| Time |
Nick |
Message |
| 06:01 |
pinesol |
News from qatests: Failed Running pgTAP tests <http://testing.evergreen-ils.org/~live//archive/2021-09/2021-09-24_04:00:03/test.42.html> |
| 07:26 |
pinesol |
[evergreen|Jason Boyer] LP1928258: Correct Upgrade Script - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=bec5712> |
| 07:27 |
|
rjackson_isl_hom joined #evergreen |
| 07:30 |
JBoyer |
berick, Will you have time today to update the failling regression test for Vandelay? It's related to the update editor / update bib sources changes from LP 1928258. |
| 07:30 |
pinesol |
Launchpad bug 1928258 in Evergreen "Add separate "update edit date / editor" fields for Vandelay merge profiles" [Wishlist,Fix committed] https://launchpad.net/bugs/1928258 |
| 07:31 |
|
mantis joined #evergreen |
| 08:42 |
|
mmorgan joined #evergreen |
| 08:50 |
JBoyer |
Seems like running an aggressive git gc and updating XCode at the same time on this machine was a poor choice, if the lamentations of this fan are any indication. |
| 08:52 |
|
rfrasur joined #evergreen |
| 08:54 |
csharp_ |
heh |
| 09:11 |
Bmagic |
JBoyer++ # bug 1174498 |
| 09:11 |
pinesol |
Launchpad bug 1174498 in Evergreen "Payment by billing type breakdown" [Wishlist,Confirmed] https://launchpad.net/bugs/1174498 |
| 09:12 |
Bmagic |
JBoyer: As I was revising that code a few weeks ago, I had the thought of using the ID that is recorded in money.account_adjustment (billing bigint) to assign those payments. Sure seems like that would take care of this scenario? |
| 09:14 |
JBoyer |
That sounds like they would very definitely still be assigned to the voided payments then, which is the current problem |
| 09:15 |
JBoyer |
The only way to care as much about matching up adjustment values to specific billings without including voided ones would be to do something like this: |
| 09:15 |
JBoyer |
gather only account adjustments for this xact in date order |
| 09:15 |
JBoyer |
gather all non-voided billings for this xact in date order |
| 09:16 |
JBoyer |
starting with the oldest aa payment look for matching billings |
| 09:16 |
JBoyer |
when found, return lines for both and discard |
| 09:17 |
JBoyer |
if you run out of billings, the remaining aa payments just get thrown on a pile with all other payments on that xact and are no longer treated specially |
| 09:17 |
JBoyer |
then at the end do the negative payment / refund stuff. |
| 09:18 |
JBoyer |
Most importantly would be that it's not done in a single 500+ line sql function |
| 09:20 |
JBoyer |
(Note also that off-the-dome pseudo code probably isn't really complete enough to be of much help, but this is apparently going to have to be done in stages rather than for payments loop ; for bills loop; end loop ; end loop; |
| 09:21 |
rhamby |
Or everyone stops charging fines and agrees that they're fine with abstracting what few payments they do take. ( I know, I know. ) |
| 09:23 |
JBoyer |
rhamby++ |
| 09:24 |
JBoyer |
That's the kind of out-of-the-box thinking we appreciate here at Software, Inc. |
| 09:26 |
rhamby |
"look, this doesn't work, change your use cases" |
| 09:26 |
rhamby |
or if you're Steve Jobs - "You're holding it wrong" |
| 09:27 |
|
stephengwills joined #evergreen |
| 09:30 |
JBoyer |
There are cases where I'd like to take more stances like that (though maybe not to those extremes, heh.) Open Source shouldn't necessarily mean infinite customizability via a choose-your-own-adventure of limitless settings. Eventually users to need to do some things differently. (This isn't related to mmpbt or any other thing in particular, but is something I've felt for a long time.) |
| 09:36 |
stephengwills |
heard. Balsam operates on very limited funds and, after 12 years, I am still part time. This necessitates my libraries conforming to common procedures. |
| 09:43 |
Bmagic |
JBoyer: It seems odd to me that we would apply account adjustment payments to billings that weren't account adjusted. That feature was implemented after this code was written, and then this code got some special cases added to accomodate. Turns out, the aa feature provides the map for us by virtue of the table giving us billing ID |
| 09:44 |
jeff |
adjustments don't "map" to bills, though. that ID there is just something that dbwells put in place to prevent "multiple adjustments of the same billing" |
| 09:45 |
Bmagic |
jeff: really? Isn't it a foreign key to money.billing? |
| 09:46 |
JBoyer |
That's how the xact balance is treated. If there are 10 account_adjustments but they were "applied" to are all voided, they're still payments so far as any other billings on that xact are concerned. |
| 09:46 |
jeff |
Bmagic: it may be, but that doesn't mean that other changes to the transaction won't mean that the adjustment is applied to a different bill. |
| 09:47 |
jeff |
if there were something that tied payments to billings like that, I don't think we'd need mmpbbt. :-) |
| 09:47 |
JBoyer |
"but they" = "but the billings they" |
| 09:47 |
Bmagic |
yeah, agreed, if the billings that were originally aa'ed get voided, then the aa has to go somewhere* |
| 09:48 |
JBoyer |
And that's the issue with the current mmpbt func, where they go isn't necessarily un-voided (and for the most part explicitly isn't in that example xact in the test) |
| 09:51 |
|
jvwoolf joined #evergreen |
| 09:52 |
Bmagic |
Well, wait a minute, it doesn't pay voided bills at all |
| 09:53 |
JBoyer |
The steps in my comment at the bottom of the bug show how that happens. Billings were adjusted, later they were voided, now things are mis-applied in mmpbt. |
| 09:54 |
jeff |
I'm going to try to make time to look at the tests. I appreciate that they're there now. :-) |
| 09:54 |
JBoyer |
They're poorly named and too basic, but can help make sure it doesn't do something silly. |
| 09:56 |
Bmagic |
if the payment is negative, then it pays voided billings as the last step. The issue is there somehow? (Because that's the only logic that deals with voided billings) |
| 10:02 |
* mmorgan |
tries to wrap head around negative payments applying to voided billings:-/ |
| 10:03 |
pinesol |
[evergreen|Terran McCanna] LP1909144 BooPAC Login Form Tabbing - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=0fb9e37> |
| 10:05 |
JBoyer |
terranm++ |
| 10:06 |
mmorgan |
terranm++ |
| 10:07 |
jeff |
Billing deals 2 arcane damage to target org unit. If this transaction's damage hits, go again. |
| 10:09 |
Bmagic |
alright, apologies, I see where voided billings are included in the results outside of negative payments. And the special case for account adjustments hunting for identical amounts. IIRC, that logic was added to fix the opposite problem, where aa payments were applying to something else. |
| 10:10 |
Bmagic |
JBoyer: back to your solution above - when aa payment lines up with a voided billing, throw them both out? And don't record them at all in mmpbt? |
| 10:11 |
JBoyer |
You don't consider voided billings at all in my above list of steps (until it's time to look at negative payments) If there are "leftover" AA payments after matching up all of the non-voided billings you can, then they're just treated as regular payments from then on, because that's how the rest of the system treats them. |
| 10:14 |
JBoyer |
And when I said discard them, I meant after returning the row matching the aa payment and non-voided bill. Since they're equal amounts you know you don't need to have either id in the future. |
| 10:16 |
JBoyer |
That was an under-specified, off the dome, stream of consciousness type list of steps, I wouldn't try to follow it *too* closely. |
| 10:17 |
Bmagic |
Fair enough, I'll take a closer look at this code again soon |
| 10:21 |
Bmagic |
It would be nice if it was easier to follow (broken up somehow) - I have to re-learn it everytime. Though, an argument can be made that splitting the login into more functions can make it confusing too :). The repeated logic could easily be a seperate function, which could reduce the total number of lines by 100 or so. Thereby making it easier to follow because the beginning and ending of a loop could fit on the same screen :) |
| 10:22 |
Bmagic |
seperate/separate login/logic |
| 10:49 |
|
jvwoolf joined #evergreen |
| 10:51 |
pinesol |
[evergreen|Bill Erickson] LP1938729 Cache "cascade" setting values - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=3887004> |
| 10:57 |
csharp_ |
I must have missed a commit where chromium and firefox are no longer installed by the -developer Makefile.install targets? |
| 10:58 |
csharp_ |
(testing the bullseye EG/OpenSRF branches and npm run test is failing |
| 10:58 |
csharp_ |
) |
| 10:59 |
JBoyer |
csharp_, That commit is the first one in the bullseye branch. |
| 10:59 |
JBoyer |
it's from a branch for a different lp though I believe. |
| 11:00 |
csharp_ |
ah - I didn't cherry-pick that one |
| 11:00 |
csharp_ |
50df96378b |
| 11:00 |
csharp_ |
oh right - it's a working repo commit so pinesol doesn't know about it |
| 11:03 |
csharp_ |
ok, I see now - it's in bug 1850156 |
| 11:03 |
pinesol |
Launchpad bug 1850156 in Evergreen "Need a -git make target" [Wishlist,Confirmed] https://launchpad.net/bugs/1850156 |
| 11:04 |
csharp_ |
I guess I'm testing a buncha muncha cruncha bugs at the same time :-) |
| 11:09 |
mmorgan |
csharp_: Buncha Muncha Cruncha Bugs sounds like a nutritious breakfast cereal! |
| 11:09 |
* berick |
hankers for a hunk o' cheese |
| 11:11 |
rfrasur |
Amish cheese? |
| 11:13 |
berick |
um, of course |
| 11:22 |
mantis |
someone say Amish cheese? |
| 11:23 |
rfrasur |
lol, always and forever. |
| 11:24 |
rhamby |
now I want cheese ... which is normal. sigh. |
| 11:27 |
rhamby |
JBoyer - regarding the infinite expanse of choose your own adventure settings chsarp and I talked last year at the Hack-a-way about an idea of creating sort of optional setting profiles where people could choose to hide granular settings for "you want it to act like this, then choose this one thing and sets all the sub ones for you" - I'd still like to find bandwidth to work on that |
| 11:27 |
rhamby |
maybe we'll find time by this year's hack-a-way |
| 11:36 |
rfrasur |
I had cheese for breakfast. Unnecessary info. It was goat cheese. I have no regrets. Or current desires for cheese. It'll return though. |
| 12:14 |
|
jihpringle joined #evergreen |
| 12:18 |
|
terranm joined #evergreen |
| 12:29 |
* mmorgan |
wonders how many long standing billing bugs could be closed if all libraries eliminated fines :) |
| 12:30 |
terranm |
lol |
| 12:31 |
rfrasur |
mmorgan, from your fingertips to God's monitor. |
| 12:32 |
|
Dyrcona joined #evergreen |
| 12:35 |
mmorgan |
rfrasur: Heh. Reminds me of https://www.memedroid.com/memes/detail/136234 |
| 12:35 |
rfrasur |
mmorgan++ # LOL!!!! |
| 12:36 |
mmorgan |
Still have that page from an old flip calendar on my bulletin board!! :) |
| 12:37 |
* mmorgan |
sometimes wishes for that Smite key! |
| 12:37 |
rfrasur |
I might wish for it more than is actually healthy (for me or anyone else). |
| 12:38 |
|
Dyrcona joined #evergreen |
| 12:39 |
mmorgan |
Yes! Probably a good thing that such a keyboard button doesn't actually exist for mere mortals. |
| 12:42 |
terranm |
I'd have used it at least 3 times this morning |
| 12:43 |
mmorgan |
terranm++ |
| 12:43 |
rfrasur |
terranm++ # times infinity |
| 12:44 |
jvwoolf |
terranm++ |
| 12:44 |
jvwoolf |
mmorgan++ |
| 13:36 |
|
jvwoolf joined #evergreen |
| 13:50 |
|
jvwoolf joined #evergreen |
| 14:06 |
|
jvwoolf joined #evergreen |
| 14:10 |
JBoyer |
terranm++ lots of good carousel work on the way. |
| 14:10 |
terranm |
And miker++ for the help with the query string |
| 14:14 |
pinesol |
[opensrf|Jason Boyer] LP1940146: Debian Bullseye Installation Support - <http://git.evergreen-ils.org/?p=OpenSRF.git;a=commit;h=85f483d> |
| 14:14 |
pinesol |
[opensrf|Jason Boyer] LP1940146: Update documentation - <http://git.evergreen-ils.org/?p=OpenSRF.git;a=commit;h=7c97b79> |
| 14:14 |
pinesol |
[opensrf|Jason Boyer] Remove XUL opensrf.math Client - <http://git.evergreen-ils.org/?p=OpenSRF.git;a=commit;h=bb8723d> |
| 14:14 |
pinesol |
[opensrf|Jason Stephenson] LP1940146: Clarify README - <http://git.evergreen-ils.org/?p=OpenSRF.git;a=commit;h=276e0d2> |
| 14:18 |
pinesol |
[evergreen|Jason Boyer] LP1850156: Remove Browsers From *-developer Targets - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=cbed258> |
| 14:18 |
pinesol |
[evergreen|Jason Stephenson] Lp1940145: Remove Installation Support for Debian Jessie - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=0ebe447> |
| 14:18 |
pinesol |
[evergreen|Jason Stephenson] LP1940146: Debian Bullseye Installation Support - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=e6097bf> |
| 14:19 |
csharp_ |
JBoyer++ Dyrcona++ |
| 14:20 |
csharp_ |
jessie is dead, long live bullseye! |
| 14:22 |
* jvwoolf |
sings "When She Loved Me" and cries |
| 14:22 |
csharp_ |
jvwoolf++ |
| 14:25 |
JBoyer |
jvwoolf++ csharp_++ |
| 14:48 |
csharp_ |
calling 1299 - I think we can do it |
| 14:52 |
pinesol |
[evergreen|Mike Rylander] LP#1901597: Allow multi-subfield match in vandelay.strip_field - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=6aecb1a> |
| 14:52 |
pinesol |
[evergreen|Chris Sharp] LP#1901597: Stamp upgrade scripts - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=439b75d> |
| 14:56 |
pinesol |
[evergreen|Tiffany Little] LP1848575 Angular boolean filters should use Yes/No - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=5672931> |
| 15:04 |
pinesol |
[evergreen|Terran McCanna] LP1881012 Typo in staff/circ/hold/app.js - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=d84c5c7> |
| 15:06 |
pinesol |
[evergreen|Shula Link] LP1834258 Replace Javascript with Placeholder Attribute - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=6114ea3> |
| 15:13 |
terranm |
Is there a way in mkurl to strip out *all* existing URL parameters? I know how to strip out specific ones by putting them in square brackets |
| 15:16 |
pinesol |
[evergreen|Galen Charlton] LP#1944765: enable the Apache module mod_headers - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=d80b607> |
| 15:16 |
pinesol |
[evergreen|Chris Sharp] LP#1944765: add headers to bullseye install script - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=97bf745> |
| 15:22 |
csharp_ |
ha! claiming 1300!! |
| 15:22 |
mmorgan |
Woo hoo!! |
| 15:26 |
rhamby |
csharp I don't know if you have an interest but I've been testing Mike Risher's angular port for action triggers, I'm going to install it in a clean test but then ... |
| 15:26 |
pinesol |
Showing latest 5 of 9 commits to Evergreen... |
| 15:26 |
pinesol |
[evergreen|Bill Erickson] LP1919465 Relabel Item Location -> Shelving Location - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=994a429> |
| 15:26 |
pinesol |
[evergreen|Bill Erickson] LP1919465 Pull List Mark Discard/Weed Action - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=8d61829> |
| 15:26 |
pinesol |
[evergreen|Bill Erickson] LP1919465 Avoid dupe Current Item's in pull list - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=e5164b9> |
| 15:26 |
rhamby |
if it works I'll post a pull request. it's kind of a funky situation though since I'm not changing any of his code, not rebuilding the follow up patch which somethign weird git-wise happened with |
| 15:26 |
pinesol |
[evergreen|Bill Erickson] LP1919465 Pull list only shows holds with current_copy - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=70c18ba> |
| 15:26 |
pinesol |
[evergreen|Chris Sharp] LP1919465: stamp upgrade script - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=f685c2a> |
| 15:27 |
terranm |
rhamby++ |
| 15:28 |
csharp_ |
rhamby: I need a break, but I'll take a look. I do remember reviewing that some time ago and I don't remember what the problem was |
| 15:28 |
rhamby |
csharp_ fwiw it tested okay for you but when galen pulled it it failed to pull cleanly, which seems to be something weird with his commit not the code itself |
| 15:29 |
rhamby |
so I'm basically redoing his second commit though the code is the same |
| 15:29 |
gmcharlt |
looks like saving bibs is currently broken in master |
| 15:30 |
gmcharlt |
1298 refers to a function, evergreen.array_remove_item_by_value, that was removed in 1264 |
| 15:35 |
gmcharlt |
ok, easy enough to fix |
| 15:35 |
* gmcharlt |
claims 1301 |
| 15:37 |
gmcharlt |
I will also momentarily push a fix to the stamping of 1300 |
| 15:42 |
pinesol |
[evergreen|Galen Charlton] LP#1919465: fix issue with DB rev stamping - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=90547d6> |
| 15:42 |
pinesol |
[evergreen|Galen Charlton] LP#1864516: (follow-up) install current version of function - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=48686a9> |
| 16:02 |
|
terranm joined #evergreen |
| 16:10 |
csharp_ |
gmcharlt: thanks |
| 16:10 |
Bmagic |
I give up, anyone know how this is possible? https://pasteboard.co/Gaxb7VsXXYVP.png |
| 16:12 |
gmcharlt |
mismatch between IDL as passed via IDL2js and reality. The sledgehammer fix is ensure that /openils/conf and /openils/var/web/reporters have the current version of fm_IDL.xml; restart services; autogen; and reload Apache |
| 16:12 |
Bmagic |
The label comes from the text in the database "cust" config.usr_setting_type |
| 16:12 |
Bmagic |
ok, good Idea! let me double check the IDL file |
| 16:12 |
JBoyer |
Bmagic, yeah, lp 1937299 |
| 16:12 |
pinesol |
Launchpad bug 1937299 in Evergreen 3.6 "Multiple issues with the Collections: Exempt user setting" [Medium,Fix committed] https://launchpad.net/bugs/1937299 |
| 16:13 |
gmcharlt |
whoops, or that. don't listen to me |
| 16:13 |
JBoyer |
need to add a .label() to that setting because it's using the whole JS object as the label rather than the text. |
| 16:13 |
JBoyer |
gmcharlt++ # fixing the build |
| 16:15 |
Bmagic |
The issue doesn't appear to be on our bugsquashing machine (master) but shows itself in 3.7.1, so it would seem, it's this bug |
| 16:17 |
Bmagic |
JBoyer++ |
| 16:17 |
JBoyer |
Yeah, because the other issue it addresses is that the setting doesn't even exist if your db is new enough. ;) |
| 16:21 |
|
jvwoolf left #evergreen |
| 16:23 |
* JBoyer |
throws himself bodily toward the weekend, hopes for the best. |
| 16:23 |
* mmorgan |
hopes JBoyer |
| 16:23 |
mmorgan |
's aim is true |
| 16:35 |
gmcharlt |
for any commtter, I commend the OAI provider support (bug 1729620) patchset for consideration as a last-minute addition to 3.8 |
| 16:35 |
pinesol |
Launchpad bug 1729620 in Evergreen "Add optional oai2 opensrf module" [Wishlist,Confirmed] https://launchpad.net/bugs/1729620 |
| 16:37 |
gmcharlt |
note my comment #7 - I have some reservations, but I think it is (a) good enough to go in as an experimental feature, (b) a purely optoinal feature that is not enabled by default, and (c) doesn't have tendrils that would affect other parts of Evergreen |
| 16:48 |
rhamby |
https://bugs.launchpad.net/evergreen/+bug/1855780 has somne history of testing already, I don't know if it'll make it in 3.8 but here it is https://bugs.launchpad.net/evergreen/+bug/1855780 |
| 16:48 |
pinesol |
Launchpad bug 1855780 in Evergreen "Angular: Notifications/Action Triggers UI Port" [Wishlist,Confirmed] |
| 16:50 |
gmcharlt |
user/rogan/lp1855780_action_triggers_port_rebased? if so, please update the bug |
| 17:01 |
gmcharlt |
rhamby: it's not quite there yet, unless somebody else wants to spend time on it; e.g., atcol needs to be registered with prcrud for the edit enviornment modal to work |
| 17:08 |
|
mmorgan left #evergreen |
| 17:11 |
terranm |
corecommitters++ Y'all beat the record for most commits since we started doing bug squash weeks! |
| 18:01 |
pinesol |
News from qatests: Failed Installing Angular web client <http://testing.evergreen-ils.org/~live//archive/2021-09/2021-09-24_16:00:03/test.29.html> |
| 18:04 |
terranm |
FYI, I'm logging out, but I'll finish up the statistics tomorrow morning to include anything else that happens today |
| 18:20 |
|
rjackson_isl_hom joined #evergreen |
| 18:53 |
|
jihpringle joined #evergreen |
| 21:50 |
|
stephengwills left #evergreen |
