| Time |
Nick |
Message |
| 00:31 |
|
sandbergja joined #evergreen |
| 01:22 |
|
sandbergja joined #evergreen |
| 06:55 |
|
agoben joined #evergreen |
| 07:05 |
|
rjackson_isl joined #evergreen |
| 07:13 |
|
rfrasur joined #evergreen |
| 08:36 |
|
mantis1 joined #evergreen |
| 08:44 |
|
Dyrcona joined #evergreen |
| 08:45 |
|
mmorgan joined #evergreen |
| 09:21 |
|
jvwoolf joined #evergreen |
| 09:29 |
|
yboston joined #evergreen |
| 09:37 |
|
sandbergja joined #evergreen |
| 09:51 |
Dyrcona |
Sometimes you have to do something before you realize it is a bad idea. |
| 09:53 |
Dyrcona |
JBoyer | sandbergja: If you're curious about what I've done on Lp 1850156, then https://git.evergreen-ils.org/?p=working/Evergreen.git;a=shortlog;h=refs/heads/user/dyrcona/lp1850156-git-makefile.install-target |
| 09:53 |
pinesol |
Launchpad bug 1850156 in Evergreen "Need a -tester make target" [Wishlist,Confirmed] https://launchpad.net/bugs/1850156 |
| 09:59 |
Dyrcona |
I should probably not have signed off on that branch.... |
| 09:59 |
Dyrcona |
Oh well... Meeting! |
| 10:34 |
|
sandbergja joined #evergreen |
| 11:33 |
Dyrcona |
Hm.. So looks like there is a problem with installing pgtap from Makefile.install. You have to know which version of PostgreSQL is installed. |
| 11:34 |
Dyrcona |
If you just install the plain old pgtap package on Ubuntu 16, it also tries to install Pg 9.6 if Pg 10 is already installed. |
| 11:48 |
Dyrcona |
And, then, you're back on a variation of the way you thought that things wouldn't work. |
| 11:49 |
|
jihpringle joined #evergreen |
| 11:50 |
Dyrcona |
Here's a question: Do we want translators to have to install the browsers for testing? I don't think so, but... |
| 11:51 |
Dyrcona |
Packagers, definitely. |
| 12:11 |
|
rjackson_isl_ joined #evergreen |
| 12:21 |
|
sandbergja joined #evergreen |
| 12:42 |
* Dyrcona |
thinks the webstaffclient tag is mostly useless, now. |
| 13:35 |
Dyrcona |
@monologue |
| 13:35 |
pinesol |
Dyrcona: Your current monologue is at least 11 lines long. |
| 13:52 |
|
khuckins joined #evergreen |
| 13:54 |
jeffdavis |
It took me 12 years of using psql, but I have finally learned how to visually distinguish null values from empty strings (\pset null). |
| 13:58 |
Dyrcona |
:) |
| 14:03 |
|
mantis1 left #evergreen |
| 14:31 |
|
LBMV joined #evergreen |
| 14:35 |
LBMV |
Hey everyone, I'm working on a little project to make the file type of format icons in search results configurable. However, I'm not sure where the best location to store this information is. So far, our thoughts are: org unit setting, opensrf.xml, or in a coded value map. What ideas do you all have? |
| 14:37 |
Bmagic |
LBMV: like you want to have the system use any arbitrary filename? Or you want to store the image in the db? |
| 14:40 |
LBMV |
Right now we are implementing icons in SVG, with the same file names as the original PNG versions. But it could also be cool to allow various different file types for the icons at once |
| 14:43 |
Bmagic |
I'd like to see config.coded_value_maps receive a new column. Maybe called "filename" or some such. So that each of the icons can be linked to any filename. Changing database structure would need weighed-in on though :) |
| 14:44 |
Bmagic |
I suppose that table doesn't only* contain icon-related rows though... |
| 14:48 |
Bmagic |
I guess everyone is gone - LBMV: at the very least, you should create a LP bug |
| 14:49 |
Dyrcona |
I agree that there should be a way to specify what icons to display so that they are more easily customized. A Launchpad bug is a good place to start a discussion, sometimes. |
| 14:54 |
jeffdavis |
speaking of bugs - bug 1861328 appears to cause some grids to incorrectly show no data |
| 14:54 |
pinesol |
Launchpad bug 1861328 in Evergreen "Error handling needed in flattened_search" [Undecided,New] https://launchpad.net/bugs/1861328 |
| 14:56 |
Dyrcona |
jeffdavis: "Error handling needed." <- FTFY. :) |
| 14:56 |
jeffdavis |
heh |
| 14:58 |
Dyrcona |
Speaking of bugs...We've got 74 bugs tagged "pullrequest" targeted at release 3.4.3, most of these were moved forward from 3.4.2. |
| 14:59 |
Dyrcona |
I plan to review some as time permits. |
| 15:02 |
mmorgan |
Dyrcona++ |
| 15:03 |
mmorgan |
Also worth mentioning, there are 26 bugs with the signedoff tag. |
| 15:03 |
Dyrcona |
Yeah, I didn't run that search but I noticed that several were signedoff. |
| 15:26 |
sandbergja |
Hi developers: I'd really appreciate some feedback on bug 1821094. In particular, gmcharlt made a very excellent point in #14 that I think warrants some discussion |
| 15:26 |
pinesol |
Launchpad bug 1821094 in Evergreen 3.3 "Item status refresh after editing can get confusingly slow" [Medium,Confirmed] https://launchpad.net/bugs/1821094 |
| 15:27 |
sandbergja |
tl;dr the AngularJS ItemSvc sometimes sends a ton of pcrud requests sequentially, which is slow for the user. Sometimes it does them simultaneously, which is fast for the individual user, but can slow things down for everybody else. What is a compromise that keeps everybody's experience performant? |
| 15:33 |
Dyrcona |
Redesign it so that it doesn't have to flood the server with requests. |
| 15:36 |
|
sandbergja_ joined #evergreen |
| 15:42 |
jeffdavis |
sandbergja: In my experience, excessive simultaneous requests are a significant performance issue. I'd rather see sequential requests, but batching would be ideal (and ought to be the standard approach in most places in the web client, IMO). |
| 15:44 |
Dyrcona |
If it's a feature than existed in XUL, maybe see what XUL did. There're a number of places where the XUL client makes a single backend call and the web staff client does it all with pcrud. Maybe this is one of those places? |
| 15:49 |
Dyrcona |
If this isn't one of those place, then implement a single backend call that the web staff client can call to do all of the pcrud more efficiently in the Perl layer. |
| 15:59 |
berick |
batched_apis_that_stream_responses++ |
| 16:26 |
jeffdavis |
I'm not sure whether there is a 1:1 XUL equivalent for the specific action in that bug, but perhaps open-ils.search.asset.copy.batch.retrieve is the API to use? |
| 16:26 |
jeffdavis |
it looks like the closest thing based on a quick look at the copy status XUL |
| 16:27 |
jeffdavis |
I guess there would need to be a new itemSvc.fetchAll method or something that invokes it |
| 16:28 |
jeffdavis |
and I don't know if it plays nice with streaming or what have you |
| 16:29 |
sandbergja_ |
jeffdavis: that's a good lead, thanks! |
| 16:29 |
jeff |
What is the scenario that leads up to needing to re-fetch the entire grid worth of items? Or, is this on an initial batch load of many items from an uploaded file of barcodes? |
| 16:30 |
jeff |
<-- skimmed just enough of the comments to probably ask silly questions |
| 16:30 |
jeffdavis |
jeff: refreshing grid from db after applying batch updates to many rows, I believe |
| 16:32 |
sandbergja_ |
jeffdavis: that's correct |
| 16:33 |
jeff |
does existing code already fetch only the changed items, or is it always re-fetching all items in the grid? i saw the comments about possibly fetching only those items on the current page... |
| 16:34 |
sandbergja_ |
It only fetches the changed items |
| 16:35 |
sandbergja_ |
oh wait, never mind. :-) |
| 16:35 |
sandbergja_ |
I think it currently fetches everything |
| 16:35 |
sandbergja_ |
And my branch only fetches the changed items |
| 17:07 |
|
mmorgan left #evergreen |
| 17:37 |
Bmagic |
Anyone know the scenario where a Purchase order can be filled up with lineitems where some copies get the barcode auto-created and others don't? I'm assuming that the lack of the barocde/callnumber is causing the activation to fail |
| 17:52 |
csharp |
ok, I just fixed the GPLS email lists issue (open-ils-general, -dev, -documentation, commits, etc.) - I deactivated some lists recently and removed a required configuration piece without noticing |
| 17:52 |
csharp |
systemd-- # for not telling me what's going on |
| 17:52 |
csharp |
so you may see an influx of messages |
| 18:01 |
pinesol |
News from qatests: Testing Success <http://testing.evergreen-ils.org/~live> |
| 18:05 |
Bmagic |
csharp++ |
| 18:26 |
|
jihpringle joined #evergreen |
| 18:28 |
jihpringle |
Bmagic: re: your acq questions when I've seen it it's because the activation timed out partway through (activating again usually fixes it) or it's run into an item it doesn't like, such as one attached to a deleted record |
| 19:01 |
|
cmalm_ joined #evergreen |
| 19:50 |
|
cmalm joined #evergreen |
| 21:00 |
|
sandbergja joined #evergreen |
| 21:36 |
|
sandbergja joined #evergreen |
| 22:31 |
|
sandbergja joined #evergreen |
