined #evergreen
17:54 pinesol [evergreen|Lisa Carlucci] docs: corrections to the 3.6 release notes - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=b2c3f81>
17:56 pinesol [evergreen|Galen Charlton] docs: release notes: adding EOLI to the list of commissioning entities - <http://git.evergreen-ils.org/?p=​Evergreen.git;a=commit;h=f900fbd>
18:01 pinesol News from qatests: Testing Success <http://testing.evergreen-ils.org/~live>
18:44 sandbergja joined #evergreen
18:46 sandbergja For anybody still around: can I make a URL that opens the patron self-registration form with a specific library selected?
18:46 sandbergja (apart from setting up another OPAC with that library as physical_loc)?
21:03 jtm-lis I am looking for a marc21 catalog, that I can query publicly.
21:31 mrisher joined #evergreen
23:12 sandbergja joined #evergreen

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