| Time |
Nick |
Message |
| 01:42 |
|
gsams joined #evergreen |
| 05:01 |
pinesol_green |
News from qatests: Test Failure <http://testing.evergreen-ils.org/~live> |
| 07:21 |
|
rjackson_isl joined #evergreen |
| 07:21 |
|
agoben joined #evergreen |
| 08:20 |
|
collum joined #evergreen |
| 08:42 |
|
kmlussier joined #evergreen |
| 08:42 |
|
mmorgan joined #evergreen |
| 09:17 |
|
yboston joined #evergreen |
| 09:21 |
|
Dyrcona joined #evergreen |
| 10:15 |
|
abowling joined #evergreen |
| 10:31 |
csharp |
@blame add $who is the SPY! |
| 10:31 |
pinesol_green |
csharp: The operation succeeded. Blame #25 added. |
| 10:32 |
Dyrcona |
@blame 25 Bmagic |
| 10:32 |
pinesol_green |
Dyrcona: Bmagic is the SPY! |
| 10:33 |
Dyrcona |
heh |
| 10:33 |
Bmagic |
nice! |
| 10:33 |
csharp |
@praise add $who LOVES the RESISTANCE! |
| 10:33 |
pinesol_green |
csharp: The operation succeeded. Praise #14 added. |
| 10:34 |
csharp |
hmm |
| 10:34 |
csharp |
open-ils.storage.actor.user.crazy_search: prepare_cached(SELECT evergreen.unaccent_and_squash(?)) statement handle DBIx::ContextualFetch::st=HASH(0x90c7f00) still Active at /usr/local/share/perl/5.18.2/OpenILS/Application/Storage/Publisher/actor.pm line 627. |
| 10:34 |
csharp |
not seen that before |
| 10:35 |
csharp |
triggered by searching for a patron by username in the web client and xul client |
| 10:35 |
Dyrcona |
unaccent_and_squash is new in 2.10, IIRC. |
| 10:35 |
Dyrcona |
So, you've got a stuck query with it? |
| 10:36 |
csharp |
yeah, looks like it |
| 10:36 |
csharp |
to the database! |
| 10:39 |
csharp |
yep, been running for > 20 mins |
| 10:40 |
Dyrcona |
Hmmm... Did they search for a blank name or something? |
| 10:40 |
berick |
select evergreen.unaccent_and_squash(next_largest_prime()) |
| 10:46 |
|
Stompro joined #evergreen |
| 10:47 |
Dyrcona |
heh |
| 10:48 |
jeff |
i am happy to find that CREATE SEQUENCE does not appear to accept a function as an argument for INCREMENT BY |
| 10:49 |
Dyrcona |
jeff: IDK, that could be a useful feature for randomizing sequences. |
| 10:49 |
Dyrcona |
Though, I guess using not a sequence for such a purpose would work. |
| 10:49 |
* jeff |
nods |
| 10:52 |
|
RBecker joined #evergreen |
| 10:53 |
jeff |
doesn't stop you from having a sequence based on prime numbers, you just have to write a handy function and do something like this: CREATE TEMPORARY TABLE evil (id INTEGER NOT NULL DEFAULT next_prime_or_so('evil_id'::text)); |
| 10:54 |
* Dyrcona |
isn't concerned about primes so much. |
| 10:54 |
Dyrcona |
It's consider a "good idea" for user ids to not be sequential, particularly if your interface is leaky. :) |
| 10:55 |
Dyrcona |
Unfortunately, I don't have any real advice for csharp at the moment. |
| 10:55 |
Dyrcona |
@blame 25 Dyrcona |
| 10:55 |
pinesol_green |
Dyrcona: Dyrcona is the SPY! |
| 10:59 |
Dyrcona |
BTW, the "particularly if your interface is leaky" was not aimed at Evergreen. |
| 10:59 |
|
Christineb joined #evergreen |
| 10:59 |
Bmagic |
another issue with 2.11 we are finding Uploading PO MARC throws an error |
| 10:59 |
Bmagic |
ERROR: new row for relation "lineitem" violates check constraint "picklist_or_po" |
| 10:59 |
Dyrcona |
I don't know of any where in the interface that we leak user ids via URL, which is what I had in mind. |
| 11:02 |
kmlussier |
Bmagic: Do you have any details on what options were selected when they were uploading the PO? |
| 11:03 |
* kmlussier |
can fire up a VM to test a PO upload. |
| 11:04 |
Bmagic |
acq.lineitem.create id= selector=201704 picklist= purchase_order= provider=18 create_time=now edit_time=now........ |
| 11:05 |
Bmagic |
shouldn't there be a purchase_order there? |
| 11:09 |
tsbere |
Bmagic: According to the DB it isn't required? |
| 11:10 |
Bmagic |
the constraint is called picklist_or_po |
| 11:11 |
Dyrcona |
Ah, bet you need either a picklist or a PO and you have neither. |
| 11:11 |
Bmagic |
ADD CONSTRAINT picklist_or_po CHECK (picklist IS NOT NULL OR purchase_order IS NOT NULL); |
| 11:11 |
tsbere |
Bmagic: I just said the purchase_order isn't required. You didn't ask about picklist. ;) |
| 11:11 |
Dyrcona |
Notice: picklist= purchase_order =, both are null in your example. |
| 11:14 |
kmlussier |
Bmagic: When uploading, did the user select the option to Create a Purchase Order or the option to add to a selection list? |
| 11:15 |
* kmlussier |
has never tried uploading records without selecting one of those options. |
| 11:15 |
Bmagic |
create purcahse order |
| 11:16 |
kmlussier |
I was able to successfully upload to a PO on master. However, I didn't get a success message for the upload. Just the progress bar. That hasn't happened to me in a while. |
| 11:26 |
|
brahmina joined #evergreen |
| 11:28 |
Bmagic |
this is interesting. The 901c is getting a non integer assignment. open-ils.cstore ERROR inserting vandelay::queued_bib_record |
| 11:28 |
Bmagic |
<datafield tag="901" ind1=" " ind2=" "><subfield code="c">Growing up, Kate Priddy was always a bit neurotic, experiencing momentary bouts of anxiety......" |
| 11:29 |
berick |
heh, wonder what sequence function generated that |
| 11:29 |
csharp |
Bmagic: that points to a problem with the file (off-by-one) - this may be relevant: bug 924952 |
| 11:29 |
pinesol_green |
Launchpad bug 924952 in Evergreen "Acquisitions: Order record loads fail when there is a null value in a holdings subfield" [High,Confirmed] https://launchpad.net/bugs/924952 |
| 11:29 |
csharp |
oh - maybe not |
| 11:29 |
kmlussier |
Oh yeah. I remember that one. |
| 11:29 |
Dyrcona |
berick: My non-linear quantum sequence generator....It just got entangled with the wrong particle. |
| 11:29 |
csharp |
that's just the holdings subvield |
| 11:29 |
csharp |
subfield, even |
| 11:30 |
Bmagic |
Looking back at the marc that I loaded, it had the 901c filled in with that. So there we go |
| 11:30 |
berick |
Dyrcona: hanging out on the wrong side of the tracks again |
| 11:31 |
* Dyrcona |
wonders if Google is using quantum computers.... ;) |
| 11:33 |
JBoyer |
Dyrcona, well, yes and no. |
| 11:33 |
Dyrcona |
Hah! |
| 11:33 |
JBoyer |
:D |
| 11:33 |
csharp |
JBoyer++ |
| 11:33 |
Dyrcona |
JBoyer++ |
| 11:33 |
mmorgan |
JBoyer++ |
| 11:34 |
csharp |
@quote add * Dyrcona wonders if Google is using quantum computers.... ;) < JBoyer> Dyrcona, well, yes and no. |
| 11:34 |
pinesol_green |
csharp: The operation succeeded. Quote #161 added. |
| 11:35 |
tsbere |
Does that collapse into a single answer once you go look at their computers? ;) |
| 11:38 |
mmorgan |
tsbere: What JBoyer said :) |
| 11:39 |
JBoyer |
Actually that one's a little more calculus. The answer approaches yes as time approaches infinity. ;) |
| 11:40 |
Dyrcona |
tsbere: Depends on which side of the brane you end up on. |
| 11:47 |
|
sandbergja joined #evergreen |
| 11:47 |
* mmorgan |
can't get a workstation to stick when trying to login to webby in Firefox. |
| 11:48 |
mmorgan |
I'm getting forced to register a workstation, and it looks like the registration is successful, but it continues to force me to register. |
| 11:48 |
mmorgan |
Works fine in chrome. |
| 11:49 |
mmorgan |
Is this expected/known? |
| 11:49 |
* mmorgan |
is happy that it's forcing a workstation, though! |
| 11:50 |
csharp |
mmorgan: I see the same thing in FF on Ubuntu |
| 11:51 |
csharp |
and works in Chrome for me too |
| 11:52 |
mmorgan |
Not sure if this might an example of bug 1546128 |
| 11:52 |
pinesol_green |
Launchpad bug 1546128 in Evergreen "eg.hatch.required not parsed correctly" [Undecided,New] https://launchpad.net/bugs/1546128 |
| 11:52 |
kmlussier |
mmorgan: I tested workstation registration the other day, but I can't recall if I tested FF. I would add a comment to the workstation registration bug. |
| 11:52 |
* kmlussier |
doesn't have the bug number at her fingertips. |
| 11:54 |
mmorgan |
This one? lp 1467663 |
| 11:54 |
pinesol_green |
Launchpad bug 1467663 in Evergreen "Cannot login to web staff client if work station does not exist in database" [Medium,Fix committed] https://launchpad.net/bugs/1467663 |
| 11:54 |
csharp |
"Firefox can’t establish a connection to the server at wss://localhost:8443/hatch." |
| 11:54 |
csharp |
^^ that's in the console |
| 11:55 |
csharp |
actually, this is in Chrome, too: |
| 11:55 |
csharp |
unable to connect to Hatch server at wss://localhost:8443/hatch |
| 11:55 |
berick |
csharp: you can ignore that |
| 11:56 |
csharp |
huh - it finally worked |
| 11:57 |
csharp |
looks like 'Use Now' might be the problem |
| 11:59 |
kmlussier |
I'm not seeing the same problem in Firefox. 'Use Now" kicks me back to the login screen, but my newly-registered workstation is now in the dropdown. |
| 11:59 |
mmorgan |
still not working for me. |
| 11:59 |
kmlussier |
My understanding is that's how the 'Use Now' button should be working now. |
| 11:59 |
csharp |
kmlussier: try in private browsing mode - that's where I saw it |
| 12:01 |
kmlussier |
It's working fine for me in private browsing mode too. |
| 12:01 |
Dyrcona |
csharp: Private browsing mode disables "tracking." I would not be surprised if that messed with things. |
| 12:01 |
kmlussier |
mmorgan: Are you in private browsing mode? |
| 12:02 |
csharp |
huh - it works for me now (even in private browsing) |
| 12:02 |
csharp |
I was using private browsing in an attempt to get around cache issues |
| 12:02 |
mmorgan |
kmlussier: No, not in private browsing mode. |
| 12:03 |
kmlussier |
mmorgan: Try clearing your cache and then re-test? |
| 12:03 |
|
krvmga joined #evergreen |
| 12:04 |
JBoyer |
Are you waiting long enough? I tried it out in my FF (49.something) private window, and it took a good while for the workstation dropdown to appear, but once it did I was able to signin using the new workstation I had just created. |
| 12:07 |
mmorgan |
JBoyer: Stared at it for a good long time, but didn't actually time it. Must've been at the login screen for at least a couple of minutes and the dropdown didn't show up. |
| 12:07 |
kmlussier |
Yeah, mine showed without a time delay. It was instantaneous. |
| 12:07 |
JBoyer |
Oh, that should have been plenty of time, yes. |
| 12:08 |
* mmorgan |
clears cache more aggressively |
| 12:09 |
JBoyer |
I just tried it again in FF 50.cough and it worked again. The only other thing I could think of is to open the inspector window and see if there's anything useful in there. :-/ |
| 12:10 |
mmorgan |
webby is veeeerrry slow for me atm. |
| 12:14 |
berick |
from hackaway... xul client deprecation warning at 3.0 around Sept 2017? have we picked a XUL client removal date/release yet? |
| 12:18 |
kmlussier |
berick: My recollection was that deprecated XUL would be available for two releases (3.0 and 3.1) and then removed in the next release, which I guess would be September 2018. |
| 12:19 |
berick |
thanks kmlussier |
| 12:19 |
kmlussier |
If it goes according to plan |
| 12:26 |
|
bmills joined #evergreen |
| 12:32 |
kmlussier |
Bah! I'm commenting on bugs under Evergreen Bug Maintenance account. |
| 12:32 |
csharp |
heh |
| 12:33 |
bshum |
Oops |
| 12:36 |
kmlussier |
Yeah, I think you all should take the keys to that account away from me and give them to somebody else who can better handle the responsibility. ;) |
| 12:37 |
csharp |
@intervention kmlussier |
| 12:37 |
pinesol_green |
csharp: Sorry, that command is only available to Evergreen Premium™ Subscribers. Please upgrade your subscription ASAP! |
| 13:33 |
Bmagic |
Importing marc acq records works bus the resulting line items in the purchase order do not link to the catalog. The import settings had a match profile and "import non-matching records" checked. All of the 8 records in my test do not match, but they are not imported. |
| 13:34 |
Bmagic |
I take that same file and use on batch import, and it works fine. |
| 13:39 |
kmlussier |
Bmagic: Has the order been activated yet? |
| 13:40 |
Bmagic |
no |
| 13:54 |
Bmagic |
hmmmm Caught error from 'run' method: Can't locate object method "content" via package "OpenSRF::DomainObject::oilsMethodException" at /usr/local/share/perl/5.22.1/OpenILS/Application/Acq/Order.pm line 526 |
| 13:54 |
Bmagic |
I think I'm onto something |
| 14:08 |
kmlussier |
Bmagic: As I mentioned earlier, when I load records, I'm getting a hanging progress bar, but the PO is created with lineitems. In my case, the lineitems are not linked to the catalog either. |
| 14:09 |
Bmagic |
ah, interesting |
| 14:09 |
kmlussier |
I saw the same behavior on master and on the community demo server, which I loaded with the 2.10 release branch yesterday, so it's pretty much 2.10.7 with a few additional branches that have recently been merged. |
| 14:10 |
kmlussier |
However, when testing on a 2.10.5 server, the upload performed as expected. The non-matching records were imported. |
| 14:18 |
kmlussier |
I get an error message that says: Caught error from 'run' method: Can't locate object method "max_chunk_count" via package "OpenSRF::AppRequest" at /usr/local/share/perl/5.18.2/OpenILS/Application/Vandelay.pm line 904. |
| 14:19 |
Bmagic |
hmmm, different |
| 14:20 |
Bmagic |
we are on 16.04 which is why we installed OpenSRF from master |
| 14:20 |
Bmagic |
which could* be playing a role in this |
| 14:21 |
dbwells |
Bmagic: looking that above errors, sounds like a good guess to me |
| 14:21 |
Bmagic |
dbwells: I'm running through installation on another server to test it against 2.4.1 tarball |
| 14:21 |
kmlussier |
tsbere could confirm, but I'm guessing my VMs are installing OpenSRF from master too. |
| 14:34 |
|
collum joined #evergreen |
| 14:34 |
dbwells |
kmlussier: looks like "chunking" was renamed to "bundling" in 01f95834835bed94 in OpenSRF. It seems like references to "max_chunk_count" in Vandelay may need to move to "max_bundle_count" to work with OpenSRF master. |
| 14:34 |
pinesol_green |
dbwells: [opensrf|Mike Rylander] LP#1612771: bundling and chunking - <http://git.evergreen-ils.org/?p=OpenSRF.git;a=commit;h=01f9583> |
| 14:35 |
dbwells |
IOW, I don't think we're quite ready for OpenSRF master yet on the EG side. |
| 14:36 |
Bmagic |
confirmed - Running OpenSRF master is causing the issue for us |
| 14:37 |
Bmagic |
so, now, is there anyone here running OpenSRF 2.4.1 on Ubuntu 16.04? |
| 14:37 |
dbwells |
Bmagic: just a hunch, but I see that alt_holds_print.js references "chunk_size", so maybe new OpenSRF was the cause of that issue as well? |
| 14:37 |
Bmagic |
it sure could be! |
| 14:38 |
Bmagic |
Which explains why it was working on my test machine and not on production |
| 14:38 |
Bmagic |
my test machine was Ubuntu 14.04 and therefore it could run 2.4.1, so I didnt have a need to run OpenSRF master |
| 14:38 |
kmlussier |
dbwells++ # Good sleuthing! |
| 15:02 |
|
mmorgan1 joined #evergreen |
| 15:20 |
|
bmills joined #evergreen |
| 15:37 |
tsbere |
kmlussier: Your VMs install OpenSRF from master by default. You can, of course, override that when building if you want to, the same basic way you tell it what Evergreen to install from. |
| 15:37 |
kmlussier |
tsbere: Thanks! I think I will override it for the community demo server. |
| 15:40 |
Dyrcona |
Bmagic: I've not tried 2.4.1 on Ubuntu 16.04, but the main changes are in prerequisite installation. |
| 15:41 |
Dyrcona |
Since you've already installed the prerequisites for 16.04, you could probably just install 2.4.1 and skip that step. |
| 15:42 |
pinesol_green |
[evergreen|Jason Stephenson] LP 1639250: Install Excel::Writer::XLSX from deb on Wheezy. - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=6db854d> |
| 15:44 |
kmlussier |
Dyrcona++ phasefx++ |
| 15:44 |
Dyrcona |
Oh, nifty! :) |
| 15:44 |
Bmagic |
Dyrcona: that is what I ended up doing and it seems like it's working ::crosses fingers:: |
| 15:46 |
Dyrcona |
Bmagic: I just fired up my 16.04 VM and I have also not updated OpenSRF since mid August, before the chunking changes went in. |
| 15:46 |
Dyrcona |
Bmagic: It should work. I don't see why it wouldn't. |
| 15:47 |
Dyrcona |
I guess a trickier option would be to revert those commits locally. |
| 15:47 |
Dyrcona |
Bmagic++ |
| 15:48 |
berick |
lol @ the irony of installing 1,396MB of stuff to compile asciidoc |
| 15:48 |
berick |
@love asciidoc |
| 15:48 |
pinesol_green |
berick: The operation succeeded. berick loves asciidoc. |
| 15:49 |
Dyrcona |
Yep, the changes were all in the Makefiles and README. |
| 15:50 |
Dyrcona |
berick: You can thank the TeXlive distribution for that "bloat." But, I *think* that's how asciidoc makes PDFs by converting to TeX and then generating PDF from that. |
| 15:50 |
* phasefx |
thinks folks should try asciidoc with action/trigger and print notices :) |
| 15:51 |
Dyrcona |
phasefx: You joke, but that is doable. |
| 15:51 |
phasefx |
Dyrcona: I yearn, you mean |
| 15:51 |
Dyrcona |
How it would work in practice.... That's another question. |
| 15:51 |
Dyrcona |
hah! |
| 15:51 |
* phasefx |
poked at it before, tripped over cover sheets and moved on |
| 15:52 |
phasefx |
no idea how fun setting margins and other specific placements would be, though |
| 15:52 |
berick |
phasefx: i love it |
| 15:52 |
Dyrcona |
Me, neither. I've not tried that with asciidoc. |
| 15:53 |
phasefx |
berick: for print notices? |
| 15:53 |
* phasefx |
loves asciidoc too, for text specs |
| 15:53 |
phasefx |
tech, even |
| 15:54 |
berick |
phasefx: yeah, for notices. i think that's a neat idea |
| 15:55 |
phasefx |
berick: ah, I misunderstood your meaning. Thought you've done it and love it :D :D |
| 15:55 |
Dyrcona |
TeX gives you excellent control over text placement in exchange for nearly unreadable markup. ;) |
| 15:56 |
phasefx |
could make folks use A/T and postscript :D |
| 15:56 |
Dyrcona |
That's going too far! ;) |
| 15:57 |
JBoyer |
To read your library notice please download and install a postscript interpreter from one of these fine sources: etc. |
| 15:58 |
JBoyer |
Damn, that font rendering, though. |
| 15:58 |
phasefx |
[% picture('Hello World') | png %] |
| 16:32 |
|
mmorgan joined #evergreen |
| 16:35 |
gmcharlt |
ack, sorry about that |
| 16:36 |
* jeff |
raises an eyebrow |
| 16:36 |
gmcharlt |
you'll see in a moment |
| 16:36 |
pinesol_green |
[evergreen|Galen Charlton] webstaff: add template for patron merge dialog - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=afbda31> |
| 16:36 |
pinesol_green |
[evergreen|Galen Charlton] Revert "webstaff: add template for patron merge dialog" - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=ad733ec> |
| 16:39 |
kmlussier |
heh |
| 16:39 |
jeff |
ah. :-) |
| 16:39 |
jeff |
gmcharlt++ |
| 16:40 |
Stompro |
Can someone give me a tip on the best way to grab the icon_format for a bib in an SQL query? I've been trying to use metabib.record_attr_flat, but I'm seeing multiple entries for icon_format there for a single bib. |
| 16:40 |
tsbere |
Stompro: Bibs can have more than one icon |
| 16:40 |
gmcharlt |
well, multiple entries is legit |
| 16:41 |
Stompro |
tsbere, gmcharlt thanks, that explains it then. |
| 16:42 |
Stompro |
I'll try and weed out the ones I don't want to see in the query. |
| 16:49 |
tsbere |
Stompro: In non-reporter queries I tend to string_agg them together, just so you know. |
| 16:50 |
* tsbere |
doesn't usually use the flat view though, as there shouldn't ever be uncontrolled versions of composites |
| 16:53 |
|
BigRig joined #evergreen |
| 17:01 |
pinesol_green |
News from qatests: Test Success <http://testing.evergreen-ils.org/~live> |
| 17:03 |
kmlussier |
Huzzah! |
| 17:05 |
|
mmorgan left #evergreen |
| 17:51 |
Bmagic |
Stompro: select string_agg(value,$$,$$) from metabib.record_attr_flat where attr=$$icon_format$$ and id=$bibid |
| 18:45 |
|
Callender_ joined #evergreen |
| 18:46 |
|
cbush06 joined #evergreen |
| 18:53 |
|
Callender joined #evergreen |
| 18:57 |
|
kmlussier joined #evergreen |
