| Time |
Nick |
Message |
| 07:03 |
|
agoben joined #evergreen |
| 07:03 |
|
cmalm joined #evergreen |
| 07:08 |
|
rjackson_isl joined #evergreen |
| 07:54 |
|
Dyrcona joined #evergreen |
| 07:55 |
|
rjackson_isl joined #evergreen |
| 08:06 |
Dyrcona |
csharp: The samples of Echo in the Canyon sound good, but not sure I want to drop $10 on it. |
| 08:07 |
Dyrcona |
Also, can you imagine the pressure being the offspring of Bob Dylan and a musician yourself? IDK, maybe there's no pressure, but ... |
| 08:39 |
|
mmorgan joined #evergreen |
| 09:25 |
|
tlittle joined #evergreen |
| 09:34 |
|
rfrasur joined #evergreen |
| 09:37 |
|
sandbergja joined #evergreen |
| 09:39 |
|
yboston joined #evergreen |
| 10:02 |
csharp |
Dyrcona: I've heard that when Jakob Dylan was starting out, he played his famous last name down as much as he could because he didn't want to trade on his father's fame |
| 10:02 |
csharp |
but given that he looks like an idealized version of 60s Dylan at his coolest it's probably hard to get away with that :-) |
| 10:03 |
Dyrcona |
Heh. |
| 10:03 |
csharp |
also, the documentary is available for streaming on netflix |
| 10:03 |
csharp |
good interviews with surviving members of those bands, including CSN and a peek of Y at the end :-) |
| 10:07 |
Dyrcona |
Well, $10 was for the soundtrack, and I don't have Netflix. I'm a bit out of it when it comes to consumer/media technology. |
| 10:08 |
csharp |
gotcha |
| 10:08 |
Dyrcona |
All right. I'm bummed at Ubuntu/Debian again. I installed graphicsmagick, and the convert program is nowhere to be found. |
| 10:09 |
Bmagic |
sandbergja: Feel free to make changes to the README - no problem |
| 10:10 |
Dyrcona |
Ah... The GraphicsMagick stuff is still accessed by a program called 'gm' so you can also have ImageMagick installed, but there's no imagemagick package any more. |
| 10:11 |
Bmagic |
sandbergja: however, I will say that your description of the issue is strange and doesn't jive with the docs like csharp said. And it's not been something I've encountered. But if you're seeing this, I'm sure someone else will. Feel free to mention it in the README |
| 10:13 |
Dyrcona |
I know I'm off topic, but that's never stopped me (or anyone else) before, so if you install graphicsmagick-imagemagick-compat, you get the individual utilities like I and my scripts expect. |
| 10:15 |
Dyrcona |
Heh. Helps NOT to typo a package name when searching for it. There are still imagemagick packages. |
| 10:16 |
csharp |
Bmagic: sandbergja: I just cloned the antora repo, installed npm, then ran the gulp command within an lxc ubuntu 18.04 container and saw the same thing sandbergja saw |
| 10:16 |
Bmagic |
ah, in a container.. might be the difference |
| 10:17 |
Dyrcona |
Oh, those pesky containers... |
| 10:17 |
Bmagic |
probably some package that makes the magic happen from a standard ISO install vs. ubuntu:18.04 docker hub image |
| 10:18 |
csharp |
...and I just saw the same happen on my 19.10 desktop |
| 10:20 |
Bmagic |
hmmm |
| 10:21 |
Dyrcona |
Can we blame npm for this? It seems to be responsible for a lot of installation issues. |
| 10:21 |
rfrasur |
Yo all. A "quick" question about org unit/org level selector in the notes/message interface, is the default level a setting or something in the underlying code? |
| 10:25 |
JBoyer |
rfrasur, does it happen to match the OU where your workstation is registered? |
| 10:26 |
rfrasur |
JBoyer, it's not my question or related to a thing I'm working on. It's the underlying question posed by others related to something they might need to think about in the future (how's that for NO answer). |
| 10:27 |
* JBoyer |
stares off into the middle distance as the hamster in the wheel slows to a stop... |
| 10:28 |
JBoyer |
I'm guessing the answer is "something in the code" |
| 10:28 |
mmorgan |
*tilt* |
| 10:29 |
phasefx |
rfrasur: in master, if there is no org depth set on the penalty type that gets chosen, it'll use the workstation library |
| 10:30 |
phasefx |
otherwise, a function in the middle layer will take workstation library and try to find an ancestor at the penalty type depth |
| 10:30 |
mmorgan |
rfrasur: The current notes/message interface does not have an org unit/org level selector visible to the staff user. |
| 10:32 |
rfrasur |
phasefx: thank you. That answers enough of my question. mmorgan: no...I know. Just trying to decide how to proceed with the thing I need to proceed with (in a last minute, timely sort of way). |
| 10:33 |
rfrasur |
Okay, my question has been answered. Thank you all. Sorry for the ambiguity. |
| 10:40 |
csharp |
Bmagic: perhaps it's because npm install is being done by a regular user without the -g flag, which would have installed for all users? |
| 10:41 |
Bmagic |
Yeah, I bet that's it. Because on my test machine, I believe I did it with -g first and then decided that we didn't want that officially |
| 10:41 |
csharp |
or maybe there's a presumption of a $PATH addition when using npm stuff? I've not used it much outside of EG |
| 10:51 |
miker |
anyone interested in taking a look at some pcrud code? https://bugs.launchpad.net/evergreen/+bug/1847805 would love some eyeballs! |
| 10:51 |
pinesol |
Launchpad bug 1847805 in Evergreen "pcrud search can fail to retrieve rows that the user has access to" [High,Confirmed] |
| 11:01 |
pinesol |
News from qatests: Failed Installing AngularJS web client - Expected 6 errors but encountered 5. <http://testing.evergreen-ils.org/~live/test.28.html#2019-11-08T11:00:45,691810346-0500 -0> |
| 11:01 |
pinesol |
News from qatests: Failed Installing Angular web client <http://testing.evergreen-ils.org/~live/test.29.html#2019-11-08T11:00:45,734863712-0500 -2> |
| 11:03 |
|
Christineb joined #evergreen |
| 11:24 |
abneiman |
Dyrcona++ # on behalf of Eric :) |
| 12:03 |
Dyrcona |
abneiman++ |
| 12:05 |
|
jihpringle joined #evergreen |
| 12:08 |
|
sandbergja joined #evergreen |
| 12:09 |
Dyrcona |
miker: I've eyeballed the pcrud cursor code and other than tabs vs. spaces in a couple of lines nothing jumps out at me. I can load the code on a vm and check if anything breaks, but I don't think I've experienced that bug. |
| 12:11 |
Dyrcona |
Actually, I might be able to reproduce it with EDI messages. I'll have a look and see. |
| 12:11 |
miker |
Dyrcona: thanks! it can happen, in a similar vein to ye olde pre-3.0 "search doesn't find items in tiny collections within a large consortium" issue, where a user has access to only a small subset of rows in a (relatively large) table, and ordering of results puts invisible rows at the front of the result set |
| 12:12 |
Dyrcona |
:) |
| 12:12 |
miker |
and its most visible in "magical" pcrud and fielder grids |
| 12:16 |
Dyrcona |
I've got a test db on 3.2.8 with 1.7 million acq.edi_messages. I'll see if I can reproduce it there. |
| 12:20 |
miker |
Dyrcona: awesome, thanks. those two commits should backport fine to 3.2 for testing fwiw |
| 12:20 |
Dyrcona |
miker: Yeah, I don't see any big changes to pcrud since then. |
| 12:21 |
Dyrcona |
Right now, though, I'm looking into something that I noticed while checking the test database. I want to adjust something in the db update that I'm going to do this weekend. |
| 13:34 |
mmorgan |
What's the shortest retention interval folks have set in action triggers? |
| 13:36 |
mmorgan |
Does a 1 day retention interval cause any problems? |
| 13:36 |
Dyrcona |
mmorgan: It depends on the type of event and the other settings. |
| 13:37 |
berick |
mmorgan: most of ours are 1 day |
| 13:38 |
Dyrcona |
We set 180 days across the board, because we often get asked about things weeks after the fact. |
| 13:38 |
berick |
yeah, some stuff sticks around longer |
| 13:39 |
mmorgan |
Ok, thanks. We're looking at 1 day for things like password resets, sms call numbers, print bib records, etc. Six months for triggers related to circ. |
| 13:40 |
mmorgan |
Dyrcona++ |
| 13:40 |
mmorgan |
berick++ |
| 13:51 |
|
yboston joined #evergreen |
| 13:52 |
Bmagic |
I'm taking a stab at making the OPAC search box in various screens autofocus. The HTML5 property autofocus does a good job but not when it's in an iframe like in the staff client. So I turned to JS. But the JS on the page inside the iframe doesn't seem to execute. Anyone solved this before? |
| 13:52 |
Dyrcona |
miker: I should have been timing page loads, but parts of acq feel faster with that cursor branch installed. |
| 13:55 |
Dyrcona |
Eh, maybe I just got "lucky." :) |
| 14:03 |
Dyrcona |
Bmagic: "A better approach is not to use IFrames." |
| 14:03 |
Bmagic |
lol |
| 14:03 |
Bmagic |
tell Evergreen that |
| 14:04 |
Dyrcona |
I...uh....just did. :) |
| 14:04 |
Dyrcona |
Kind of like when I lookup mod_perl issues and the top answer is almost always: Don't use mod_perl. :) |
| 14:05 |
Bmagic |
yeah! Haha, lotta help. Thanks internet |
| 14:06 |
Dyrcona |
Well, they are the most widely accepted answers by the JavaScript and Perl communities, respectively. |
| 14:07 |
Dyrcona |
And, now, I have a bone to pick with Chromium 78. It's acting up again and won't show me my browser history. |
| 14:09 |
Dyrcona |
Restarting helps. I think I'm going to disable automatic installation of security updates. I'm blaming this weirdness on snap updates. |
| 14:10 |
Dyrcona |
Oh, look... Another Chromium snap update..... |
| 14:11 |
Dyrcona |
And, it's missing the history where I found that "answer" on stackoverflow, of course. |
| 14:37 |
|
stephengwills joined #evergreen |
| 14:42 |
miker |
Dyrcona: I would not be surprised at all if some interfaces are faster, esp if they use streaming pcrud responses, because the cursor should return the first result faster on balance |
| 14:42 |
pinesol |
[evergreen|Bill Erickson] LP1851831 Group perm editor null descriptions OK - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=dfd1fd9> |
| 14:47 |
Dyrcona |
I always have regarded cursors as slow, but maybe that's ancient Sybase baggage from the '90s. :) |
| 14:53 |
csharp |
we're trying out carousels in PINES - one thing we're wondering is the requirement for shelving locations - should we need those if the carousel is populated from a bucket? |
| 14:55 |
csharp |
log message is Carousel for bucket 1918604 is misconfigured; copy location filter expected but none specified |
| 15:02 |
csharp |
oof - looks like the admin UI for carousels is doing a perm check for every record bucket in the DB - which in PINES is BAD |
| 15:03 |
csharp |
checking CREATE_BIB_BUCKET and ADMIN_BIB_BUCKET for each one |
| 15:04 |
csharp |
terran is reporting that this causes the modal to take minutes to load |
| 15:09 |
mmorgan |
One more question about action trigger retention - if no retention interval is set, they'll stay, forever. True? |
| 15:13 |
Dyrcona |
mmorgan: True. |
| 15:14 |
mmorgan |
Thanks! |
| 15:14 |
Dyrcona |
I was going to point out earlier that for passive hooks, retention interval also has to be > (max_delay - delay). |
| 15:15 |
Dyrcona |
If not, event get recreated. |
| 15:16 |
mmorgan |
Oh, good point! |
| 15:58 |
miker |
csharp: hrm.... we should restrict whatever is asking for the list of ALL THE BUCKETS to just those of the proper type, if we're not, I assume. (I don't have the code in front of me ATM to look ... and am about to turn into a pumpkin ... but if you want somewhere to look ;) ) |
| 16:27 |
csharp |
miker: thanks |
| 16:27 |
* csharp |
converts to pumpkin mode too |
| 16:47 |
gmcharlt |
csharp: miker: at first glance, this appears to be a fm-record-editor bug |
| 16:49 |
gmcharlt |
specifically, even though the modal has a separate template (bucketTemplate) for displaying the bucket, rather than the default widget (which would be a combobox of all the buckets), it's instructing the data source for the non-existant combobox anyway |
| 16:49 |
gmcharlt |
and initiating a search of all rows in the linked table |
| 16:49 |
gmcharlt |
berick: ^^ may be of interest to you |
| 16:50 |
berick |
indeed |
| 16:50 |
gmcharlt |
berick: and as a gut reaction, I think preloadLinkedValues needs to default to false rather than true |
| 16:51 |
berick |
gmcharlt: agreed. i was thinking about preloadLinkedValues earlier this week.. kind of wonder if it should go away. it's essentially a sync data source at that point. |
| 16:51 |
berick |
it just delays loading |
| 16:51 |
berick |
which may be of value |
| 16:52 |
gmcharlt |
and certain until we have https://bugs.launchpad.net/evergreen/+bug/1830804 ... |
| 16:52 |
pinesol |
Launchpad bug 1830804 in Evergreen ""small", "medium", and "large" IDL table attributes might be useful" [Wishlist,New] |
| 16:52 |
gmcharlt |
(or the like) |
| 16:52 |
berick |
yeah |
| 17:00 |
berick |
gmcharlt: oops, my comments about making stuff go away... I was thinking about combobox's asyncSupportsEmptyTermClick (not preloadLinkedValues) |
| 17:00 |
gmcharlt |
ah, OK |
| 17:00 |
gmcharlt |
but yeah, preloadLinkedValues is still dangerous, at least if used indiscrimnately |
| 17:01 |
berick |
agreed |
| 17:01 |
berick |
IIRC, it was set to true to make the pile of incoming admin pages JustWork |
| 17:01 |
gmcharlt |
I've opened bug 1851884 to separate the general issue from the specific carousels admin issue |
| 17:01 |
pinesol |
Launchpad bug 1851884 in Evergreen "eg-fm-record-editor: take care wiring up comboboxes" [Medium,New] https://launchpad.net/bugs/1851884 |
| 17:01 |
berick |
(until one fetches every record in the database, at least) |
| 17:02 |
berick |
gmcharlt++ |
| 17:08 |
|
mmorgan left #evergreen |
| 17:26 |
gmcharlt |
whoops, I've kinda ended up doing a stream of conciousness on that bug |
| 18:12 |
csharp |
gmcharlt++ # thanks for the investigation |
| 18:14 |
|
stephengwills left #evergreen |
| 18:49 |
|
dbwells_ joined #evergreen |
| 21:02 |
|
sandbergja joined #evergreen |
| 22:41 |
|
sandbergja joined #evergreen |
| 23:01 |
pinesol |
News from qatests: Failed Installing AngularJS web client - Expected 6 errors but encountered 7. <http://testing.evergreen-ils.org/~live/test.28.html#2019-11-08T23:00:53,948801253-0500 -0> |
| 23:01 |
pinesol |
News from qatests: Failed Installing Angular web client <http://testing.evergreen-ils.org/~live/test.29.html#2019-11-08T23:00:53,989972818-0500 -2> |
