Time |
Nick |
Message |
01:22 |
|
dluch joined #evergreen |
01:22 |
|
Bmagic joined #evergreen |
01:38 |
|
Bmagic joined #evergreen |
01:38 |
|
dluch joined #evergreen |
06:00 |
pinesol |
News from qatests: Failed Installing Evergreen database pre-requisites <http://testing.evergreen-ils.org/~live//archive/2022-02/2022-02-22_04:00:02/test.39.html> |
07:16 |
|
rjackson_isl_hom joined #evergreen |
07:32 |
|
collum joined #evergreen |
08:26 |
|
rfrasur joined #evergreen |
08:27 |
|
mantis joined #evergreen |
08:33 |
|
mmorgan joined #evergreen |
08:34 |
|
jvwoolf joined #evergreen |
09:05 |
|
Dyrcona joined #evergreen |
09:07 |
|
Keith__isl joined #evergreen |
10:04 |
csharp_ |
somebody commit something to master so I can see whether the @rss version of commit announcement suffices for our purposes :-) |
10:06 |
csharp_ |
Dyrcona: I missed that you had included removing PG 9.6 on your branch - makes sense, but I wasn't seeing it before :-/ |
10:11 |
csharp_ |
Dyrcona: I'm interested in seeing that branch go in... to fully test it, would I want to run all the PGtap tests on PG 10-14? or is there another way? |
10:13 |
csharp_ |
(this is why I prefer more granular bugs/fixes, btw - not trying to be critical since you've done so much work though :-) ) |
10:28 |
Dyrcona |
csharp_: Commit the branch that updates the PostgreSQL prerequisite targets. That should test the rss feeds. :) |
10:29 |
Dyrcona |
My branch changes the default client to Pg 14, IIRC. I'm using it on almost all of my VMs with Pg 10 databases with no issues. |
10:30 |
Dyrcona |
As for more granular bug fixes, I mostly agree. I don't like the giant omnibus branches that sometimes appear. I also prefer more, smaller commits over larger commits that touch more lines/files. |
10:30 |
Dyrcona |
In the case of my branch, I think it's scope is fine. I don't see why removing Pg 9.6 and adding support for newer Pg version should be separate. |
10:32 |
csharp_ |
Dyrcona: makes sense - I wasn' |
10:32 |
csharp_ |
t speaking specifically about removal of 9.6 - it just looks like a whole lot to test at once :-) |
10:33 |
csharp_ |
but, I'm going for it, so we'll see :-) |
10:33 |
Dyrcona |
Nah, not so much. Just inst all Pg 10, if the tests pass, that's all you've really got to worry about. I say int he README that the other versions are not fully tested and may have performance issues. |
10:34 |
* Dyrcona |
wishes his fingers would cooperate. |
10:34 |
Dyrcona |
csharp_++ |
10:34 |
csharp_ |
ok, that's where I was starting anyway |
10:35 |
Dyrcona |
If you want to test a different Pg version, I'd recommend Pg 14. |
10:35 |
Dyrcona |
It has the most changes from Pg 9.6/10, and it is the latest. |
10:37 |
csharp_ |
will do |
10:40 |
Dyrcona |
I should use the other installations on my development db server more. I've mostly just been using Pg 10 lately. |
10:41 |
csharp_ |
Dyrcona: do you have the alternate versions running together on different ports? |
10:42 |
Dyrcona |
csharp_: Yes. When I want to test one version in particular, I'll swap out a configuration file with optimized parameters and reboot. |
10:42 |
Dyrcona |
If I'm not too fussed for performance, I'll just change the ports in my client config. |
12:00 |
|
jihpringle joined #evergreen |
12:04 |
csharp_ |
Dyrcona: I see this when running pg_prove on PG 10/master + your branch: https://pastebin.com/cYA4qzDW |
12:18 |
Dyrcona |
It's missing an @ I think. |
12:18 |
Dyrcona |
I had different code there and miker said I should change it. |
12:18 |
Dyrcona |
I'll fix it. |
12:20 |
csharp_ |
Dyrcona++ |
12:22 |
Dyrcona |
NB: I'm not blaming miker for the bug. That looks like one of the places where I changed it to use namespaces, and miker thought we should use either namespaces or local-name() everywhere. I switched to local-name() because there were more instances of it that there were namespaces. |
12:24 |
Dyrcona |
Actually, I think it is missing an * not an @. |
12:25 |
miker |
I think the problem is a missing * between // and [ |
12:25 |
miker |
heh, what Dyrcona said |
12:26 |
* miker |
fades back into the north Georgia mountain mists (vacation!) |
12:28 |
Dyrcona |
Yeah, exactly what miker said. |
12:28 |
Dyrcona |
I'll check the upgrade script, too. It probably has the same error. |
12:29 |
Dyrcona |
Huh. It's correct in the upgrade script. I must not have copied and pasted. |
12:33 |
Dyrcona |
Greping the commit diff, that looks like the only place that omission happened. |
12:34 |
Dyrcona |
csharp_: Do you prefer I add a commit with the fix and push it, or do you prefer I rebase it and squash the fix into the previous one? |
12:40 |
Dyrcona |
csharp_: I just did the former, push a new commit. You should be able to just pull the branch, unless you've merged or rebased it on something else in the meantime. |
12:42 |
csharp_ |
thanks |
12:44 |
Dyrcona |
csharp_++ Thank you for catching this. |
13:00 |
csharp_ |
Dyrcona: now I'm getting https://pastebin.com/LjE1DtxJ (I cherry-picked your latest commit on top and rebuild the DB) |
13:02 |
csharp_ |
s/rebuild/rebuilt/ |
13:02 |
csharp_ |
gonna step away for a few |
13:09 |
Dyrcona |
Another typo. I thought I tested these after making the changes.... |
13:10 |
Dyrcona |
Again, the db upgrade script is correct. Maybe I only ran the tests after running the db upgrade with the switch from namespace to local-name()? |
13:14 |
Dyrcona |
OK. This one looks like the last issue. I've poured over the diffs and don't see any other changed XPath that looks bad. |
13:16 |
Dyrcona |
Fortunately I put the XPath canges all in one commit. |
13:16 |
Dyrcona |
s/canges/changes/ |
13:16 |
Dyrcona |
typos-- |
13:19 |
Dyrcona |
csharp_: Same deal I pushed a new commit. |
13:22 |
* Dyrcona |
sighs. Some days are more typo-prone than others. |
14:41 |
csharp_ |
yay!: |
14:41 |
csharp_ |
Result: PASS |
14:41 |
csharp_ |
ok, now onto PG 14 |
15:01 |
csharp_ |
Dyrcona: ok, PG 14 pgtap tests pass too - is there anything else I should test? |
15:07 |
Dyrcona |
You could try the Perl tests. Do anything you'd normally do, like circulation or whatever. |
15:08 |
csharp_ |
10-4 |
16:07 |
Dyrcona |
So, I'm trying to figure out how the bootstrap OPAC knows to load main_logo.png. So far it just seems to be magic. My motivation is, I want to replace it with CWMARS' Logo, name home-logo.png. I'd rather not just replace main_logo.png, but I will if I have to. |
16:08 |
* csharp_ |
looks around for terranm but doesn't see her |
16:08 |
csharp_ |
she's our resident boostrap OPAC person :-) |
16:08 |
csharp_ |
@seen terranm |
16:08 |
pinesol |
csharp_: terranm was last seen in #evergreen 2 years, 23 weeks, 6 days, 4 hours, 39 minutes, and 31 seconds ago: <terranm> Bmagic++ for setting up bug squashing test server! https://docs.google.com/spreadsheets/d/1qYNGrJBt42_ArQzbwlcnKxS3TobDnmquSXZzopOkZh4/edit#gid=0 |
16:08 |
csharp_ |
huh I guess she's using a different nick lately :-/ |
16:10 |
Dyrcona |
My suspicion is it has something to do with glide.js, but my suspicions are frequently incorrect. |
16:11 |
csharp_ |
https://gapines.org/opac/images/main_logo.png - she must have figured out how to change it |
16:12 |
csharp_ |
oh - we don't have a main logo in place - we just have the carousel |
16:12 |
csharp_ |
that's probably why |
16:13 |
* csharp_ |
stops talking about stuff he doesn't know :-) |
16:13 |
Dyrcona |
Hmm. So I'm wronger than I thought. I get the same image for the equivalent URL on my server, but that's not the logo that shows up in the bootstrap OPAC. |
16:13 |
Dyrcona |
csharp_++ |
16:13 |
csharp_ |
https://catalog.cwmars.org/opac/images/home-logo.png?749f4c |
16:13 |
Dyrcona |
It's quantum computing... |
16:13 |
csharp_ |
that's what you're trying to change? |
16:14 |
Dyrcona |
I'm trying to change to that from the default logo in the OPAC on our training server: https://training.cwmars.org/eg/opac/home |
16:15 |
csharp_ |
ah, I see now |
16:16 |
csharp_ |
https://training.cwmars.org/images/Evergreen-Final-Logo.png - is that the file you're trying to replace? |
16:17 |
Dyrcona |
I guess so. |
16:17 |
Dyrcona |
Ah, found it! |
16:18 |
Dyrcona |
It's set in Open-ILS/src/templates-bootstrap/opac/css/style.css.tt2 as a background image. |
16:18 |
Dyrcona |
csharp_++ |
16:18 |
csharp_ |
ah |
16:18 |
csharp_ |
CSS IS AWESOME |
16:18 |
Dyrcona |
I was searching for the wrong things: main_logo, logo, all in lower case. |
16:19 |
Dyrcona |
Heh. |
16:20 |
Dyrcona |
That would make a good series for introductory computer books for kids: [Blank] IS AWESOME |
16:21 |
Dyrcona |
HTML IS AWESOME, CSS IS AWESOME, PYTHON IS AWESOME, PERL IS AWESOME, JAVA IS AWESOME.... :) |
16:22 |
Dyrcona |
I don't think our logo is going to work as a background image. |
16:23 |
Dyrcona |
I'll give it a go and see what happens. It's git, I can always back it out. |
16:25 |
csharp_ |
per Terran: Actually, on terran-master it looks like it's in templates-bootstrap/opac/parts/topnav_logo.tt2 |
16:27 |
csharp_ |
she corrected herself - the topnav location isn't what you're talking about |
16:27 |
Dyrcona |
Right. |
16:27 |
Dyrcona |
I'm likely going to modify the topnav location at some point, too. |
16:28 |
Dyrcona |
Well, it's on the list. |
16:28 |
Dyrcona |
Actually, it doesn't look so bad if you go to our training OPAC, now. |
16:28 |
Dyrcona |
It's been scaled up, so it looks a little jagged. |
16:29 |
Dyrcona |
It was anti-aliased, so I'm not sure that I can fix that without the original from which the png was made. |
16:30 |
Dyrcona |
I suspect that we're either going to remove the logo, or move it below the carousels. We put it below the carousels in ttopac. |
16:41 |
Dyrcona |
We might keep it, but make it smaller. I don't think we have as many carousels in production. It looks like some has been playing around on training. |
16:42 |
Dyrcona |
s/some/someone/ |
16:47 |
|
eady joined #evergreen |
16:50 |
|
eady joined #evergreen |
16:52 |
|
tsadok joined #evergreen |
16:52 |
|
jvwoolf left #evergreen |
17:02 |
|
jihpringle joined #evergreen |
17:09 |
|
mmorgan left #evergreen |
18:00 |
pinesol |
News from qatests: Failed Installing Evergreen database pre-requisites <http://testing.evergreen-ils.org/~live//archive/2022-02/2022-02-22_16:00:02/test.39.html> |