| Time |
Nick |
Message |
| 06:51 |
|
jamesrf joined #evergreen |
| 06:55 |
|
JBoyer joined #evergreen |
| 07:02 |
|
agoben joined #evergreen |
| 07:07 |
|
rjackson_isl joined #evergreen |
| 07:34 |
|
bdljohn joined #evergreen |
| 08:16 |
|
bos20k joined #evergreen |
| 08:36 |
|
Dyrcona joined #evergreen |
| 08:44 |
|
mmorgan joined #evergreen |
| 09:13 |
JBoyer |
Dyrcona, If you have a couple minutes for Overdrive API questions I have a couple. (Mostly what needed to be requested for CW/MARS' setup, not low level stuff) |
| 09:14 |
Dyrcona |
JBoyer: Go ahead. |
| 09:15 |
JBoyer |
I was looking at the APIs we have access to (LIB, META, AVAIL, and SRCH) and I have an updated and hand-tested api secret, but no part of the Ebook API displays anything in the logs. Did you request additional APIs or have to do anything special to get things working there? |
| 09:16 |
JBoyer |
(hand-tested as in I can throw the secret at curl and get an oauth bearer token back, I haven't done much else) |
| 09:17 |
Dyrcona |
Yes. Let me check. |
| 09:20 |
Dyrcona |
You need to have Client and Patron authenticaton API access. |
| 09:21 |
Dyrcona |
BTW, where do you see the list of APIs that you have access to? I'm not seeing that in the member center on the developers' site. |
| 09:22 |
JBoyer |
Try running this with your info: curl --basic -d grant_type=client_credentials -u CLIENTNAME:CLIENTSECRET https://oauth.overdrive.com/token |
| 09:23 |
JBoyer |
gives your token an API access |
| 09:23 |
JBoyer |
an->and. |
| 09:24 |
|
yboston joined #evergreen |
| 09:28 |
JBoyer |
Actually, if you could run that and throw the list in here that would be helpful since I'm sure I'm going to have to go back and forth with them about what we need. |
| 09:28 |
Dyrcona |
It's the same as yours. |
| 09:28 |
JBoyer |
That's annoying. |
| 09:29 |
Dyrcona |
I don't think the circulation APIs show up. |
| 09:29 |
JBoyer |
Maybe other endpoints say or do different things. I guess I'll see what happens with this latest request. |
| 09:29 |
JBoyer |
Thanks for the info! |
| 09:29 |
JBoyer |
Dyrcona++ |
| 09:33 |
|
krvmga joined #evergreen |
| 09:33 |
miker |
csharp: you're right, it was gmcharlt that fixed that function (I remember having the conversation with him, mis-remembered writing that and took credit) http://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=85efd3b40bc3d53d7d004b8bd02a89bd7ceeaa1f |
| 09:33 |
krvmga |
it's been so long since i've done it, i can't remember where to configure which languages appear in the advanced search language box |
| 09:33 |
pinesol |
miker: [evergreen|gmc] do not use TRUNCATE when refreshing reporter.materialized_simple_record - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=85efd3b> |
| 09:35 |
JBoyer |
Now that I look a little closer I suspect expires_in:0 is probably not ideal. |
| 09:38 |
|
jvwoolf1 joined #evergreen |
| 09:39 |
Dyrcona |
JBoyer: My response said expires_in:3600. Zero may not mean immediately. It may mean never. |
| 09:41 |
JBoyer |
Possibly, but when I try some of their other endpoints (such as https://api.overdrive.com/v1/libraries/{LibID}) it gives me an auth failure. |
| 09:42 |
|
jvwoolf joined #evergreen |
| 09:42 |
JBoyer |
The bearer token also being 870 characters long seems unusual but I don't touch OAuth much. |
| 09:45 |
|
terran joined #evergreen |
| 09:46 |
csharp |
miker: cool |
| 09:47 |
csharp |
I just changed it locally and all's good |
| 09:48 |
Dyrcona |
charp|miker: I was gonna bug that truncate, but could one of you do it? I'm in the middle of something else at the moment. |
| 09:48 |
Dyrcona |
grr. csharp ^ |
| 09:53 |
Dyrcona |
krvmga: I'm pretty sure the list comes from config.coded_value_map where ctyp = 'item_lang' in the database. |
| 09:53 |
* csharp |
creates bug 1811689 |
| 09:53 |
pinesol |
Launchpad bug 1811689 in Evergreen "Use TRUNCATE, not DELETE which refreshing reporter.materialized_simple_record" [Undecided,New] https://launchpad.net/bugs/1811689 |
| 09:53 |
Dyrcona |
s/ctyp/ctype/ |
| 09:53 |
Dyrcona |
csharp++ |
| 09:54 |
Dyrcona |
It's too cold in this office to type properly. They shouldn't turn the heat off over the weekends. |
| 09:58 |
* csharp |
pictures Dyrcona hunched over desk like Bob Cratchitt |
| 09:59 |
Dyrcona |
:) |
| 10:00 |
Dyrcona |
krvmga: If you're looking to disable any languages in particular, we can turn them off by setting opac_visible to false. |
| 10:00 |
|
collum joined #evergreen |
| 10:02 |
krvmga |
Dyrcona, for some reason i thought there were more than 22 language options available |
| 10:02 |
* Dyrcona |
should get some gloves with the finger tips cut off to look like Bob Cratchitt. |
| 10:05 |
|
jvwoolf left #evergreen |
| 10:05 |
* mmorgan |
has a pair of those gloves. |
| 10:06 |
* Dyrcona |
has some gloves that are too big to type with, or do much of anything else with. |
| 10:13 |
Dyrcona |
Would anyone else be interested in a Lp bug to add Hebrew to the list of languages in config.coded_value_map? Looks like we're going to add it locally. |
| 10:29 |
|
sandbergja joined #evergreen |
| 10:49 |
|
yboston joined #evergreen |
| 11:02 |
|
yboston joined #evergreen |
| 11:04 |
|
bdljohn joined #evergreen |
| 11:06 |
|
jvwoolf joined #evergreen |
| 11:25 |
|
yboston joined #evergreen |
| 11:33 |
|
bdljohn1 joined #evergreen |
| 11:42 |
gmcharlt |
dbwells: are you actively working on bug 1712854? |
| 11:42 |
pinesol |
Launchpad bug 1712854 in Evergreen 3.1 "web client: Sorting for view record holds and holds shelf interfaces" [High,Confirmed] https://launchpad.net/bugs/1712854 - Assigned to Dan Wells (dbw2) |
| 11:49 |
|
sandbergja joined #evergreen |
| 11:55 |
dbwells |
gmcharlt: No, I am not. I don't quite recall why I assigned myself, but I think it was related to some release-time bug wrangling. I am now removed. |
| 11:57 |
|
aabbee joined #evergreen |
| 12:04 |
|
beanjammin joined #evergreen |
| 12:07 |
|
jvwoolf joined #evergreen |
| 12:12 |
|
jihpringle joined #evergreen |
| 12:20 |
|
Christineb joined #evergreen |
| 12:30 |
|
mmorgan joined #evergreen |
| 12:37 |
|
nfburton joined #evergreen |
| 12:37 |
|
jamesrf joined #evergreen |
| 12:39 |
nfburton |
Does anyone have experience setting up Playaways as a format? |
| 12:41 |
Bmagic |
we do |
| 12:42 |
Bmagic |
here is an example: |
| 12:42 |
Bmagic |
missourievergreen.org/eg/opac/record/607734 |
| 12:42 |
Bmagic |
http://missourievergreen.org/eg/opac/record/607734 |
| 12:43 |
nfburton |
Oh that's exactly what I want! Do I require a reingest for the label to show? I've been trying to create my own coded value map for the OPAC Format Icon. Could you share what the specifics for that are? |
| 12:44 |
nfburton |
We are setting up a new vendor for them and this has been wanted for a while now |
| 12:44 |
nfburton |
But I can't seem to get the codes just right |
| 12:45 |
Bmagic |
yeah, we defined the MARC bits. We ended up with a combination of two 007's |
| 12:46 |
nfburton |
I see NCCardinal did that too. I've been trying to replicate from their documentation but must be missing something |
| 12:47 |
Bmagic |
Since it's not officially recognized in LOC, it's the wild west, and you/catalogers can decide how to catalog playaways. Our catalogers decided it needed to have two 007's and some other stuff. Then you define that in the Evergreen ILS, call it playaway. Then create an icon for it with the same name |
| 12:48 |
Bmagic |
the icon/picture file needs to be located in the same folder as the rest of the picture files. The naming convention needs to match the name of the format (but all lowercase) as defined in the ILS |
| 12:49 |
Bmagic |
and finally.. reingest |
| 12:49 |
Bmagic |
you need only reingest a test record for troubleshooting before you go and reingest the whole database |
| 12:53 |
nfburton |
Would I be able to see your tree for the Format Icon? |
| 12:54 |
Bmagic |
yeah, let me see if I can get a screenshot |
| 12:54 |
nfburton |
Thanks |
| 12:56 |
Bmagic |
https://imgur.com/a/9KhoJzS and https://imgur.com/a/p76HC9q |
| 13:01 |
nfburton |
Do the SMD indications exist or were those created for this? |
| 13:06 |
Bmagic |
I really dont remember |
| 13:07 |
Bmagic |
I may have needed to create those definitions... Probably did it in the database if I did |
| 13:08 |
nfburton |
Okay |
| 13:08 |
nfburton |
Thanks for the direction |
| 13:08 |
nfburton |
Bmagic++ |
| 13:09 |
Bmagic |
no prob |
| 13:11 |
Bmagic |
I've gotta run, but I'll dig into the config.* tables and see if I can find anything else helpful along these lines |
| 13:12 |
nfburton |
Sounds good thanks |
| 13:18 |
|
yboston joined #evergreen |
| 14:31 |
|
jvwoolf joined #evergreen |
| 14:36 |
csharp |
khuckins_: we were just testing your fix for bug 1777677 and hit a snag - if I'm a non- |
| 14:36 |
pinesol |
Launchpad bug 1777677 in Evergreen "Test notification method" [Wishlist,New] https://launchpad.net/bugs/1777677 |
| 14:37 |
csharp |
"admin" user, I am prompted with a perm override box for OPAC_LOGIN even though my user has that perm at the proper level |
| 14:37 |
csharp |
gonna see if I can get that working, but wanted you to know (I also update the bug report) |
| 14:38 |
csharp |
s/update/updated/ |
| 14:50 |
Bmagic |
nfburton: it looks like the SMD stuff is stock in config.marc21_physical_characteristic_type_map |
| 15:02 |
jeff |
csharp: was the user you were testing as a user with super_user = true set? |
| 15:02 |
csharp |
I think I may have a handle on the problem. It's possible that the "home_ou" attribute is not being dereferenced "enough" (obviously not conversant in Perl to this level) |
| 15:02 |
jeff |
csharp: can you see what arguments were being passed to open-ils.actor.event.test_notification? |
| 15:02 |
csharp |
jeff: the original testers were superusers and that works perfectly |
| 15:03 |
jeff |
and once that's fixed, there's some other permissions-related flaws which would need to be addressed before that's tested and merged. |
| 15:03 |
csharp |
but a non-superuser with OPAC_LOGIN set still doesn't see it |
| 15:03 |
csharp |
http://git.evergreen-ils.org/?p=working/Evergreen.git;a=blob;f=Open-ILS/src/perlmods/lib/OpenILS/Application/Actor.pm;h=f8aa7491bf1ff8d097229c484a26201b1a84c1b4;hb=refs/heads/user/khuckins/lp1777677-test-notification-method#l4231 |
| 15:04 |
csharp |
that's the sub it's using and I think $$args{home_ou}) on line 4241 is the problem |
| 15:04 |
jeff |
it's because a user with super_user = true (which should probably be deprecated) gets TRUE returned for permission checks on non-existent org unit contexts. |
| 15:04 |
csharp |
I'm not great at using Data::Dumper, but it comes back as Fieldmapper::actor::org_unit=ARRAY(0x5b40070) |
| 15:05 |
csharp |
ah - makes sense |
| 15:05 |
jeff |
can you see in your browser what arguments are being passed? |
| 15:05 |
csharp |
er... |
| 15:05 |
jeff |
oh. |
| 15:06 |
csharp |
not showing the args in the console, no |
| 15:06 |
jeff |
is this in place on a test system i can log into, or would i need to install the branch to get that? |
| 15:06 |
JBoyer |
I would wonder how well it works if that line is just thrown away. I barely remember when a perms check came up for that and the question was basically "What's something that everybody has?" which to me sounds a lot like |
| 15:07 |
JBoyer |
"Maybe we don't need to gate this one" |
| 15:07 |
csharp |
jeff: let me fix the SSL cert on it and I'll let you it (it's a concerto server) |
| 15:07 |
csharp |
s/let you it/let you in/ |
| 15:07 |
Dyrcona |
chsharp: You are correct if $$args{home_ou} is an object, you need the id field. |
| 15:07 |
JBoyer |
(Though it is also helpful to know why things aren't working; I'm not trying to dissuade either of you from figuring that part out_) |
| 15:08 |
Dyrcona |
bleh... |
| 15:08 |
jeff |
JBoyer: yeah, i think you're alluding to what i was alluding to. :-) |
| 15:09 |
Dyrcona |
If that line is thrown away, the perm check will happen at the workstation ou of the user. |
| 15:09 |
* jeff |
was attempting to find/parse the original requirements and the bug comments before declaring it a glaring hole |
| 15:10 |
csharp |
jeff: https://csharp-master.gapines.org and I used Scott Brock's login: https://wiki.evergreen-ils.org/doku.php?id=qa:concerto_logins |
| 15:11 |
Dyrcona |
Erm, I meant if the $args->{home_ou} bit were dropped. I'm nor sure why we'd check for OPAC_LOGIN. |
| 15:11 |
csharp |
(BR1) |
| 15:12 |
Dyrcona |
csharp: If you change line 4241 to have $$args{home_ou}->id instead of just $$args{home_ou} that check should work. |
| 15:12 |
JBoyer |
jeff++ # due diligence where I occasionally charge blindly ahead. |
| 15:13 |
csharp |
Dyrcona: I'll give it a shot |
| 15:13 |
Dyrcona |
My recommendation would be to drop that line and the preceding line. |
| 15:13 |
Dyrcona |
But, I've not looked at the specifications or the bug. :) |
| 15:13 |
|
khuckins joined #evergreen |
| 15:13 |
khuckins |
home_ou is passed in for the sake of checking the permission, so if we drop the OPAC_LOGIN check, that should probably be dropped as well |
| 15:14 |
khuckins |
csharp++ jeff++ Drycona++ JBoyer++ |
| 15:14 |
Dyrcona |
khuckins: Any reason for checking for OPAC_LOGIN? |
| 15:15 |
jeff |
Just to be clear, the status of this is that it is not currently included in any released or reployed Evergreen install, other than perhaps a test install with either no live data or no ability for non-trusted individuals to log in? :-) |
| 15:16 |
* Dyrcona |
thinks its on our training server where we may have looked at it, but yeah. :) |
| 15:16 |
khuckins |
Early on on the lp ticket it was suggested to have a permissions check to avoid potential abuse, initially using the UPDATE_USER permission, then brought down to OPAC_LOGIN when realizing users should be using this as well as staff |
| 15:16 |
khuckins |
But in retrospect any user who's going to be able to access that API call would be logged in |
| 15:17 |
Dyrcona |
I was just about to ask if it was accessible without being logged in. |
| 15:17 |
jeff |
the issue where you get a perm check in the staff interface is likely due to service.send_test_message needing to use patron.home_ou.id instead of patron.home_ou. |
| 15:17 |
jeff |
the issue works with a superuser account, but would fail with most/all other accounts. |
| 15:19 |
|
terran joined #evergreen |
| 15:19 |
Dyrcona |
CStoreEditor->allowed could be made smarter, i.e. to check if it got an object or an int and then derefence the id if got an object. Some other places make similar checks. |
| 15:20 |
jeff |
and if my quick scan of the requirements is correct, you're going to want to checkauth, but then instead of just checking for OPAC_LOGIN or skipping a perm check, you'll want to verify that the user is the same as the target, OR if the user is different from the target, that the user has an appropriate staff permission at the right level. |
| 15:21 |
Dyrcona |
An "appropriate permission"... VIEW_USER perhaps? |
| 15:21 |
jeff |
the trouble with the code as currently written is that it permits any valid logged in user to send test notifications to anyone, just by supplying a value for "target" |
| 15:21 |
jeff |
Dyrcona: originally it was UPDATE_USER |
| 15:21 |
|
aabbee left #evergreen |
| 15:21 |
|
aabbee joined #evergreen |
| 15:22 |
Dyrcona |
Well, that makes sense if the test is done in conjunction with changing the notification method's value. |
| 15:22 |
Dyrcona |
jeff++ |
| 15:23 |
jeff |
Also, it allows a wider than intended array of events to be fired. |
| 15:24 |
jeff |
by passing arbitrary values as event_def_type -- though it looks like it will only pass an (arbitrary!) user object to $U->fire_object_event |
| 15:27 |
csharp |
jeff++ |
| 15:32 |
|
bdljohn joined #evergreen |
| 15:36 |
|
jamesrf joined #evergreen |
| 16:02 |
|
khuckins_ joined #evergreen |
| 16:03 |
|
khuckins__ joined #evergreen |
| 16:04 |
|
Glen__ joined #evergreen |
| 16:27 |
|
nfburton joined #evergreen |
| 17:10 |
|
jvwoolf left #evergreen |
| 17:13 |
|
mmorgan left #evergreen |
| 17:13 |
Bmagic |
@later tell nfburton the source code shows the DB inserts for the SMD stuff. 950.data.seed-values.sql lines (roughly) 6238 - 6400 |
| 17:13 |
pinesol |
Bmagic: The operation succeeded. |
| 17:17 |
|
khuckins joined #evergreen |
| 17:35 |
|
khuckins joined #evergreen |
| 21:08 |
|
ddale joined #evergreen |
| 22:20 |
|
beanjammin joined #evergreen |
| 22:57 |
|
beanjammin joined #evergreen |
