| Time |
Nick |
Message |
| 00:38 |
|
jeff__ joined #evergreen |
| 00:44 |
|
rjackson-isl joined #evergreen |
| 02:35 |
|
dbwells_ joined #evergreen |
| 02:35 |
|
remingtron_ joined #evergreen |
| 07:35 |
|
rjackson-isl joined #evergreen |
| 07:57 |
|
Dyrcona joined #evergreen |
| 08:00 |
|
collum joined #evergreen |
| 08:28 |
|
mrpeters joined #evergreen |
| 08:35 |
Dyrcona |
Here we go: Testing a void_all_bills fix in the billing branch. |
| 08:38 |
|
Shae joined #evergreen |
| 08:39 |
mrpeters |
morning all -- trying to load some bib record SQL generated by paralell_pg_loader.pl and getting this error -- http://pastie.org/8755529 -- not sure which file's "line 31" it's balking at... |
| 08:45 |
|
mmorgan1 joined #evergreen |
| 08:45 |
|
mmorgan1 left #evergreen |
| 08:46 |
Dyrcona |
Could not load <x-jsd:source?location=oils%3A%2F%2Fremote%2Fxul%2Frel_3_0_20140220_184016%2Fserver%2FOpenILS%2Fdata.js&instance=68> |
| 08:47 |
Dyrcona |
That can't be good, but it only happens when running with Venkman. |
| 08:51 |
mrpeters |
here's a little more from postgresql.log Re: the error loading bibs -- http://pastie.org/8755570 -- still doesn't quite clue me in to where this errors out, exactly |
| 08:53 |
|
mmorgan joined #evergreen |
| 08:54 |
|
ericar joined #evergreen |
| 08:54 |
Dyrcona |
mrpeters: I've seen someone (jl-?) post that can't add attribute error before or very similar. I'd say search the IRC logs for the past couple of weeks, maybe they got an answer. |
| 08:56 |
Dyrcona |
On my "could not load" error: The file is sitting right there, too, on the server. |
| 08:57 |
Dyrcona |
Same ownership and permissions as the other files. |
| 08:57 |
|
akilsdonk joined #evergreen |
| 09:01 |
mrpeters |
Dyrcona: yeah, not turning up anything in the google search sadly |
| 09:03 |
Dyrcona |
mrpeters: Well, I'm afraid I won't be much help. I've not actually seen it before myself. |
| 09:04 |
mrpeters |
no worries, i havent either and ive done hundreds of migrations hah |
| 09:04 |
mrpeters |
the load of the bre.sql runs for several hours, and then dumps with that....so i figure it's just a bad record somewhere along the line |
| 09:05 |
|
timf joined #evergreen |
| 09:06 |
Dyrcona |
It looks like something going on a xsl transform if that helps. I think it might have something to do with ingest. |
| 09:07 |
mrpeters |
that would make sense, with the timeline at which it occurs during the load |
| 09:10 |
Dyrcona |
You could turn the ingest off while you do the load. I forget the actual global flag off the top of my head. |
| 09:10 |
Dyrcona |
Then, you could try ingesting the records individually to see which one croaks. |
| 09:10 |
Dyrcona |
It will take a while if you have a lot of records, though. |
| 09:11 |
|
senator joined #evergreen |
| 09:13 |
Dyrcona |
Man, I really dislike working in the staff client. |
| 09:14 |
mrpeters |
hah, yeah that would take mega time Dyrcona, but its a thought |
| 09:14 |
Dyrcona |
Well, you could write a sql or perl script that outputs the id before doing each ingest call. |
| 09:15 |
Dyrcona |
Capture that to a file and have a look a day later to see where the problem records are. |
| 09:16 |
Dyrcona |
I wanna know why changing a field name used to determine a map filter on a list causes other calls to lose the session id of the logged in user. |
| 09:16 |
Dyrcona |
Venkman is just getting in my way. |
| 09:17 |
Dyrcona |
hmm.. maybe money.billing.void is my actual problem and the staff client is masking that from me. |
| 09:17 |
Dyrcona |
but, I swear that was working before this little change. |
| 09:19 |
* Dyrcona |
hangs his head in shame. |
| 09:19 |
jeff |
change in the IDL resulting in something that was formerly an int being represented as an object can break previous assumptions about the data. i'd expect that to result in a noticeable error in the opensrf logs, though. |
| 09:19 |
Dyrcona |
jeff: It is simpler than that, I think. |
| 09:20 |
jeff |
or might not error at the server level, but would give data that your xul/js level wasn't expecting, etc. |
| 09:20 |
Dyrcona |
it is money.billing.void.... |
| 09:20 |
jeff |
Dyrcona: simple is good. :-) |
| 09:20 |
Dyrcona |
PEBKAC/typo-like |
| 09:21 |
tsbere |
I suppose I should say something to get the @later I don't need to read |
| 09:25 |
Dyrcona |
heh. |
| 09:25 |
Dyrcona |
And, it works! |
| 09:26 |
|
jboyer-laptaupe joined #evergreen |
| 09:26 |
Dyrcona |
jeff: You are correct about me changing field mapper entries, and those were contributing to the void_all_bills issues, but I fixed them earlier this morning. |
| 09:26 |
Dyrcona |
Thanks for the suggestion, though. It's good to be in the logs somewhere. :) |
| 09:27 |
Dyrcona |
I think I'll leave these as separate commits when I bring them from my dev branch to the working branch, since 1 commit hits the client and the other Circ/Money.pm. |
| 09:35 |
|
jwoodard joined #evergreen |
| 09:38 |
Dyrcona |
dbwells and to anyone who cares: I've resolved the void_all_bills issue AFAICT, I've pushed some fixes to the collab branch. |
| 09:38 |
Dyrcona |
Be warned that I rebased it last night to resolve a conflict with master when a collections function was removed. |
| 09:39 |
phasefx |
qa tests were angrier this morning than usual: http://testing.evergreen-ils.org/~live/test.html Need to add a hueristic for the db init section: http://testing.evergreen-ils.org/~live/test.17.html syntax error |
| 09:39 |
phasefx |
this is with the stripe code |
| 09:42 |
|
adbowling-isl joined #evergreen |
| 09:44 |
remingtron |
phasefx: I'll push a fixed version to my signoff branch |
| 09:44 |
dbs |
hmm. shouldn't that be caught if you run "eg_db_config.pl --create-schema --create-database" (with an optional "| grep ERROR") before committing? |
| 09:44 |
|
Wyuli joined #evergreen |
| 09:45 |
|
akilsdonk_ joined #evergreen |
| 09:45 |
remingtron |
dbs: yes, but I sadly neglected the fresh db install from the branch, only did the upgrade script on a master db |
| 09:45 |
* remingtron |
is learning his lesson |
| 09:45 |
dbs |
remingtron: we all have much to learn, don't worry! |
| 09:46 |
phasefx |
learning++ tests++ |
| 10:00 |
pinesol_green |
[evergreen|Remington Steed] Fix syntax error in seed data - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=0bb4915> |
| 10:08 |
|
dluch joined #evergreen |
| 10:10 |
* mrpeters |
attempts loading the records in 10k chunks (split FTW) to see if i can narrow it down any |
| 10:11 |
|
akilsdonk joined #evergreen |
| 10:12 |
dbs |
mrpeters: bonus is that all the chunks that do load successfully won't need to be reimported |
| 10:13 |
mrpeters |
dbs++ |
| 10:13 |
mrpeters |
ill just keep breaking it down until i find the record i suppose |
| 10:14 |
mrpeters |
this has been some tricky data to work with |
| 10:18 |
remingtron |
Dyrcona: thanks for the fix. one more issue I found is in the print and email receipt templates in seed data. |
| 10:18 |
remingtron |
IF mb.voided == 'f' |
| 10:19 |
Dyrcona |
Recipts.... OK. I'll have a look. Guess we should add some information to the release notes, too, that people will need to check their receipt templates. |
| 10:19 |
|
kbeswick joined #evergreen |
| 10:28 |
Dyrcona |
Ah. action trigger email receipt... |
| 10:30 |
|
dluch joined #evergreen |
| 10:34 |
remingtron |
right, sorry wasn't clear about that |
| 10:38 |
Dyrcona |
Interestingly, the template is treating the money.billing as a fleshed object, but the call that fires the event doesn't explicitly flesh money.billing, so I wonder what is going on there. |
| 10:44 |
Dyrcona |
Ah, that's right. I remember now. |
| 10:53 |
* dbs |
added the eg_db_config.pl check to http://wiki.evergreen-ils.org/doku.php?id=dev:signoff_review_checklist |
| 11:01 |
|
mcooper joined #evergreen |
| 11:01 |
Dyrcona |
So, do the changes go in yet another upgrade script or should I mush in with the schema changes? |
| 11:02 |
Dyrcona |
Hmm.. YAUS, because someone might have customized their templates. |
| 12:01 |
|
zerick joined #evergreen |
| 12:13 |
|
jboyer_laptaupe joined #evergreen |
| 12:14 |
|
jihpringle joined #evergreen |
| 12:16 |
pastebot |
"csharp" at 64.57.241.14 pasted "No AppSession object returned from server_build()" (1 line) at http://paste.evergreen-ils.org/23 |
| 12:17 |
|
Alias_Fakename joined #evergreen |
| 12:17 |
csharp |
^^so what is happening when that error occurs? This error's time correlates with a library attempting a keyword search and ending with an Internal Server Error message |
| 12:18 |
csharp |
I'd like to dig deeper, but no threadtrace... |
| 12:19 |
Alias_Fakename |
So I have a question that may be easy for my fellow evergreeners. We have a couple new people here and every time they check in an item evergreen wants to send it in transit to the branch they are sitting in. I thought the problem was working location, but that didn't seem to solve anything. |
| 12:19 |
Alias_Fakename |
Anyone else had this problemo? |
| 12:19 |
csharp |
oh - wait - I found one - more soon |
| 12:20 |
csharp |
Alias_Fakename: probably a transit that needs manually resolving |
| 12:20 |
Alias_Fakename |
But it happens on every checkin |
| 12:20 |
Alias_Fakename |
With multiple books and employees |
| 12:21 |
csharp |
Alias_Fakename: e.g., UPDATE action.transit_copy SET recv_time = now() WHERE id = <transit id>; |
| 12:21 |
Dyrcona |
there is a capture holds as transits feature. That is probably turned on. |
| 12:21 |
csharp |
or what Dyrcona says ;-) |
| 12:21 |
Alias_Fakename |
Where is that tickbox located? There were no holds on the items that I could see but I'll give it a try. |
| 12:22 |
Dyrcona |
Look on the Checkin Modifiers popup to the bottom right of the checkin screen. |
| 12:23 |
Alias_Fakename |
Ok thanks. |
| 12:24 |
pastebot |
"csharp" at 64.57.241.14 pasted "further log messages - same error" (2 lines) at http://paste.evergreen-ils.org/24 |
| 12:26 |
Alias_Fakename |
Nope, nothing was checked in any of those boxes |
| 12:26 |
csharp |
Alias_Fakename: do you have database access? |
| 12:27 |
|
yboston joined #evergreen |
| 12:27 |
Alias_Fakename |
Not right now, and even so I'm pretty sure it's not problem that requires it for fixing. I just can't find where the fix is! lol |
| 12:28 |
csharp |
Alias_Fakename: suit yourself ;-) |
| 12:40 |
csharp |
so I see the relevant code in OpenSRF, so what circumstances would put OpenSRF in that state? |
| 12:42 |
csharp |
http://git.evergreen-ils.org/?p=OpenSRF.git;a=blob;f=src/perl/lib/OpenSRF/AppSession.pm;h=9e371c7a71f38874a6fa5bc80882cd9934435e7f;hb=HEAD#l123 |
| 12:47 |
* csharp |
shrugs and gives up |
| 12:48 |
Alias_Fakename |
were you talking to me csharp? I apologize if so |
| 12:52 |
|
senator joined #evergreen |
| 12:54 |
csharp |
Alias_Fakename: nope - no problem - I was hoping someone with OpenSRF chops could assist |
| 12:55 |
Alias_Fakename |
Ah, ok |
| 13:10 |
eby |
jeff: MeL question if around? |
| 13:10 |
jeff |
eby: sure thing! |
| 13:10 |
eby |
jeff: you doing ncip to them or double booking manually? |
| 13:11 |
eby |
can pm if that is better ;) |
| 13:11 |
jeff |
We are using NCIP with MeLCat, though "to them" is perhaps inaccurate. |
| 13:11 |
Dyrcona |
Heh. It's more they're doing it to you, eh? :) |
| 13:12 |
Dyrcona |
more [like]... |
| 13:12 |
jeff |
we send extracts with bib and item data. patron lookup and hold placement and checkouts and checkins and such is handled by the MeLCat system sending NCIP messages to a CGI running on our Evergreen system. |
| 13:12 |
jeff |
Staff use the MeLCat DCB Client to process materials. |
| 13:14 |
eby |
so staff mediated still |
| 13:14 |
jeff |
yes. MeLCat items do not, cannot self check. |
| 13:15 |
jeff |
they're also not on the public holdshelf. |
| 13:15 |
jeff |
and i think that "not on the public hold shelf" thing is possibly a MeLCat policy. |
| 13:15 |
jeff |
not certain, though. |
| 13:15 |
eby |
yeah that is same with iii |
| 13:15 |
eby |
was just curious if updating status and stuff manually on your own holdings |
| 13:15 |
jeff |
Nope, NCIP takes care of that. |
| 13:17 |
jeff |
so, when we process our items to send out, when processed in the DCB client an NCIP message is sent from MeLCat servers to TADL's Evergreen instance, and the NCIP responder handles things like checking the outgoing item out to an account representing "this is out on loan to a melcat library" |
| 13:17 |
jeff |
(that account is similar to what we use for standard interlibrary loan to other libraries, and is treated specially for stats reporting, etc.) |
| 13:18 |
jeff |
we have the option of checking the items out to an account per-library, but there hasn't been a lot of interest in that) |
| 13:20 |
eby |
thanks. was curious if they got the ncip working before you went live with it or not |
| 13:20 |
|
smyers_ joined #evergreen |
| 13:22 |
jeff |
renewals need to be done in MyMeLCat, and then the renewal date is updated in Evergreen. |
| 13:22 |
jeff |
that's pretty much it. |
| 13:22 |
jeff |
there are a number of quirks, but those are the major points. |
| 13:23 |
jeff |
and yes, we did not go live without NCIP -- that was a pretty firm requirement on our end. |
| 13:25 |
jeff |
we took the NCIP responder that MCLS had contracted and did some cleanup and modification with the help of atz, Dyrcona, and dbwells: https://github.com/jonscott/iNCIPit is the repo (dbwells and j_scott have commit access, i think) and https://github.com/tadl/iNCIPit is TADL's fork. |
| 13:25 |
* Dyrcona |
has commit access to, IIRC. |
| 13:25 |
Dyrcona |
too |
| 13:25 |
jeff |
I believe at present we are running this branch in production: https://github.com/tadl/iNCIPit/tree/handle_invalid_xml (with a pull request for the parent repo here: https://github.com/jonscott/iNCIPit/pull/7 ) |
| 13:26 |
jeff |
there was a bit of work involved in cleaning things up. this was the large pull request: https://github.com/jonscott/iNCIPit/pull/4 |
| 13:28 |
bshum |
Hmm, I wasn't watching the conversation earlier but with Alias_fakename, but if he's getting transit on every checkin, I wonder if maybe they just have a workstation registered to the wrong location for that computer. Like they think they're X, but they're registered as Y. |
| 13:31 |
|
hbrennan joined #evergreen |
| 13:32 |
dbs |
bshum++ # happens to our libraries all the time |
| 13:32 |
bshum |
Someone asked me to calculate our general uptime from the last year. |
| 13:33 |
bshum |
Stats are such lies. |
| 13:33 |
bshum |
It's all about how you frame the question and answer... |
| 13:33 |
Dyrcona |
Well, "there's lies, damned lies, and there's statistics." |
| 13:34 |
dbs |
Right. I mentioned we had 99.9% uptime to our IT Services people and they guffawed (they miscalculated, but that's a different matter) until they realized that I was including maintenance downtime in that figure |
| 13:34 |
dbs |
"Oh, if we schedule downtime for maintenance we don't count that as downtime" |
| 13:34 |
dbs |
WHAT? |
| 13:34 |
dbs |
:) |
| 13:35 |
bshum |
Yeah I was just roughing that even if we're down for say, 5 days (like full 24 hours total added up), then that's still only like 1.3% of a year. Or something like that. |
| 13:35 |
bshum |
And we weren't down for even that much with unexpected + scheduled down |
| 13:36 |
bshum |
I don't know, it just makes me laugh :) |
| 13:39 |
Dyrcona |
"Quantify your impact on the organization." |
| 13:40 |
Dyrcona |
Well, you're able to do your job, so I must be doing mine. |
| 13:42 |
jcamins |
Dyrcona: multiply the number of circs by, say, 1.25*10^3 and say "as a result of my work keeping the catalog up, more than X tons of books were circulated from member libraries." |
| 13:42 |
jcamins |
That's assuming an average weight of ~10oz. |
| 13:43 |
jeff |
You circ sixteen tons, and whaddya get... |
| 13:43 |
* jeff |
. o ( sixteen tomes? ) |
| 13:44 |
|
Bmagic joined #evergreen |
| 13:45 |
|
smyers__ joined #evergreen |
| 13:49 |
|
davidcandlestick joined #evergreen |
| 13:54 |
berick |
dbwells: fyi, pushed another small patch to collab/miker/mvf_cra_metarecords-squash-signoff-rebase. I'll stop now. |
| 13:59 |
hbrennan |
I need to give a shout out to eeevil++ and phasefx++ for solving a problem with checkbox stickiness AND creating the patch for it yesterday! |
| 14:00 |
hbrennan |
I appreciate developers who understand the frustrations with a little thing like that |
| 14:03 |
berick |
jeff: hey, i'm working on a proof of concept java print/storage service using jetty, just to make sure some key assumptions are true. |
| 14:03 |
berick |
mainly that communication from the browser works fine and that jetty is a good tool (hint: yes to both) |
| 14:04 |
jeff |
where do you see jetty fitting in? locally? |
| 14:04 |
jeff |
that is, jetty on the client/workstation? |
| 14:04 |
berick |
jeff: yeah, locally, as the web service running on the PC |
| 14:06 |
berick |
before I get into anything meaty, i'd love to hear your thoughts on the project (any aspect of it) |
| 14:06 |
berick |
since I know you had some interest there |
| 14:08 |
berick |
e.g. browser xmlttprequest -> localhost:8080/foobar?baz -> jetty -> java does whatever -> responds to browser |
| 14:09 |
jeff |
yeah. it has some advantage over some other options, like browser downloads file with set mime type specifying a print job, browser fires off app to print. |
| 14:09 |
berick |
i've confirmed that cross-origin XHR to localhost works just fine |
| 14:09 |
berick |
and that java is perfectly willing to print (spawn dialogs, etc.) from the jetty handler |
| 14:10 |
jeff |
neither jetty nor the special-mime-type option have good support for chromebook/ipad/etc, but just because one option doesn't fit all doesn't mean it should be disqualified. |
| 14:10 |
phasefx |
hbrennan: to be clear, I didn't create the patch, just applied it to your system; it was bshum who fixed the mistake |
| 14:10 |
jeff |
i.e., there might be "jetty for these clients", "something-else for these-other-clients" |
| 14:10 |
* dbs |
might be concerned about Java's rather horrible security track record, but that's mostly as a browser plugin |
| 14:11 |
dbwells |
berick: I had been wondering about cross-origin XHR to localhost, glad to hear it works |
| 14:11 |
jeff |
well, it does add the requirement of "JRE needs to be installed and kept up to date"... |
| 14:11 |
dbs |
And until we get to bcrypt / salted passwords, we can't really talk |
| 14:11 |
berick |
jeff: yes, my assumption is this would not be running on mobile devices (nor would it be required). they would just have less flexible (or more manual) print options |
| 14:11 |
hbrennan |
phasefx: Ohhhh. Well, as if bshum++ needs another boost. :) Still, it was like magic. |
| 14:12 |
jeff |
i'd like them to have the same flexibility, just via different conduit (which might be seen as limiting flexibility), but that could be pie-in-the-sky. |
| 14:12 |
jeff |
printing aside, storage... for fixating workstation ID? |
| 14:12 |
jeff |
anything else? |
| 14:12 |
berick |
offline xacts |
| 14:12 |
jeff |
got it. desire not to use localstorage there is... too fragile, or something else? |
| 14:13 |
jeff |
sorry if it sounds like i'm just firing back with "why don't you use X" -- hopefully I'm coming across as brainstorming / inquiring, not criticizing. :-) |
| 14:14 |
berick |
localstorage is just easy enough to manually clear, possibly unententionally |
| 14:14 |
berick |
though i doubt we'd be hitting the 4M mark (or whatever the size limit is), but there is a size limit to localstorage as well |
| 14:15 |
berick |
if it's in a shared space, different browsers can access it (fwiw) |
| 14:15 |
berick |
jeff: you are coming across as brainstorming :) |
| 14:15 |
jeff |
another option i had considered with regard to workstation is that of client-side ssl certs. pros/cons there as well, but has some interesting side benefits. |
| 14:16 |
berick |
sounds familiar, but I admit I never fully grasped the big picture with that |
| 14:18 |
berick |
i'm guessing there will be other types of data we want to store on the filesystem as well. i know the xul client puts a bunch of stuff in the profile dir. don't recall what all it is, atm, though. |
| 14:19 |
dbs |
Lots of local preferences IIRC |
| 14:20 |
dbs |
Or state (e.g. last picked New MARC template, etc) |
| 14:20 |
berick |
jeff: the print via mime approach.. that would still require staff to select a printer or use a default, right? the beauty of java approach is that we can send to different printers automatically, which is a requirement. |
| 14:20 |
berick |
dbs: ah, right, probably lots of persist stuff |
| 14:21 |
jeff |
no, the print via mime would still have a locally-present helper app that knows how to print to different printers, with different options, etc. it's just a slightly different means of getting there. |
| 14:22 |
jeff |
one is localhost xhr to jetty listening port, the other is mime type causes browser to hand off to helper app |
| 14:22 |
|
smyers_ joined #evergreen |
| 14:22 |
berick |
ohhhh |
| 14:22 |
berick |
gotcha |
| 14:24 |
jeff |
argh. another quirk due to overdues being future-dated. :P |
| 14:25 |
jeff |
(unrelated to current conversation) |
| 14:28 |
jeff |
berick: i like proofs of concept. i'd be interested in seeing/hearing more. i'd also advocate that the jetty option not be the only option, or at least not be a hard requirement. |
| 14:29 |
berick |
definitly not a hard requirement |
| 14:30 |
berick |
i imagine a general purpose JS api within the client for storing / printing and when the jetty service is there, use it, if not use localstorage / window.print() / etc |
| 14:32 |
berick |
for printing, i find myslef leaning towards client-fleshed templates. by that, i mean, we don't want to make a server call (e.g. to action/trigger) every time we want to print. |
| 14:32 |
berick |
unless the server call can be insanely fast |
| 14:32 |
berick |
we used A/T for printing in the self-check and it's.. not fast |
| 14:34 |
berick |
the idea behind client-fleshed print data being that you have the data in hand already, so you can easily insert the data into an angular template. |
| 14:34 |
berick |
but i could be crazy. |
| 14:35 |
berick |
my grasp of the Evergreen printing requiremetns is right at that "a little bit of knowledge is a dangerous thing" level |
| 14:36 |
berick |
which is why i'm in here monologuing |
| 14:36 |
* berick |
will clean up the jetty stuff and push to random in a bit |
| 14:37 |
jeff |
i had been leaning toward client-side templates as well, but for slightly different reasons. mostly, to let the client decide what's best for its printer, while still getting the templates or a set of templates from the server, there would be more local control similar to how we have local templates in the client now. |
| 14:37 |
jeff |
and just because the current selfcheck A/T print setup is slow doesn't mean that all server-side options should be ruled out. :-) |
| 14:38 |
berick |
jeff: +1 to all of that |
| 14:40 |
jeff |
receipt printers used to cut, adjust font size, kick the cash drawer, etc via escape or control codes. that went out around the time it stopped being easy to print directly to LPT1 on a windows machine. |
| 14:40 |
jeff |
nowadays, there's usually a printer font called "control" or similar, and a "P" printed in the "control" font does something like cut. |
| 14:41 |
jeff |
but on top of that, and slightly more common are windows print drivers that are just configured to cut after every job (configurable on or off, or sometimes by installing a "cut" vs "no cut" printer object) |
| 14:45 |
jeff |
and the "change in font size" type things are more often done by the receipt printer working more like a "graphics-supporting" printer and less like a line/text-only printer. :-) |
| 14:50 |
|
davidcandlestick joined #evergreen |
| 14:51 |
berick |
jeff: you coming to the EG conf? :) |
| 14:51 |
jeff |
I am. |
| 14:51 |
berick |
excellent |
| 15:00 |
hbrennan |
eg2014++ |
| 15:06 |
|
stevenyvr joined #evergreen |
| 15:09 |
|
Bmagic|2 joined #evergreen |
| 15:11 |
Wyuli |
Perhaps a silly question! |
| 15:12 |
Wyuli |
If I create a list in Evergreen, I see there is a way to download it as a CSV file. |
| 15:12 |
Wyuli |
I don't suppose there's a not terribly inconvenient way to take a CSV file and upload it to make a list? :) |
| 15:12 |
hbrennan |
That would be fun |
| 15:14 |
phasefx |
Wyuli: as staff, you can pull it up in the Record Buckets interface and export the displayed columns as CSV |
| 15:15 |
bshum |
Wyuli: As for importing back in, there's a wishlist LP for that |
| 15:17 |
Wyuli |
Excellent, thanks! phasefx++ bshum++ |
| 15:18 |
Wyuli |
We are trying to come up with a way to easily point folks to newly cataloged items. First thought was take a report CSV of newly cataloged items -> upload it as a Public list -> link to list on website -> eat cake -> profit |
| 15:18 |
* dbs |
thinks... doesn't freshmeat have a TPAC output now? |
| 15:18 |
Wyuli |
Is there an obvious way to accomplish that that I may be overlooking? |
| 15:20 |
dbs |
http://hostname/opac/extras/feed/freshmeat/opac/biblio/import/10/ |
| 15:21 |
dbs |
substitute /opac/ for /rss2/ or /atom/ if you want to pull that format instead |
| 15:21 |
dbs |
and /10/ for /20/ or whatever for a longer list |
| 15:24 |
eeevil |
and there's the item_age feed as well |
| 15:24 |
eeevil |
to order by copy.create_date instead of record.create_date |
| 15:25 |
Wyuli |
dbs++ eeevil++! |
| 15:25 |
Wyuli |
Thanks, I will take a look at this. :) |
| 15:26 |
dbs |
eeevil: right! |
| 15:30 |
* eeevil |
looks for an example |
| 15:31 |
hbrennan |
This looks like magic |
| 15:32 |
Wyuli |
scratchpad:random_magic_spells |
| 15:32 |
hbrennan |
I'd love to use this to get the word out about new items, however the problem has been that there still isn't a way for staff to be notified when something is placed on hold that might not be available yet |
| 15:33 |
Wyuli |
So yeah, comes from a book of modern magic spells, essentially. |
| 15:33 |
hbrennan |
We could take forever to finish processing a new book, so we want to know if someone puts it on hold. Otherwise then they're angry when it's not immediately ready |
| 15:34 |
hbrennan |
Still need a way to grab items "In Process" and with holds attached |
| 15:34 |
hbrennan |
and as far as I know, there still isn't |
| 15:37 |
phasefx |
reporting could do it |
| 15:37 |
hbrennan |
We were told it wasn't possible in 2.3.4 |
| 15:37 |
hbrennan |
Because that's what we did previous to Evergreen |
| 15:37 |
phasefx |
I could be mistaken; I tend to think in sql |
| 15:37 |
hbrennan |
We just ran a daily report of items waiting to be processed for holds |
| 15:38 |
hbrennan |
well, of course it's possible in SQL. I know nothing about it, but I know it does anything! |
| 15:38 |
phasefx |
and if it's possible in sql, then it's possible in reporting with a sql view, etc. |
| 15:39 |
hbrennan |
Who's doing the Reports tutorial at the conference? I better warn them. I have a growing list of wishes |
| 15:40 |
hbrennan |
And I can get permission to ask for information from SQL, without having the power to actually get it to change anything, correct? |
| 15:41 |
bshum |
hbrennan: montgoc1 and jventuro are doing the Reports Tutorial. |
| 15:42 |
eeevil |
Wyuli: http://training-eg25b.evergreencatalog.com/opac/extras/browse/atom-full/item-age/- ... and, see the <link rel="alternate" ... /> elements. the href in those go directly to the bib in question. replace "-" with an org unit shortname. limit and offset work like other feeds ( add /$offset/$pagesize) |
| 15:42 |
bshum |
hbrennan: I'm next to jventuro now in a meeting, I'll pester her to check the IRC log :) |
| 15:43 |
hbrennan |
Thanks, bshum++ |
| 15:43 |
hbrennan |
As someone who knows basically nothing about programming, it seems so possible it almost hurts. |
| 15:43 |
hbrennan |
Find items with status "In Process" + holds waiting to be trapped |
| 15:44 |
hbrennan |
Ahh, not trapped.. captured. Sirsi slip-up |
| 15:44 |
mmorgan |
Seems like information on IN PROCESS items with holds should be more accessible in the client |
| 15:44 |
mmorgan |
you shouldn't need to run an sql or a report. |
| 15:44 |
hbrennan |
Exactly |
| 15:44 |
hbrennan |
And it's keeping us from showing off our new stuff |
| 15:45 |
hbrennan |
because we don't really want people to put holds on things if we can't be notified |
| 15:45 |
mmorgan |
In the client you can click on a menu option to see your holdshelf, the same sort of thing should be available for holds on the new items |
| 15:45 |
dbwells |
grabbing 0864, 0865, 0866, 0867 |
| 15:45 |
hbrennan |
But we don't want to hide everything from the TPAC, because it's important for patrons to see that we are getting something, even if it's not ready yet |
| 15:47 |
hbrennan |
I see something from eeevil to the open listserv 3/7/13 that looks promising. "Report to find unfilled holds" |
| 15:48 |
eeevil |
mmorgan / hbrennan: I'm not certain I'm understanding, but wouldn't you just want to make In Process both opac-visible and holdable |
| 15:48 |
Wyuli |
eeevil++, you are a rockstar. |
| 15:48 |
hbrennan |
Yes, they are opac-visible and holdable |
| 15:48 |
hbrennan |
But we aren't notified when the hold is placed |
| 15:48 |
eeevil |
well... ok |
| 15:49 |
hbrennan |
In Process is like our purgatory. We never catch up on the processing, so we want to get wanted items out faster |
| 15:49 |
eeevil |
but if it's In Process, doesn't that mean that the cataloging dept isn't finished with it? or are you just looking for a way to poke catalogers to finish high-demand items? |
| 15:49 |
hbrennan |
but we don't know what is wanted, unless patrons ask at the desk and we write it on a yellow pad |
| 15:49 |
eeevil |
ah, gotcha |
| 15:49 |
eeevil |
ok |
| 15:49 |
hbrennan |
So in a way, yes we're bending over backwards for patrons |
| 15:49 |
hbrennan |
but we're a public library! |
| 15:50 |
eeevil |
no, that's great! |
| 15:50 |
eeevil |
for patrons ;) |
| 15:50 |
hbrennan |
And it's a service we did before |
| 15:50 |
hbrennan |
It encourages people to place holds |
| 15:50 |
eeevil |
I just want to make sure I understand the purpose |
| 15:50 |
eeevil |
because sometimes (though, I don't think in this case) there is existing functionality that could be leveraged |
| 15:51 |
hbrennan |
That's what I thought when I saw your comment about a report to find things on hold that still haven't been captured by a certain date |
| 15:52 |
|
davidcandlestick joined #evergreen |
| 15:52 |
hbrennan |
instead of a date, enter "With status of In Process" |
| 15:52 |
mmorgan |
hbrennan: so it could be like a pull list, but for in process items? |
| 15:52 |
|
kbeswick joined #evergreen |
| 15:52 |
hbrennan |
Exactly |
| 15:53 |
hbrennan |
Oh, and if it was in the Circ drop-down, how awesome would that be? |
| 15:53 |
hbrennan |
I keep looking for it there :) |
| 15:53 |
* phasefx |
likes it |
| 15:53 |
eeevil |
well, "a status in in process" is a copy thing (of course), and not a simple attribute of the hold ... if, for a given hold, there are 10 non-in-process copies and one in-process copy, do you want to be alerted? |
| 15:53 |
eeevil |
s/in/of/ |
| 15:53 |
hbrennan |
Ahh, the joy of being a simple-minded one-branch library, with usually only one copy in each record |
| 15:54 |
mmorgan |
or possibly any item that's in a status that can't fill a hold. Missing, lost, mending... |
| 15:54 |
hbrennan |
It took me forever to figure out copy/volume/record because for us it's often all the same thing |
| 15:55 |
eeevil |
FWIW, I think this would be great fodder for a wishlist bug if it's not already |
| 15:56 |
hbrennan |
If only I could get other staff into using Launchpad, or anything besides just the staff client |
| 15:56 |
hbrennan |
it would have been already if I was cataloging and in charge of pulling holds! |
| 15:57 |
mmorgan |
eeevil: IMO if there are non-in process copies and yours in in process, you would want to be alerted if it's your hold. |
| 15:58 |
eeevil |
mmorgan: gotcha ... now, to define "your hold" in a consortium ;) ... |
| 15:58 |
eeevil |
which is not to be flip |
| 15:58 |
eeevil |
it's just something we have to define |
| 15:59 |
eeevil |
IOW, does that mean "you are the pickup library, and all your copies are in-process"? (that seems like a sane definition to me) |
| 15:59 |
mmorgan |
certainly! your hold = pickup at your library, in our consortium :) |
| 15:59 |
mmorgan |
eeevil: Yes. |
| 16:00 |
eeevil |
mmorgan: what about within your system where you've set up proximity adjustment to make all branches of a system "equal"? |
| 16:01 |
eeevil |
(I'm serious about this stuff ... let's define the boundaries so the code can be reasonably scoped, and then see how broadly applicable the feature would be. If it's "everyone wants exactly that" well, we've got a plan!) |
| 16:02 |
mmorgan |
if all branches in a system are equal, then an In Process item at branchA should be rushed to fill a hold at branchB in the same system. |
| 16:03 |
hbrennan |
I have nothing to add for consortium-level stuff, but I'm intrigued! |
| 16:04 |
eeevil |
mmorgan: ok, cool. so, it's really about "apparent proximity". how about "give me a list of holds where potential copies all have a pickup lib proximity of 0, filtered by status" |
| 16:04 |
eeevil |
we might be closing in on a general purpose holds UI |
| 16:05 |
eeevil |
holds browsing UI, rather |
| 16:05 |
eeevil |
toss in an org filter, and a copy count filter (to find orphaned holds) ... |
| 16:06 |
pinesol_green |
Showing latest 5 of 34 commits to Evergreen... |
| 16:06 |
pinesol_green |
[evergreen|Bill Erickson] LP#1053397 staff client MR results paging repair - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=3fd258f> |
| 16:06 |
pinesol_green |
[evergreen|Bill Erickson] LP#1053397 MR list return to grouped results - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=0a6b913> |
| 16:06 |
pinesol_green |
[evergreen|Bill Erickson] LP#1053397 icon / metarcord format additions - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=00c36d7> |
| 16:06 |
pinesol_green |
[evergreen|Bill Erickson] LP#1053397 Prevent display non-opac-visible icons / hold formats - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=c7503c7> |
| 16:06 |
pinesol_green |
[evergreen|Dan Wells] Stamping 0864-0867 for MVF, CRA, and TPAC MRs - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=81d720f> |
| 16:06 |
eeevil |
only 34? I shouldn't have squashed all those ... ;) |
| 16:06 |
hbrennan |
eeevil: Exactly. It doesn't have to DO things, just produce a lst |
| 16:06 |
hbrennan |
list* |
| 16:06 |
hbrennan |
and since basically any screen output can be printed easily.. |
| 16:06 |
hbrennan |
sounds great |
| 16:07 |
eeevil |
hbrennan: well, producing the list is actually the hard part ... because the definition of what should be on the list is not super-well defined. but if there was a wishlist bug that did that, it might create some momentum |
| 16:07 |
hbrennan |
I can't find one discussing holds in a certain status |
| 16:08 |
hbrennan |
lots about something in Clear Shelf-Expired |
| 16:08 |
hbrennan |
"one" meaning a Launchpad report |
| 16:09 |
eeevil |
hbrennan: yeah ... step 1) fix LP's search ;) |
| 16:09 |
hbrennan |
I'm using Google! |
| 16:10 |
hbrennan |
EG Lesson 1: Developers don't trust, recommend, or use Reports |
| 16:10 |
hbrennan |
EG Lesson 2: Launchpad search doesn't work |
| 16:11 |
mmorgan |
eeevil: I'm a little unsure on the proximity of 0 part - we use proximity of 2 for age hold protection and the like to keep things in our systems. |
| 16:12 |
eeevil |
mmorgan: ah. I'm referring to adjusted proximity |
| 16:12 |
eeevil |
you may well not use that |
| 16:13 |
mmorgan |
yeah, no immediate plans to use that. |
| 16:13 |
eeevil |
but, for someone that does, their pickup lib can be made equal to, well, any others, for hold targetting order ... and that's more or less the definition of "here" for evergreen |
| 16:14 |
eeevil |
"it costs 0 to transit to X", where that is true both for pickup-lib==copy-lib and, optionally, other locations |
| 16:15 |
eeevil |
so it's something to take into account |
| 16:19 |
eeevil |
Dyrcona: correct me if I'm wrong, but I though that the new marc_export was merged to master already... |
| 16:19 |
eeevil |
AH! ... nm, it was |
| 16:24 |
|
Bmagic joined #evergreen |
| 16:28 |
|
RBecker joined #evergreen |
| 16:33 |
dbwells |
grabbing 0868 |
| 16:36 |
* phasefx |
loses his mind and catalogs every launchpad bug as a MARC record in Evergreen ;) |
| 16:36 |
hbrennan |
mmorgan: Do you want to start a Lauchpad wish for the holds capturing? You have an understanding of how it would work for a consortium |
| 16:37 |
berick |
phasefx: meta-bug searching! ;) |
| 16:37 |
phasefx |
berick++ |
| 16:37 |
bradl |
great way to ensure MARC never dies |
| 16:38 |
pinesol_green |
[evergreen|Bill Erickson] LP#1281744 Aged hold purging column repair - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=11d70bc> |
| 16:38 |
pinesol_green |
[evergreen|Dan Wells] Stamping 0868 - 'behind_desk' fixes - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=e406e36> |
| 16:38 |
phasefx |
need to shorten the allowed length of a MARC record every year; that'll kill it. Maybe when it matches the length of a tweet |
| 16:39 |
phasefx |
catalogers can just text each other their bib records |
| 16:40 |
phasefx |
must be Friday :) |
| 16:40 |
hbrennan |
phasefx: I sure hope so |
| 16:41 |
berick |
tom sawyer: lovable scamp, almost dies in a cave. |
| 16:41 |
Dyrcona |
Like all bad ideas, MARC will never die. |
| 16:43 |
Dyrcona |
Y'know it is kind of hard to fine patrons who a) have overdues and b) owe fines on those overdues when you're just searching the staff client by putting in last names and going through the lists of the first 50 or so patrons that turn up. |
| 16:44 |
berick |
i've had similar frustrations looking for demo patrons |
| 16:44 |
eeevil |
Dyrcona: toss the web SC prototype on your prod server. at least your browsing will be quicker! ;) |
| 16:45 |
Dyrcona |
☺ |
| 16:45 |
Dyrcona |
:) |
| 16:45 |
mmorgan |
hbrennan: sure, I can take a stab at it, won't get to it til Monday, though. can't put words together so well on Friday afternoon :), and loose ends to tie. |
| 16:45 |
Dyrcona |
hit the compose key. |
| 16:45 |
Dyrcona |
this is on my dev server, so I should throw it up sometime. |
| 16:46 |
hbrennan |
mmorgan: No worries. We upgraded from 2.3.4 to 2.5.2 on Wednesday, so I'm surrpised I can still function. But the conversation was excellent today so I didn't want to lose the momentum |
| 16:46 |
dbwells |
Dyrcona: You made the world's smallest smiley-face in my IRC client :) |
| 16:46 |
* berick |
needs to rebase it to master soon |
| 16:46 |
hbrennan |
mmorgan: If you send something to the list when it's there, I'll add my comments |
| 16:47 |
Dyrcona |
So I find 1 patrons that owes a buck and some change, but has nothing checked out. |
| 16:48 |
mmorgan |
hbrennan: yes, feedback from the general list would be good. It puzzles me why no one has asked the question before. |
| 16:48 |
Dyrcona |
Then, another patron with 5 overdues who owes nothing, because that library doesn't charge fines. |
| 16:48 |
mmorgan |
hbrennan++ |
| 16:48 |
hbrennan |
mmorgan: I searched high and low and found nothing. We questioned this during migration, then just got too busy |
| 16:49 |
hbrennan |
mmorgan++ eeevil++ phasefx++ for the great holds conversation today! I feel hopeful! |
| 16:49 |
eeevil |
Dyrcona: select usr from money.billable_xact_summary where total_owed > 0.0 and balance_owed <> total_owed; -- not sure on that table name... |
| 16:50 |
Dyrcona |
yeah. |
| 16:50 |
mmorgan |
could be everyone has a workaround. Our sysadmin wrote a titles with holds script for our previous system, and rewrote it after we migrated. |
| 16:50 |
hbrennan |
now, a serials questions: Where the heck do I change the notes that show in Issues Held in the OPAC? Subtitled Holdings Summary. Instead of "live" holdings info, I had our notes put in... perhaps gmcharlt could point me in the right direction since he put the notes there |
| 16:51 |
jeff |
Dyrcona: select usr from money.materialized_billable_xact_summary where last_billing_type = 'Overdue materials' and balance_owed > '0.00' order by xact_start desc limit 100; :-) |
| 16:52 |
jeff |
oh, eeevil already went that direction. |
| 16:52 |
Dyrcona |
Yeah, I think I will have to do something like that. |
| 16:54 |
jeff |
the staff client isn't really designed to facilitate casual browsing of financial or patron data. :-) |
| 16:57 |
Dyrcona |
Heh... seclect barcod [sic]. |
| 16:57 |
Dyrcona |
yeah. I was hoping I could luck into finding someone to use as a guinea pig. |
| 16:57 |
Dyrcona |
Anyway, I took jeff's query and joined actor.card. |
| 16:58 |
Dyrcona |
added a distinct and now I have 100 patrons to play with. |
| 16:59 |
Dyrcona |
Ah, one hitch. |
| 16:59 |
Dyrcona |
I need ones where the overdue material is still checked out. |
| 16:59 |
Dyrcona |
We don't block privileges until $10 is owed, so a lot of patrons and staff let smaller amounts slide. |
| 17:03 |
jeff |
AND EXISTS (SELECT 1 FROM action.open_circulation ocirc WHERE id = ocirc.id) |
| 17:04 |
jeff |
which will include lost-and-still-owes and such. limit to stop_fines of MAXFINES if you only want "vanilla overdues"? |
| 17:04 |
Dyrcona |
I just wrote something that joins actor.card, action.circulation and money.billing. |
| 17:05 |
jeff |
you're probably all set then. :-) |
| 17:05 |
Dyrcona |
Yeah. I guess I'm not lucky. :) |
| 17:06 |
Dyrcona |
I now have a list of 9,997 barcodes. I expected more. |
| 17:06 |
Dyrcona |
Sorry, 9,977. |
| 17:08 |
berick |
jeff: it's crusty, rough, and poorly documented ATM, but it answered my questions: http://git.evergreen-ils.org/?p=working/random.git;a=shortlog;h=refs/heads/collab/berick/hatch |
| 17:08 |
|
mrpeters left #evergreen |
| 17:11 |
Dyrcona |
dbwells: I'm not able to reproduce the scenario with the paying fines and then a back dated checkin that you mentioned this morning in a private chat. |
| 17:11 |
|
mmorgan left #evergreen |
| 17:11 |
Dyrcona |
I'll keep looking, though. |
| 17:12 |
Dyrcona |
jeff: I should have excluded some stop_fines... |
| 17:16 |
Dyrcona |
hmmm... xact_finish is null, checkin_time is null, patron owes three dollars, doesn't show in list of patrons items out..... |
| 17:19 |
gmcharlt |
hbrennan: look under basisc summary in the distributions tab of the serials control view |
| 17:19 |
jeff |
Dyrcona: what is stop_fines on that example patron? |
| 17:20 |
Dyrcona |
maxfines |
| 17:20 |
Dyrcona |
However, the checkin_time is not null when I look at the circ, so something is wrong with my query. |
| 17:20 |
jeff |
ah |
| 17:20 |
hbrennan |
gmcharlt: Not seeing it. Should it be one of the tabs across the top? Legacy Record Entry looks promising, but it's empty |
| 17:22 |
eeevil |
AAAAAAAAAAAAHHHHHHHHHHHHH WEB-INF AAAAAAAAAAAAAAAAAAAHHHHHHHHH MY EYESSSSSSSSSSSSSSSS! THE GOGGLES, THEY DO NOTHING! |
| 17:22 |
Dyrcona |
ah. need parentthesis. |
| 17:23 |
eeevil |
berick: is that "hatch" like the hatch from Lost? |
| 17:23 |
jeff |
upgraded JasperReports Server here recently. I've been seeing WEB-INF a lot. :P |
| 17:24 |
berick |
eeevil: *ding ding ding* |
| 17:24 |
eeevil |
nice :) |
| 17:24 |
* csharp |
fears THE OTHERS |
| 17:25 |
eeevil |
needs a method that returns the current value of the countdown clock |
| 17:25 |
berick |
eeevil++ |
| 17:25 |
berick |
totally |
| 17:25 |
csharp |
or a nuke that blows up Evergreen? |
| 17:25 |
berick |
csharp: just sends it back in time |
| 17:25 |
csharp |
berick++ |
| 17:26 |
phasefx |
that's kind of deep.. the bomb shelter |
| 17:26 |
hbrennan |
Right when I'm about to go to lunch.. now I want to nerd out on Lost |
| 17:27 |
eeevil |
berick: do you get websockets for free? ... I guess yes |
| 17:28 |
eeevil |
hrm.. ws://localhost:8080/... maybe |
| 17:28 |
eeevil |
ok ... I'm out |
| 17:28 |
berick |
eeevil: yeah, but you have to enable the module |
| 17:28 |
berick |
which is not enabled by default |
| 17:28 |
berick |
oh, and it would need a slightly different servlet handler |
| 17:29 |
berick |
almost free |
| 17:31 |
|
akilsdonk_ joined #evergreen |
| 18:56 |
|
mtj_ joined #evergreen |
| 21:07 |
|
RBecker joined #evergreen |
| 22:18 |
|
mtj_ joined #evergreen |
