| Time |
Nick |
Message |
| 01:42 |
|
Jillianne joined #evergreen |
| 04:31 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 06:40 |
|
rlefaive joined #evergreen |
| 07:19 |
|
agoben joined #evergreen |
| 07:24 |
|
rjackson_isl joined #evergreen |
| 08:28 |
|
kmlussier joined #evergreen |
| 08:43 |
kmlussier |
@coffee [someone] |
| 08:43 |
* pinesol_green |
brews and pours a cup of El Salvador Nombre de Dios, and sends it sliding down the bar to stompro |
| 08:43 |
kmlussier |
@tea [someone] |
| 08:43 |
* pinesol_green |
brews and pours a pot of Earl Grey Creme Black Tea, and sends it sliding down the bar to pinesol_green (http://ratetea.com/tea/teavana/earl-grey-creme/1257/) |
| 08:47 |
|
collum joined #evergreen |
| 08:48 |
kmlussier |
Happy Friday! Looks like my email and web site are down. :( |
| 08:49 |
* kmlussier |
believes it's a sign that she should start her weekend early. |
| 08:49 |
dbs |
kmlussier: I concur! |
| 08:50 |
|
mmorgan joined #evergreen |
| 08:57 |
|
Dyrcona joined #evergreen |
| 08:58 |
|
rlefaive_ joined #evergreen |
| 08:59 |
|
mmorgan1 joined #evergreen |
| 09:08 |
dbs |
Should we add --with-websockets-port=443 to the default OpenSRF config install instructions? I think proxying is the way of the future, no? |
| 09:26 |
|
yboston joined #evergreen |
| 09:29 |
Dyrcona |
Think I'll play some Bob Dylan, since I seem to be talkin' to myself in a monologue. |
| 09:29 |
Dyrcona |
dbs: I don't really have a strong opinion on your question. |
| 09:33 |
dbs |
Dyrcona: hey I was listening to A Hard Rain is Gonna Fall on the way into work this morning. Because a hard rain *was* falling |
| 09:33 |
|
jvwoolf joined #evergreen |
| 09:33 |
Dyrcona |
Heh. |
| 09:33 |
dbs |
At some point we should make opensrf.math respond with "42" for everything. |
| 09:33 |
Dyrcona |
I'm listening to newer Dylan, from teh '90s. |
| 09:34 |
Dyrcona |
:) |
| 09:34 |
Dyrcona |
maybe for opensr.math.life ? |
| 09:34 |
Dyrcona |
bleh... missed the f. |
| 09:39 |
dbs |
Or maybe on Towel Day |
| 09:40 |
Dyrcona |
I think it would be funny to add opensrf.math.life_the_universe_and_everything as a hidden request. :) |
| 09:40 |
dbs |
+1 |
| 09:42 |
Dyrcona |
So, we should submit the bug and branch on April 1st and commit it on May 25. |
| 09:53 |
dbs |
kmlussier: on Bug 1704396 it sounds very much like a cstore connection is dying on you :/ |
| 09:53 |
pinesol_green |
Launchpad bug 1704396 in Evergreen "Slowness for metecord and one-hit searches in 2.12" [High,New] https://launchpad.net/bugs/1704396 |
| 10:00 |
|
mmorgan joined #evergreen |
| 10:05 |
kmlussier |
I just replicated it on the C/W MARS training server too. Dyrcona: That's on 2.12, right? |
| 10:05 |
Dyrcona |
Yes, on 2.12.3 plus customizations. |
| 10:06 |
Dyrcona |
I've seen issues with ISBN searches before, but it seemed randomish and I never tracked it down. |
| 10:08 |
kmlussier |
Dyrcona: Yeah, the one I replicated was with the metarecord searches. Those are a little easier to make happen. Were the ISBN issues you saw prior to 2.12? |
| 10:09 |
Dyrcona |
kmlussier: I'm not certain. It may have been on test servers. I'm not sure I've seen it in production on 2.10, still. |
| 10:20 |
Dyrcona |
So, when are we moving search out of the database and using a real search engine? |
| 10:20 |
* Dyrcona |
ducks the flying duck decoys. :) |
| 10:25 |
dbs |
Also on nginx proxying maybe we rely on nginx to do all of the TLS work and just proxy Apache on port 80? |
| 10:26 |
dbs |
Dyrcona: yeah, sigh :/ |
| 10:26 |
Dyrcona |
I've not really looked into proxying, yet, but think I'll have to soonish, like in the fall. |
| 10:27 |
Dyrcona |
I'm looking at options to possibly replace the load balancer, but I might not be able to do that. |
| 10:54 |
* csharp |
has to start proxing very soon for our Labor Day 2.12 upgrade |
| 10:55 |
csharp |
proxying, even |
| 10:55 |
csharp |
however, I haven't had luck getting it working on a standalone yet - pretty sure it's cert issues |
| 10:55 |
* csharp |
hasn't looked in a while |
| 10:58 |
* dbs |
will have to fire up the shiny and inara PWA dev instances again to copy & paste from his past smarter self |
| 11:42 |
|
maryj joined #evergreen |
| 11:46 |
dbs |
~search_path |
| 11:46 |
dbs |
@search_path |
| 11:46 |
pinesol_green |
dbs: I'm sorry, Dave. I'm afraid I can't do that. |
| 11:47 |
dbs |
*sigh* |
| 12:05 |
|
jihpringle joined #evergreen |
| 12:21 |
* dbs |
imagines "Recalculating bib fingerprints; this may take a while" will indeed take quite a while with 2 million bibs |
| 12:24 |
|
rlefaive joined #evergreen |
| 13:14 |
Dyrcona |
jeffdavis: For the overdive integration, does the subfield n of the 037 have to match the beginning of the URL in the 856 $u? |
| 13:16 |
jeffdavis |
No, if there's an ID in the 037 it should use that regardless of what's in the 856. |
| 13:16 |
kmlussier |
I noticed a couple of bugs for Bug Squashing Week are xul fixes. I know xul will be deprecated in 3.0, but we'll still be providing support for xul in 2.11 and 2.12 |
| 13:16 |
jeffdavis |
With more recent records from Overdrive the URL won't contain a usable ID. |
| 13:16 |
* dbs |
wonders how http://search.cpan.org/dist/JSON/ is not an Ubuntu Xenial package |
| 13:16 |
Dyrcona |
OK. The $n in 037 in our Overdrive records has http://www.overdrive.com |
| 13:17 |
Dyrcona |
The id is in $a |
| 13:17 |
kmlussier |
I'm wondering how far we should be merging xul fixes as they happen. Is there a certain release where we'll be cutting them off? Or a particular date? |
| 13:17 |
Dyrcona |
dbs: On 17.04: i libjson-perl - module for manipulating JSON-formatted dat |
| 13:18 |
jeffdavis |
Dyrcona: yeah, I was just doublechecking the subfields. I think we have some older Overdrive records with an 037 but no ID in $a. |
| 13:19 |
Dyrcona |
dbs: It's a package on Xenial, but I think we maybe installing JSON::XS. |
| 13:19 |
dbs |
Dyrcona: yeah, found https://packages.ubuntu.com/xenial/libjson-perl but JSON was installed by CPAN during plain old make -f ubuntu-xenial |
| 13:19 |
Dyrcona |
Was it? |
| 13:19 |
dbs |
https://packages.ubuntu.com/xenial/libjson-perl is there too |
| 13:19 |
dbs |
err https://packages.ubuntu.com/xenial/libjson-xs-perl |
| 13:19 |
dbs |
stupid chrome browser disabling initial highlight for middle-click copy/paste |
| 13:20 |
Dyrcona |
Yeah.. Guess that's a hangover from earlier. |
| 13:20 |
Dyrcona |
I see the packages, too. |
| 13:20 |
* dbs |
runs to a meeting but will try to remember to bug that, and that the upgrade docs point at 2.12.1 instead of 2.12.3 |
| 13:20 |
Dyrcona |
jeffdavis: Thanks. I'm trying to figure out what our users are not seeing checked out and on hold items. |
| 13:20 |
jeffdavis |
Dyrcona: Specifically it will try to pull an ID from 037$a if and only if 037$b contains "OverDrive" - if both conditions aren't met, it will try to parse URLs in the 856 that match your base_uri settings from config.tt2 |
| 13:21 |
Dyrcona |
jeffdavis: OK. That's good to know. Is the OverDrive part case sensitive? |
| 13:21 |
jonadab |
kmlussier: Or a particular version number, perhaps. |
| 13:21 |
Dyrcona |
And, ours says OverDrive, Inc. |
| 13:22 |
jeffdavis |
IF marc037_source.match('OverDrive') |
| 13:22 |
jeffdavis |
so, case sensitive but "OverDrive, Inc." should be a match |
| 13:22 |
Dyrcona |
OK! Thanks! I think we may need to set the password required setting to true. |
| 13:24 |
jeffdavis |
Also, if it goes with the 856 URL, it will just grab whatever comes after "ContentDetails.htm?ID=" in the first URL that matches your base_uri settings and use that as the ID. AFAIK that consistently matches the format of URLs in older OverDrive records, but perhaps you've got different records. |
| 13:24 |
jeffdavis |
any improvements are welcome :) |
| 13:25 |
Dyrcona |
Cool. You read my mind! :) |
| 13:25 |
Dyrcona |
I was gonna ask about the URI scanning. |
| 13:26 |
Dyrcona |
So, from looking at this record, it should be working, and the OPAC correctly shows 0 of 1 available. |
| 13:26 |
Dyrcona |
The user who has it checked out, doesn't see it in items out. |
| 13:27 |
kmlussier |
jonadab: Yeah, I don't know that I'm framing my question correctly. |
| 13:27 |
Dyrcona |
jeffdavis++ |
| 13:29 |
Dyrcona |
What does the ebook_api.ebook_test.enabled setting do? |
| 13:29 |
Dyrcona |
Oh never mind. I think I see what it does. |
| 13:29 |
Dyrcona |
It enables a test provider. |
| 13:30 |
jeffdavis |
Yeah, same one the live tests use. |
| 13:32 |
jeffdavis |
If you load the concerto dataset and do an OPAC search for "tolkien" with the test provider enabled, you should see it at work. |
| 13:32 |
Dyrcona |
Well, we've got it mostly working, so far. Are the users check outs and holds supposed to be showing up? |
| 13:33 |
jeffdavis |
Yeah, transactions should be displayed in separate "E-Items Checked Out" / "E-Items on Hold" tabs in My Account |
| 13:33 |
jeffdavis |
actually performing checkouts and holds from the OPAC should be possible in 3.0, I have a working branch for that to be pushed today actually |
| 13:34 |
Dyrcona |
If we require a login to check something out at OverDrive, then it should require the user password to see what is checked out and on hold, right? |
| 13:35 |
* Dyrcona |
decides to look at the code and see how the settings are used. |
| 13:35 |
jeffdavis |
Yes. It should grab the password that the user enters on login (which I hate, but there it is). |
| 13:35 |
Dyrcona |
Thanks. This has been most helpful. I think I need to change a setting. |
| 13:42 |
|
rlefaive joined #evergreen |
| 13:56 |
|
Jillianne joined #evergreen |
| 14:08 |
stompro |
dbs++, thanks for pointing out the other nowrap fix. |
| 14:37 |
|
jvwoolf joined #evergreen |
| 15:23 |
jeffdavis |
I've shared a working branch for bug 1673870. I'll be away for the next couple of weeks, but if anyone wants to test what's there, please feel free. |
| 15:23 |
pinesol_green |
Launchpad bug 1673870 in Evergreen "Add support for ebook API transactions in OPAC (OverDrive/OneClickdigital)" [Undecided,New] https://launchpad.net/bugs/1673870 - Assigned to Jeff Davis (jdavis-sitka) |
| 15:24 |
kmlussier |
jeffdavis++ |
| 15:24 |
kmlussier |
That's exciting! |
| 15:24 |
JBoyer |
jeffdavis++ |
| 15:25 |
* JBoyer |
needs to look into what we need to do to get all of that setup... |
| 15:25 |
JBoyer |
turned on, rather. |
| 15:29 |
jeffdavis |
There's some documentation in section 5.7.4 of the release notes for 2.12: https://evergreen-ils.org/documentation/release/RELEASE_NOTES_2_12.html |
| 15:30 |
|
rlefaive joined #evergreen |
| 15:39 |
JBoyer |
Oh yeah, one of the questions I had was what kind of API keys to request from OD since they appear to have multiple "levels." But if OPAC checkout is coming in 3.0 the answer is "the one that can do everything!" :) |
| 15:44 |
jeffdavis |
You want client and patron authorization. EG doesn't support granted auth yet. |
| 15:44 |
jeffdavis |
The API uses their discovery and circulation APIs in addition to those authentication endpoints. |
| 15:45 |
jeffdavis |
I don't recall if you need to request discovery/circ access separately - I think you get them both when you have patron auth. |
| 15:59 |
JBoyer |
jeffdavis++ |
| 15:59 |
JBoyer |
Something to look into in a couple weeks when I have time again. |
| 16:31 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 17:03 |
|
mmorgan left #evergreen |
| 17:09 |
|
jvwoolf left #evergreen |
| 18:53 |
|
Bmagic joined #evergreen |
| 19:27 |
dbs |
*sigh* ERROR: index row size 2760 exceeds maximum 2712 for index "browse_entry_sort_value_value_key" on upgrade to 2.12 presumably because of our extra-specially long entries in a few records |
| 19:27 |
dbs |
guess I could slap a substr() on it like we do for mfr - just sucks that right now that rolls back the whole upgrade |
| 19:27 |
dbs |
pingest to the rescure |
| 19:28 |
dbs |
ERROR: index row size 2760 exceeds maximum 2712 for index "browse_entry_sort_value_value_key" too, for good measure |
| 20:01 |
jeffdavis |
ugh, yeah, I've run into that before :( |
