| Time |
Nick |
Message |
| 07:27 |
|
rjackson_isl_hom joined #evergreen |
| 07:37 |
|
collum joined #evergreen |
| 07:53 |
|
BDorsey joined #evergreen |
| 08:06 |
|
rfrasur joined #evergreen |
| 08:10 |
|
kworstell-isl joined #evergreen |
| 08:21 |
|
mantis1 joined #evergreen |
| 08:37 |
|
mmorgan joined #evergreen |
| 08:57 |
|
Dyrcona joined #evergreen |
| 09:17 |
|
jvwoolf joined #evergreen |
| 09:39 |
|
kworstell-isl joined #evergreen |
| 10:14 |
Dyrcona |
Anyone ever passed an array to a psql query using placeholders? |
| 10:17 |
Dyrcona |
Well, I just did, and looks like the "obvious" thing works: psql -U evergreen -h localhost -p 35433 -f lookup2.sql -v expire=2020-09-01 -v orgs="{45,51}" |
| 10:19 |
csharp_ |
miker: the Shibboleth development came after OA stuff was in place here, so we never tested it against it |
| 10:32 |
jmurray-isl |
practice.evergreen.lib.in.us |
| 10:33 |
jmurray-isl |
Oops. Wrong window. |
| 10:39 |
JBoyer |
Dyrcona, I'd expect that to work, depending on how the variable is referenced in the sql (:var vs :'var' or :"var", I don't recall which you need for arrays. :) ) |
| 10:40 |
Dyrcona |
JBoyer: I used :'orgs' |
| 10:40 |
Dyrcona |
I usually do that for strings, and arrays seem to work as strings. |
| 10:41 |
JBoyer |
Obviously couldn't be :"var" I guess. And yeah, makes sense treating arrays like strings, postgres doesn't treat {} like a quoting operator. |
| 10:42 |
Dyrcona |
Otherwise, you have to double quote the argument on the command line like -v orgs="'{45,51}'" or -v date="'2020-07-01'" |
| 10:42 |
JBoyer |
+1 |
| 10:43 |
JBoyer |
That's why I keep meaning (and failing) to put in an LP about changing up the upgrade scripts going forward to use :'var' instead of :var. |
| 10:43 |
Dyrcona |
I thought of that once, too. |
| 10:47 |
Dyrcona |
I think I needed the double quotes in my original to hide {} from the shell. Even if I didn't, it doesn't hurt. :) |
| 11:00 |
JBoyer |
Oh, yeah. one set on the cmdline is just good hygiene. :) |
| 11:00 |
JBoyer |
sandbergja++ Just saw the lint branch was pushed |
| 11:01 |
Dyrcona |
sandbergja++ |
| 11:05 |
pinesol |
News from commits: LP#1959048: add release notes <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=01f87025e11c16bd94e4bb5ed8bb4a29bbec7080> |
| 11:05 |
pinesol |
News from commits: LP1959048: manual ng lint fixes <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=c1faaf054fcc7c28e1037ef8b620f817f3853fe5> |
| 11:05 |
pinesol |
News from commits: LP1959048: Run ng lint --fix <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=58238af879b5b4b361d824eb3013b3918e4b05c3> |
| 11:05 |
pinesol |
News from commits: LP1959048: run ng g @angular-eslint/schematics:convert-tslint-to-eslint <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=01c5f2336110b3cea8c003befb7507f810959921> |
| 12:02 |
|
jihpringle joined #evergreen |
| 12:05 |
miker |
csharp_: "working code wins" :) gotcha |
| 12:07 |
Dyrcona |
For certain definitions of "working." |
| 12:23 |
|
rjackson_isl_hom joined #evergreen |
| 12:47 |
csharp_ |
miker++ |
| 12:47 |
csharp_ |
assuming OA stuff gets committed, maybe a wishlist bug to dedupe what we can? |
| 13:18 |
miker |
csharp_: sure, that's sane. and certainly, I don't know if there is OA stuff (that is needed in your instance) that can't be done via SAML. if that's the case then we'd have to keep both, obv. if not ... there's a lot less to configure for Shib-based SAML SSO, so less code and maintenance! |
| 13:30 |
JBoyer |
Given that we're up to 3-4 different auth types now (native db, ldap, SAML, and potentially OA), maybe it's time to consider a more modular auth plugin system? So you could hook up whatever to your system provided the .so exposed the necessary interface. Could also allow adding new auth types without waiting for a release or rebuild. |
| 13:32 |
JBoyer |
(before someone wants to add OpenIDConnect or some other new option.) |
| 14:36 |
|
Stompro joined #evergreen |
| 16:18 |
|
jvwoolf left #evergreen |
| 17:07 |
|
mmorgan left #evergreen |
