Time |
Nick |
Message |
06:01 |
pinesol |
News from qatests: Failed Running autogen.sh <http://testing.evergreen-ils.org/~live//archive/2022-04/2022-04-06_04:00:02/test.44.html> |
07:16 |
|
rjackson_isl_hom joined #evergreen |
07:43 |
|
collum joined #evergreen |
08:07 |
|
rfrasur joined #evergreen |
08:37 |
|
mantis joined #evergreen |
08:38 |
|
mmorgan joined #evergreen |
08:48 |
|
terranm joined #evergreen |
09:08 |
|
Dyrcona joined #evergreen |
10:09 |
|
jvwoolf joined #evergreen |
10:29 |
|
terranm joined #evergreen |
10:31 |
|
Keith-isl joined #evergreen |
10:46 |
Dyrcona |
The test server failures look like something peculiar to the test server environment. When I run autogen.sh and the tests on a local VM, they work. |
11:10 |
bshum |
Hmm, what I've found is that on the first attempt to run, it fails. But subsequent attempts work. |
11:10 |
bshum |
First try -- Unable to write to /openils/var/web/opac/extras/ac, please check |
11:10 |
bshum |
Second try, autogen goes through |
11:13 |
bshum |
I think that's cause the main directory isn't being created initially, but the other changes to add the subfolders for the cover small/r, etc. are actually creating the whole path |
11:13 |
bshum |
Including the top level folder |
12:00 |
|
jihpringle joined #evergreen |
12:23 |
JBoyer |
It may be better to just create those empty dirs and drop a .placeholder in them so they can just be committed and not special-cased. |
12:24 |
JBoyer |
Though I guess depending on how explicit the release notes were re: creating those dirs yourself I can see some appeal to auto-creating them somehow. Not a big fan though. |
12:40 |
* Dyrcona |
didn't read the release notes to be honest. I was only looking into it because bshum mentioned that the test was failing to me the other day. |
12:42 |
Dyrcona |
bshum: I have installed Evergreen before on the VM where I checked, so that may have something to do with the differences in what you and I see. I didn't try on a clean virtual machine. |
12:55 |
Dyrcona |
bshum: You are correct. The $COVERDIR is not created before the check for its existence in autogen.sh, then it is created when the small, medim, and large subdirectories are created. |
12:56 |
Dyrcona |
My inclination is to remove the check for the COVERDIR being writable and just let it be created if necessary when the small, medium, and/or large subdirectories are created. |
12:57 |
miker |
unrelated to -^, and I'll be following up on the dev list, but make_release is running now to create a beta tarball (finally) |
12:57 |
Dyrcona |
miker++ |
12:57 |
* miker |
taps fingers waiting for the ang SC to build so he can extract and upload strings to POEditor... |
12:59 |
|
jihpringle joined #evergreen |
12:59 |
Dyrcona |
rxjs is still really slow. |
13:00 |
miker |
given that bug squashing is ongoing, I'm not inclined to force the "release"-related commits for beta into master, but do that at RC1 instead. we'll be doing the i18n dance again anyway, and that's the one part that isn't totally recreated at each make_release run AFAICT |
13:02 |
Dyrcona |
Yeah, the i18n dance has to be repeated each time. It could benefit from automation/addition into make_release. |
13:02 |
mantis |
Working on this bug: https://bugs.launchpad.net/evergreen/+bug/1901895 |
13:02 |
pinesol |
Launchpad bug 1901895 in Evergreen "Docs: Carousel documentation clarification" [Undecided,Confirmed] |
13:03 |
mantis |
Is this an appropriate explanation of the age limit field? |
13:03 |
mantis |
"filters items by how long ago they were adjusted in the system and if it fits within the age limit." |
13:03 |
mantis |
or is just using 'date active' fine? I'm unsure if every carousel type checks that or another date field |
13:07 |
miker |
(note for release watchers: the relator code list from LC /claims/ it still hasn't changed since 2013, so we've got that going for us, which is nice) |
13:07 |
Dyrcona |
:) |
13:09 |
miker |
huh, the little spinner "icon" in the console showing progress in the npm compilation is cool. I guess it uses a braille font or unicode characters? |
13:13 |
miker |
welp, poeditor has changed enough that the instructions for angular transations can't be followed, AFAICT |
13:15 |
miker |
(or, more likely, I don't have access to the import/export functions for the project. so, /I/ can't follow the instructions) |
13:20 |
miker |
and in my absence from releasing, I think I aged out of web server access as well |
13:25 |
miker |
time for some lunch |
14:00 |
phasefx |
berick: as far as you know, Field Documentation only shows up in the Patron Editor? |
14:08 |
|
collum joined #evergreen |
14:09 |
|
jihpringle joined #evergreen |
14:15 |
berick |
phasefx: as far as I recall, yes |
14:17 |
phasefx |
berick: gracias; I was pondering whether we should filter that class list to just the fm objects for supported interfaces, but I'll leave that for another day, another bug. Right now I'm working on your suggeestions for additions on the angular port ticket |
14:17 |
berick |
phasefx: yeah, agreed filtering would be ideal |
14:30 |
|
terranm joined #evergreen |
15:50 |
Dyrcona |
I'm going to a 3.9-rc target on Lp since 3.9beta is essentially done. |
15:50 |
Dyrcona |
to add a 3.9-rc target... :/ |
16:04 |
|
terranm joined #evergreen |
16:09 |
pinesol |
News from commits: LP1920039-Bootstrap Opac: More Details button doesn't change on use. <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=9c1855e0b9eac816d982e5e2029115da08f0ea20> |
16:27 |
miker |
Dyrcona++ |
16:39 |
pinesol |
News from commits: LP#1954923: Fix current date issue in Boostrap OPAC Circ History CSV export <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=e08d76965aaa5c30d1344e6d778577803b2d0723> |
17:05 |
|
mmorgan left #evergreen |
17:30 |
|
jvwoolf left #evergreen |
18:01 |
pinesol |
News from qatests: Failed Installing Angular web client <http://testing.evergreen-ils.org/~live//archive/2022-04/2022-04-06_16:00:03/test.29.html> |
18:13 |
|
jmurray-isl joined #evergreen |
18:13 |
|
rjackson_isl_hom joined #evergreen |
18:41 |
|
rjackson_isl_hom joined #evergreen |