09:12 |
Dyrcona |
I'm using FBReader and the Hyperionics (sp?) Text to Speech plugin. |
09:13 |
krvmga |
has anyone run across a problem with IE11 where format filters in advanced search show up as a blank box? |
09:16 |
|
rfrasur joined #evergreen |
09:20 |
csharp |
krvmga: bear with me while I upgrade IE in my windows VM and I'll test |
09:20 |
csharp |
oh and |
09:20 |
csharp |
ie-- |
09:20 |
csharp |
@karma |
09:20 |
pinesol_green |
csharp: Highest karma: "dbs" (742), "bshum" (609), and "tsbere" (561). Lowest karma: "ie" (-49), "^" (-26), and "----------------------------------" (-18). You (csharp) are ranked 12 out of 2060. |
09:20 |
csharp |
ie-- |
09:20 |
csharp |
@karma ie |
09:27 |
rfrasur |
jboyer-isl: is there somewhere to look at past daily reconciliations? |
09:28 |
csharp |
krvmga: http://picpaste.com/ie11-advanced-search-crkHqcF3.png - a screenshot of our advanced search (2.3.6) in IE11 |
09:28 |
csharp |
I don't see anything missing |
09:30 |
krvmga |
csharp: yes, this is the same experience that i had when i tested it. |
09:30 |
krvmga |
csharp: a library contacted me and told me the blank box happened to them. i was not able to duplicate the error. |
09:31 |
krvmga |
csharp: they sent me a screenshot. everything was as you'd expect except the format filter box was blank. |
09:31 |
Dyrcona |
@xkcd 583 |
09:31 |
pinesol_green |
Dyrcona: CNR: http://xkcd.com/583 |
09:32 |
jboyer-isl |
rfrausr: Not really. I can get older info if you need it, though. |
09:47 |
krvmga |
jeff: interesting thought. checking. |
09:47 |
jeff |
but it can make a major difference if it's set incorrectly (on when should be off, off when should be on, etc) |
09:51 |
jeff |
if the screen shot in question included the address bar, you can probably see the indicator -- though it's gotten more difficult to read in recent versions, iirc. |
09:51 |
krvmga |
jeff: i just tested with compatibility mode on and off and it didn't make a noticeable difference |
09:52 |
krvmga |
jeff: plus, if it's not showing up in the staff client, isn't that mozilla-based? |
09:52 |
jeff |
you can probably safely rule that one out, then. |
09:52 |
jeff |
krvmga: i don't follow your last inquiry. |
09:52 |
krvmga |
jeff: the format filters aren't displaying in search in the staff client either for them |
13:08 |
|
akilsdonk_ joined #evergreen |
13:14 |
|
dMiller__ joined #evergreen |
13:17 |
|
stevenyvr2 joined #evergreen |
13:24 |
paxed |
testing on a mini laptop == slightly painful. |
13:25 |
rfrasur |
why a mini laptop? |
13:26 |
|
mcooper joined #evergreen |
13:26 |
paxed |
traveling |
13:26 |
|
dMiller_ joined #evergreen |
13:27 |
rfrasur |
makes sense. and you have to work during the travel? |
13:28 |
paxed |
i'm working on a new Added Content handler and want to finish that. and this is work-related travel anyway |
13:28 |
paxed |
well, the code is done, now just need to test... |
13:29 |
rfrasur |
;-), in that case...good luck and hopefully the tests go well so you don't have to fuss with snack-size keyboards and monitors. |
13:30 |
|
hopkinsju joined #evergreen |
13:32 |
* Dyrcona |
finds the staff client very cumbersome, and some interfaces unusable, on a resolution less than 1920x1080. |
13:32 |
bshum |
There are resolutions lower than that? :P |
13:40 |
paxed |
couple hours, and i have working code for bug 1133464 ... now to make it prettier, and configurable. |
13:40 |
pinesol_green |
Launchpad bug 1133464 in Evergreen "Use cover image/blurb URL from field 856" (affected: 3, heat: 14) [Wishlist,Triaged] https://launchpad.net/bugs/1133464 |
13:43 |
bshum |
Fancy |
13:43 |
paxed |
i was surprised how easy that was, actually. of course everything is currently hardcoded and whatnot, but our test server catalog looks much more colorful now... |
13:44 |
* rfrasur |
likes colorful things |
13:44 |
paxed |
hmm. now how do i query OUS in perl? |
13:45 |
paxed |
or perhaps this should be moved into it's own added content handler ... |
14:49 |
Dyrcona |
senator++ |
14:49 |
eeevil |
bshum: I can't argue with the counter-example of acq, except to say that that's about as complicated as you can get ... I can't think of another similar example |
14:49 |
csharp |
senator: +1 |
14:49 |
gmcharlt |
#info Discussion of the proposal to be moved to open-ils-dev |
14:49 |
gmcharlt |
#topic Requiring automated tests for new code/database modifications. |
14:49 |
eeevil |
anyway, my point was, am I alone in seeing the "arbitrary waypoints as versioned releases" as a way around much of this? |
14:50 |
eeevil |
and with that, I'll hush |
14:50 |
gmcharlt |
eeevil: no, you're not, but that implies some things that we don't have yet but hopefully will accumulate over time |
14:51 |
* Dyrcona |
ducks. |
14:52 |
Dyrcona |
I think dbs email was a call for this to be more formal and eeevil and others have made strides in that direction. |
14:52 |
Dyrcona |
This, I hope, we could actually "vote" on. |
14:53 |
dbs |
I think we could make pgTAP tests for database changes a matter of policy. Not sure if we're ready for requiring tests for general code modifications though |
14:54 |
jeff |
I think that all-or-nothing "no commits without matching unit tests" is impractical given our current testing infrastructure. Starting with database changes where we have pgTAP seems like a good beginning. |
14:54 |
jeff |
"Encouraged" elsewhere, and we can transition to "required" as our testing methodologies mature. |
14:54 |
dbs |
FWIW, I thought August's dev meeting had already voted on the pgTAP test requirement? |
14:54 |
dbs |
jeff++ |
14:55 |
Dyrcona |
Was that a vote on a requirement? |
14:55 |
* phasefx2 |
thought that was an "ease into it" |
14:56 |
dbs |
phasefx2 is right |
14:56 |
dbs |
http://evergreen-ils.org/meetings/evergreen/2013/evergreen.2013-08-27-14.04.log.html |
14:56 |
RoganH |
As someone who handles contracts for development rather than development myself I wish it was a requirement. When I get competing bids and vendor X says we can cut costs for you by not doing tests that seems attractive to some of my members - the same ones who complain about unexpected behavior of course but still. |
14:57 |
Dyrcona |
As a developer, I'd have to learn to do the tests... |
14:57 |
Dyrcona |
I think the rumblings are that devs in general would like to see these be required. |
14:58 |
jeff |
browsing "git log -- Open-ILS/src/sql/Pg/upgrade", i don't know if i can say without a doubt that every recent change immediately lends itself to having a unit test. |
14:58 |
gmcharlt |
#startvote Shall pgTAP test cases (where applicable) be required for new commits starting with 2.6? Yes, No, Postpone |
14:58 |
pinesol_green |
Begin voting on: Shall pgTAP test cases (where applicable) be required for new commits starting with 2.6? Valid vote options are Yes, No, Postpone. |
14:58 |
pinesol_green |
Vote using '#vote OPTION'. Only your last vote counts. |
14:59 |
phasefx2 |
#vote Yes |
14:59 |
berick |
gmcharlt: to clarify, 2.6 means staring now w/ all new features? |
14:59 |
jeff |
Is this meaningless if we don't have a good idea of what "where applicable" means? |
14:59 |
bshum |
"where applicable" is vague enough that given my general infamiliarity with pgTAP use, I'm not sure I'm ready to commit to that. |
14:59 |
phasefx2 |
you can make an assertion about any upgrade script; at the very least, it might help someone check that they actually ran a required upgrade script |
14:59 |
RoganH |
#vote Yes |
14:59 |
phasefx2 |
but coming up with mock environments to test behavior for stored procedures.. that might be a bit much for some folks |
15:00 |
bshum |
But I'm among the "have to learn to do the tests..." folks |
15:00 |
gmcharlt |
berick: yes -- though the RM may have the option to exercise discretion |
15:00 |
berick |
thanks |
15:00 |
berick |
#vote YES |
15:00 |
phasefx2 |
if doing an independent unit test as oppossed to relying on concerto test data |
15:00 |
eeevil |
gmcharlt: s/may/should/ IMO |
15:00 |
berick |
tis inevitable |
15:00 |
gmcharlt |
bshum: in my view, a requirement will encourage folks to learn the framework |
15:01 |
dbwells |
#vote Yes |
15:01 |
eeevil |
#vote yes |
15:02 |
dbwells |
I am anxious to see how this will play out in practice, but it certainly can't hurt to try. |
15:02 |
jeffdavis |
gmcharlt: or conversely, folks who might otherwise contribute a patch might decide it's too difficult |
15:02 |
jeffdavis |
(not that I'm opposed to unit tests) |
15:02 |
gmcharlt |
jeffdavis: the requirement applies to the commit, not the person, IMO |
15:02 |
eeevil |
dbwells: in practice, you'll get to threaten revert unless a test shows up :) |
15:03 |
gmcharlt |
IOW, a more experienced dev helping write unit tests for a patch originated by somebody else would be a Good Thing |
15:03 |
csharp |
can buildbot run the tests? |
15:03 |
berick |
dbwells: you are this -><- close to be elected benevolent overlord |
15:03 |
csharp |
(thinking of pgTAP) |
15:03 |
csharp |
berick++ |
15:03 |
phasefx2 |
csharp: ultimately yes |
15:04 |
jeff |
If I'm crafting a database upgrade script and base-schema changes to add a new org unit setting type, 1) is that "where applicable" and requires a pgTAP test? and 2) What is tested -- test for successful insertion / existence in config.org_unit_setting_type? |
15:04 |
jeffdavis |
which could mean experienced devs spend more time on others' patches, or that those patches languish |
15:04 |
jeffdavis |
again, I'm not opposed, but I do worry a bit about the downsides |
15:04 |
eeevil |
jeff: that's a case where it seems useless. we have config.upgrade_log for that |
15:05 |
dbs |
#vote yes |
15:05 |
bshum |
In principle, I agree; I just won't say when I'm ready till I know more. |
15:05 |
bshum |
#vote yes |
15:05 |
jeff |
jeffdavis: I think that "patches languishing" can be evaluated at a time afterward. If there's an issue, the decision can be re-evaluated or additional steps can be taken to make things easier to test, etc. |
15:05 |
berick |
jeff: good question. imo, seed data that's not used by the DB (only the app) is not really worth the effort of testing in pgtap; app-level tests, sure. |
15:05 |
dbs |
code reviewers will certainly help where needed methinks |
15:06 |
eeevil |
also, my as yet unsent plan for non-changing base schema would reduce the overhead of this... |
15:06 |
gmcharlt |
and requiring unit tests can help promote more code review |
15:06 |
* eeevil |
-- |
15:06 |
phasefx2 |
jeff: there's also automated test creation for simple things like that, that we may want to think about |
15:06 |
gmcharlt |
(and I've seen evidence of that in Koha-land) |
15:06 |
dbs |
(For 2.6 I'd kind of like to make good on the threat others have for pulling the fake org_unit stuff out of seed data and into concerto & friends) |
15:07 |
jeff |
Okay. Further questions about "where applicable" and such to be sought on open-ils-dev, where you can either find help in conceptualizing / creating a test, or concensus on "not applicable"? |
15:07 |
dbs |
(But that's a different subject :) ) |
15:07 |
Dyrcona |
#vote postpone # cause there's no "abstain" option other than not voting. |
15:07 |
pinesol_green |
Dyrcona: postpone # cause there's no "abstain" option other than not voting. is not a valid option. Valid options are Yes, No, Postpone. |
15:10 |
gmcharlt |
Dyrcona: you didn't seen the Officail Badge dbwells has been wearing for the past few months? |
15:10 |
gmcharlt |
anyway |
15:10 |
gmcharlt |
#endvote |
15:10 |
pinesol_green |
Voted on "Shall pgTAP test cases (where applicable) be required for new commits starting with 2.6?" Results are |
15:10 |
pinesol_green |
Yes (10): jeff, berick, RoganH, dbwells, gmcharlt, bshum, senator, dbs, phasefx2, eeevil |
15:10 |
pinesol_green |
Postpone (1): Dyrcona |
15:11 |
gmcharlt |
#agreed pgTAP test cases (where applicable) are required for new commits start with 2.6 |
15:11 |
phasefx2 |
yay |
15:11 |
gmcharlt |
any other last minute insertions before we end the meeting? |
15:12 |
gmcharlt |
... |
09:21 |
jeff |
(purely imo) |
09:22 |
tsbere |
jeff: Security, or just private for now? |
09:23 |
Dyrcona |
On the upside, all the issues turning up related to this bug indicate the best fix is a complete rewrite of OpenILS::Appication::Circ and friends! |
09:23 |
Dyrcona |
This is also NOT the kind of thing the proposed automated testing would turn up. |
09:27 |
jeff |
tsbere: i'd recommend private security. we can worry about making some clear policy on what the difference is later. |
09:33 |
tsbere |
Well, for those who can see such things: https://bugs.launchpad.net/evergreen/+bug/1249367 |
09:33 |
pinesol_green |
tsbere: Error: malone bug 1249367 not found |
10:17 |
jeff |
so, 10 overdue billings, a cash payment, a forgive (poor man's partial void) for the rest, then a lost item billing, then the lost item is returned, another forgive for the lost amount, and a re-bill of the forgiven overdues amount to "un-void" the unpaid overdues. |
10:18 |
jeff |
i filled a whiteboard or two with this a while back. it was "fun". |
10:18 |
* Dyrcona |
gives jeff a lollipop. :) |
10:19 |
jeff |
in bringing mmpbbt back out of the freezer, i'm considering how we can test billing things. probably starting with finding some clean and some pathological examples from our live data for mmpbbt itself. hope to learn something that can be applied elsewhere. |
10:21 |
RoganH |
Billing is simple. I didn't give staff forgive or void permissions. |
10:21 |
dbwells |
When it comes to the fines situation in EG, I think we could benefit from some more careful semantics. We can't really solve problems until we are talking about the same things. |
10:22 |
Dyrcona |
RoganH: Lucky you. Ours would kill us if we didn't. |
10:22 |
jeff |
dbwells: standardizing terminology and expectations and such, yes. |
10:22 |
RoganH |
Only a few managers have forgive permissions and they know they can be held accountable and no one has void (except me). |
10:22 |
Dyrcona |
dbwells: When it comes to fines, just remove it from Evergreen, and say, we don't do fines. |
10:23 |
jeff |
dbwells: i was thinking that this would likely be necessary for testing as well, since if you and i define "correct behavior" differently, my tests will not match your expectations. |
10:23 |
dbwells |
For instance, I would say that any system which allows "partial voids" is a path to madness. Not that we don't need something *like* that, but shouldn't void mean "void" (aka, it never happened)? |
10:23 |
Dyrcona |
I would now like to point to something Edsger Djikstra said about testing, but you can Google it yourself. |
10:24 |
jeff |
dbwells: right. better "written off", etc. |
10:25 |
dbwells |
jeff: exactly! |
10:25 |
jeff |
but if you bill someone overdue fines, then you make those fines no longer due, is that the same as "voided, never happened"? hard line to draw. |
12:48 |
linuxhiker |
Howdy folks, anyone ever seen this?? Can't locate object method "has_a" via package "actor::user" (perhaps you forgot to load "actor::user"?) at /usr/lib/perl5/site_perl/5.8.8/OpenILS/Application/Storage/CDBI.pm line 433. |
12:50 |
jeff |
linuxhiker: more context would be helpful -- when/where are you encountering this? |
12:51 |
bshum |
Also, 5.8.8? Yeesh |
12:52 |
linuxhiker |
bshum: centos5.10 :P |
12:52 |
linuxhiker |
jeff: good question, I am checking, I was just given the file to test, I believe it is on load (we are getting an 550 from the website) |
12:54 |
Dyrcona |
Fieldmapper isn't loaded or properly inited. Usually because the OpenILS::System->bootstrap failed. Usually because it can't find or was not given the bootstrap config. <- Just a wild guess. |
12:54 |
Dyrcona |
;) |
13:39 |
linuxhiker |
when running autogen.sh I am getting: Session Error: routerapache05-priv/opensrf.settings IS NOT CONNECTED TO THE NETWORK!!! |
14:30 |
Dyrcona |
since, even... |
14:32 |
sseng |
berick: Dyrcona: bug submitted (https://bugs.launchpad.net/evergreen/+bug/1249476) |
14:32 |
pinesol_green |
Launchpad bug 1249476 in Evergreen "Api to retrieve an authority record not working" (affected: 1, heat: 6) [Undecided,New] |
14:33 |
Dyrcona |
Ok. I'll make a branch that you and anyone else can test. I'll throw it on my dev server, too. |
14:33 |
sseng |
Dyrcona: great, thanks!! |
14:34 |
Dyrcona |
sseng++ |
14:42 |
|
gsams joined #evergreen |
14:43 |
mceraso |
dbwells: Just did a clean installation of 2.5.0 and it worked |
14:44 |
jeff |
mceraso++ thanks for testing! |
14:44 |
mceraso |
dbwells: using Ubuntu 12.04 |
14:45 |
bshum |
mceraso++ |
14:45 |
dbwells |
mceraso++ # thanks so much! |
16:20 |
Dyrcona |
But. I do have a mad scheme to do my ingesting entirely in parallel... |
16:21 |
Dyrcona |
One forked process doing all of the records with browse ingest, and then several forked processes doing the other reingests in groups of 10 000 records! |
16:21 |
Dyrcona |
Let the madness commence! |
16:23 |
Callender |
dbwells: Just FYI, I've loaded 2.5.0 onto our public demo box, login information here (http://evergreen-ils.org/dokuwiki/doku.php?id=community_servers) and ran it through a lot of random tests and everything is looking good to me |
16:24 |
dbwells |
Callender++ # awesome, thanks! |
16:56 |
|
linuxhiker joined #evergreen |
16:58 |
|
mdriscoll left #evergreen |
17:50 |
|
dMiller joined #evergreen |
17:50 |
TominMichigan |
bsum: I did, but I don't see anything obvious. |
17:51 |
bshum |
I haven't done this in awhile, but does the settings-tester script give you anything? |
17:52 |
TominMichigan |
I get the same error when testing. If I run the script before autogen, it seems to come through clean. But if I run it after that, then the opensrf tests fail. |
17:53 |
bshum |
That's... weird sounding. |
17:54 |
dbwells |
TominMichigan: after restarting all services, are you able to do the srfsh login test in the install doc? |
17:54 |
TominMichigan |
dbwells: sometimes. |
17:55 |
TominMichigan |
I can't figure out when it fails. Like right now, I can to srfsh, but I get a received no data from server error when I do the math test. |
17:55 |
TominMichigan |
Other times when I've tried, I get a completely different error and srfsh doesn't launch at all. Unfortunately, I didn't copy that error message. |
17:56 |
dbwells |
TominMichigan: I am not sure if the math service runs after Evergreen is installed, since it sets up its own services (the math one is just for testing OpenSRF). |
17:56 |
TominMichigan |
dbwells: the login test comes back with Received no data from server |
17:57 |
TominMichigan |
dbwells: that's insteresting. I think it works after a fresh reboot. I'm going to try that to see if I'm chasing my tail with the math test. |
17:58 |
TominMichigan |
dbwells: I know it's late ... don't let me keep if you want to go. I can come back here on Monday. |
17:58 |
dbwells |
TominMichigan: it looks like the 'math' service is in the Evergreen default configs too, so that should be working for you. |
18:00 |
TominMichigan |
dbwells: I'm double-checking now that the math works on a fresh restart. |
18:00 |
dbwells |
How much memory is on this test box? |
18:00 |
TominMichigan |
1G |
18:00 |
bshum |
1 G is too low |
18:00 |
bshum |
You probably need at least 2 GB |
18:07 |
dbwells |
TominMichigan: 1) If you want to tinker and see if memory is the problem, you could try hacking away some services and see if you get more reliable results. I think it would mean just commenting some lines from the <activeapps> section in opensrf.xml, but that's without looking. |
18:08 |
bshum |
(also look at top to see what you're using) |
18:08 |
* bshum |
really wanders off |
18:08 |
TominMichigan |
dbwells: I could try that. Can you think of some services that would be worth cutting out as a test. |
18:08 |
dbwells |
e.g. Just run opensrf.settings and opensrf.math and see if the math service works every time, then maybe try the same with open-ils.auth for the login test. |
18:08 |
dbwells |
Evergreen won't work, obviously, but at least you would know the plumbing was in place correctly. |
18:09 |
TominMichigan |
dbwells: I'll try that. I'm going to reconfigure the server with more RAM, and then try commenting out services if I still have trouble. |
18:09 |
TominMichigan |
You've both been very helpful. Thanks for staying late to help me out |
18:11 |
dbwells |
dang, thought two was only half typed, oh well |
14:15 |
krvmga |
yboston: i think that's good. |
14:15 |
kbutler |
yboston++ |
14:15 |
rfrasur |
so, 10-5 |
14:15 |
yboston |
I will plan to stay until like 6 zero to test the asciidoc produced, to look for syntax errors, etc |
14:16 |
yboston |
thanks |
14:16 |
yboston |
I will move on to other issues, if that is OK |
14:16 |
yboston |
yes 10-5 PM |
14:17 |
yboston |
next issue: "location coordinators" |
14:18 |
yboston |
I just realized I want to set up communication with the "hosts" at the individual communications. lets share telephone numbers by email |
14:18 |
yboston |
sorry, that should say individual locations |
14:19 |
yboston |
So far I am hosting folks, and ESI is hosting folks. Anybody else acting as a host? |
14:20 |
rfrasur |
yboston: I'm fine with that, but I'll be working from home that day and plan to use the hangout. I don't think, in Indiana, there's going to a centralized location this year. |
14:21 |
yboston |
So I think it is just Erica and me hosting. If something goes wrong I can look up Erica's number. ESI should have my cell on file too |
14:22 |
akilsdonk |
yboston: sounds good. I can send you the best contact # for ESI |
14:46 |
jecs |
oh sorry |
14:46 |
yboston |
sorry I meant Angela from ESI |
14:47 |
akilsdonk |
yboston: so far no one has said they would be joining us. :( but we would love to host anyone in the area who is going to be participating |
14:47 |
yboston |
#action yboston will send email to list to get RSVPs for those coming to Berklee for DIG hack-a-way |
14:47 |
yboston |
akilsdonk: thanks again for your offer. This should be a yearly event, so there is always next year. |
14:48 |
yboston |
also, can you send me the login credentials to the test server when it is ready? |
14:48 |
akilsdonk |
yes, I will do that as soon as it is ready |
14:48 |
yboston |
thanks |
14:49 |
yboston |
BTW, it is already 2:48, we should talk a bit about what we can work on, based on what is already listed on the hack-a-way page or any new ideas you may have |
09:51 |
rfrasur |
Now if it just had a fold out double/triple monitor that didn't make it unbalanced... |
09:59 |
paxed |
how come this search finds only two hits: http://62.148.106.92/eg/opac/results?query=title%3A^decade%24 ...when there's also this: http://62.148.106.92/eg/opac/record/213825 |
10:02 |
paxed |
select * from metabib.real_full_rec where tag='245' and subfield='a' and value like 'decade'; returns all three. |
10:02 |
kmlussier |
paxed: IIRC, when I previously tested anchored searching, and exact match was an exact match. I suspect the forward slash is the reason why you aren't retrieving that record in your results. |
10:03 |
kmlussier |
s/and/an |
10:04 |
paxed |
oh. yes, you're right. |
10:04 |
paxed |
ffs |
10:06 |
paxed |
i expected it to find the exact title as shown in the search results screen. (without the slash at the end) |
13:32 |
rfrasur |
dbs: that's what I saw from afar as well. |
13:33 |
Dyrcona |
jcamins: Just bust the locks and take it over. |
13:34 |
rfrasur |
Dyrcona++ #my thoughts exactly |
13:43 |
pinesol_green |
[evergreen|Mike Rylander] Live pgTAP test for 0847, authority overlay generator fixes - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=3e7f6f1> |
13:44 |
rfrasur |
rjackson-isl: we haven't gotten an all clear for cataloging yet, have we? |
13:46 |
dbs |
eeevil: does that pgTAP test assume that concerto has been loaded, given the "FROM authority.record_entry WHERE id = 1" clause? |
13:48 |
|
mjingle joined #evergreen |
13:49 |
eeevil |
dbs: yes. I chose to follow the statement in the README for the stuff in live_t/ |
13:52 |
jeff |
would there be any objection / issues with a patch to allow password resets to accept a single value, "enter your library card number or username", with org unit regex backed tests for detecting "that's a barcode!" vs "not a barcode, try it as a usrname!" logic? |
13:53 |
jeff |
(similar to the logic used for logging in) |
13:53 |
rjackson-isl |
rfrasur - waiting green light from Anna |
13:53 |
rjackson-isl |
she is at Montpelier doing the go live |
13:53 |
|
smyers_ joined #evergreen |
13:53 |
rfrasur |
rjackson-isl: okie. will relay. |
13:54 |
dbs |
eeevil: eh. okay. I'm not a big fan of that, personally, as we can't run that test on a production system. |
13:57 |
pinesol_green |
[evergreen|Mike Rylander] Only one of these tests was valid, removing the bad one - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=a32f742> |
13:59 |
eeevil |
dbs: understood, and I'm not a /big/ fan either, but the mock env for that is (IMHO) more complicated (read: easy to get wrong) than the code it's testing or the test itself, and IIRC, we (as a community) generally aggreed to concerto as a testbed data set |
14:00 |
eeevil |
that said, the live test I just updated doe set up a nice mock env |
14:06 |
jeff |
storing auth sessions in memcached is handy because it's fast, and expiry is pretty much taken care of automatically. |
14:06 |
jeff |
it's less-handy because a memcached restart invalidates all sessions. |
14:06 |
jeff |
and in case of a full memcached slab, unexpired auth sessions will be evicted. |
14:07 |
phasefx_ |
I think there's room for independent tests and db-dependent tests.. I think the latter will be easier for most tests |
14:07 |
jeff |
i'm interested in backing sessions with a postgres table, while still using memcached for caching the sessions. if a session isn't found in memcached, we'd consult the database table next before returning an event. |
14:09 |
dbs |
The problem with the db-dependent tests is that as soon as you touch the database, you no longer can have confidence that your test suite is valid |
14:09 |
jeff |
expired sessions could be excluded from retrieval by a WHERE clause, and expired sessions could be routinely purged from the db table by a cron script. updating "last used" time could perhaps be done on a "every X interval" basis rather than on every use. |
14:10 |
jeff |
but i'm getting in the way of testing conversations. sorry. :-) |
14:10 |
dbs |
Which means that if you have a running instance, you have to go through a whole osrf-control --stop-all / eg_db_config --create-schema --load-all-sample / osrf-control --start-all routine for a clean test |
14:10 |
dbs |
I would strongly urge as little dependence on the sample db for unit tests as possible |
14:11 |
dbs |
jeff: we have room for concurrent discussions / monologues here :) |
14:12 |
eeevil |
jeff: if we have sessions in an unlogged table, I'm for it. those are new, of course, which is why tables weren't considered before |
14:13 |
dbs |
Sorry. I spent years with some responsibility for a sample database meant primarily for demonstrating a product that ended up not being able to be touched because our regression team built a ton of tests depending on it. |
14:13 |
eeevil |
dbs: urgings accepted. I'll toss reworking that test onto my pile unless someone else would like to work on a pre-existing test, for practice |
14:14 |
eeevil |
jeff: of course, unlogged tables, being unlogged, would not be replicated, so a db failure would take them out just like a memcache restart. though, in practice, those are /much/ rarer |
14:14 |
jeff |
eeevil: with unlogged tables and streaming replication, wouldn't all sessions cached in memcached lose their corresponding db table entry? |
14:14 |
jeff |
heh |
14:15 |
jeff |
your concern with wanting it unlogged is wal churn / replication traffic overhead? |
14:16 |
jeff |
well, logging the table seems reasonable to me, but AUDITING the table just seems insane. ;-) |
14:16 |
jeff |
s/insane/like a very bad idea/ |
14:17 |
jeff |
and of course, depending on your WAL retention you really can keep it all. ;-) |
14:17 |
phasefx_ |
dbs: yeah, for ease of testing tests, I think independence is something to strive for.For testing production instances of EG, I'm not as interested in that.I'd rather see the barrier to entry for test creation be very low, and since it's not easy to do mock environments with EG's technology stack... :-/ with the vm I want to point ~live at, it'll be restoring to a pristine pre-EG snapshot |
14:17 |
phasefx_ |
every night |
14:17 |
eeevil |
but the other concern is, of course, memcache is orders of magnitude faster than pg |
14:18 |
* eeevil |
runs away for a few |
14:19 |
phasefx_ |
for pgTAP, our life is a bit easier, because of ROLLBACK's, etc. |
14:25 |
jeff |
also, to avoid memcached caching your previous attempt, you might need to use a new search term, or clear your previous search out of memcached (easiest is by restarting memcached, but that does do other things like invalidate auth sessions) |
14:25 |
jeff |
easiest is to add -xyzzy to your terms or something. |
14:25 |
paxed |
i'll restart everything - we're not live after all. |
14:26 |
phasefx_ |
speaking of tests, it looks like my test for closed dates and fine generation caught something (or it could be a faulty assumption within the test itself) |
14:26 |
jeff |
tests++ |
14:27 |
phasefx_ |
maybe something faulty involving DST |
14:27 |
phasefx_ |
DST-- |
14:27 |
paxed |
jeff: right, it required the restart. thanks :) |
15:26 |
Dyrcona |
I also figure why bother when I can just open a text editor, write a query, and run it in the database directly. |
15:26 |
rfrasur |
;-) |
15:27 |
rfrasur |
you probably get cleaner and more efficient information |
15:27 |
* Dyrcona |
deletes yet another email to the dev list about testing before sending it. |
15:27 |
Dyrcona |
No, I get the same information. I just have to navigate Byzantium to get it. |
15:28 |
rfrasur |
:-) |
15:28 |
Dyrcona |
Damn my fingers. |
15:29 |
rfrasur |
lol, that makes more sense |
15:29 |
* Dyrcona |
flips the toast on his laptop and calls it a day. |
15:32 |
* rfrasur |
loves this library's shelving location "wholesome reading" |
15:46 |
Dyrcona |
+1 to rewrite Evergreen to support unit testing! |
15:47 |
eeevil |
Dycona: in java? |
15:48 |
Dyrcona |
Nope, erlang! |
15:48 |
phasefx_ |
I think that's a good choice |
15:49 |
eeevil |
it's already typeless in Perl, so, sure |
15:49 |
Dyrcona |
phasefx_: My +1 was an impromptu response your latest dev list email. |
15:49 |
phasefx_ |
hello_world() -> io:fwrite("hello world\n"). |
15:49 |
Dyrcona |
I agree that most of the testing should use Concerto. |
15:50 |
phasefx_ |
I really feel for the other POV; my instinct is to run tests over and over really quickly |
15:51 |
Dyrcona |
My own instinct is to say tests only test what they test, and when you run a production system your users are far more creative than your testers. |
15:51 |
phasefx_ |
production systems are tests in their own right :) |
16:02 |
|
gsams joined #evergreen |
16:04 |
gsams |
csharp: thanks for the info on reports. I had a feeling it was about like that. |
16:22 |
* rfrasur |
thinks about how to automate ALL her reports using recurrance. |
16:56 |
jeff |
it would be an interesting thing if more libraries used the kinds of retention methods that other institutions use. "haven't seen you in a while, check out some of our newest events/services/items" email / postcard, etc. |
16:57 |
rfrasur |
it's something we hope to implement, but the record keeping has been such a time sink in the past (and, by extension, a money sink) that those kind of things weren't feasible. As we get a little more streamlined and a little smarter in general, we can start doing that. |
16:58 |
Dyrcona |
I wish some of them were a little less efficient. |
16:58 |
jeff |
actually, we have enough patrons and an email service, we could totally run a test or two -- one group of patrons as control. one group of patrons that get an automated email... examine and compare those that saw activity after X weeks of receiving the email vs those that were not sent an email... |
16:58 |
rfrasur |
I mean, we're running a food for fines thing right now that, honestly, we should have looked up every person that it applies to and direct mailed them. |
16:58 |
jeff |
rfrasur: yeah. we did a big amnesty mailing a while back. direct first class mail to all patrons with outstanding items saying "bring 'em back, slate's wiped clean" |
16:58 |
rfrasur |
(our community interaction sucks...but that's my fault now) |
11:53 |
bshum |
We have to either remove that dir or change the ownerships to make it work. |
11:53 |
|
graced_ joined #evergreen |
11:53 |
bshum |
Has anyone seen anything like that on Debian Wheezy or is it just a Ubuntu Precise issue? |
11:56 |
paxed |
i don't think i've encountered that on debian - i run testing. |
11:57 |
bshum |
So maybe it's just a weird Ubuntu issue |
11:57 |
dbs |
bshum: at some point we'll cut over to running Apache as apache / www-data instead of opensrf, and then such errors will go away |
11:57 |
bshum |
dbs: True that |
13:13 |
rfrasur |
Only 7 from Indiana (and none from Hagerstown...sigh) |
13:13 |
graced_ |
collum++ |
13:15 |
csharp |
ah yes... the moment when you're 95% done with a reports template, only to realize that one of the requested fields is not avaialble via your chosen source... |
13:15 |
kmlussier |
Can anybody else confirm this? In testing offline circ in 2.4, we found that the "Skip offline checkout if newer item status changed time" was backwards. When set to false, it skipped the offline checkout. When set to true, it did not skip the offline checkout. |
13:18 |
jcamins |
rfrasur: does your library provide online access to JSTOR? |
13:19 |
rfrasur |
no |
13:19 |
rfrasur |
And it's not available through our statewide db list either. |
13:31 |
jcamins |
He went by Bozhidar, but his actual name was Bogdan Petrovich Gordeev. |
13:31 |
jcamins |
I found a great biography in JSTOR. :) |
13:33 |
jcamins |
It's by Janacek, too, which makes it doubly excellent. |
13:33 |
* rfrasur |
is testing INspire (and all it's Ebsco heavy awesomeness. *cough) |
13:35 |
rfrasur |
well, I found a record, but can't view it because apparently the library doesn't have enough permission to access statewide database stuff. |
13:36 |
csharp |
@blame walled gardens |
13:36 |
pinesol_green |
csharp: walled gardens musta been an Apple employee. |
13:36 |
kmlussier |
Massachusetts uses geolocation to authenticate to statewide databases, which is pretty cool. We don't have JSTOR, though. |
10:11 |
bshum |
I'd have to look closer to see how it renders. |
10:11 |
|
hopkinsju joined #evergreen |
10:12 |
berick |
bshum: oh, i see, i wasn't aware stuff was busted in there. |
10:13 |
bshum |
berick: We used to use field docs, but after our upgrade to newer masters (around 2.3's time), it stopped working. |
10:13 |
bshum |
Nobody missed it enough to really complain yet though. |
10:13 |
bshum |
Plus there's the new YAOUS to show examples next to all the stuff we used to use field docs for showing examples for. |
10:14 |
bshum |
But yeah, I'm not entirely sure the CSS renders right for the comments anymore. |
10:14 |
* bshum |
hasn't tested it lately |
10:18 |
bshum |
if I get a moment later, I'll poke at it some and file a bug if I still see issues. |
10:18 |
* bshum |
wanders off for meetings |
10:38 |
csharp |
bshum++ berick++ # thanks |
10:51 |
|
ningalls joined #evergreen |
10:54 |
csharp |
oh cool - now I see the alert |
14:33 |
gmcharlt |
bshum: I suggesting removing the Google checkout link and adding it to the menu soonish |
14:33 |
gmcharlt |
we can always fine-tune it later |
14:34 |
bshum |
gmcharlt: Sounds like a good plan. I'll go poke at the page now and then get the link up there. Fine-tuning it can go onto the next web team meeting agenda (not that I know when the next meeting is) |
14:37 |
csharp |
I'm beginning to look at the link checker and am looking for sane defaults for the four related library settings |
14:39 |
csharp |
right now I have max redirect attempts set to 4, max wait time for url lookup, 30 secs, number of parallel URLs to lookup: 10, and number of seconds between URL test attempts: 2 |
14:39 |
csharp |
if you're using this, would you mind sharing your settings? |
14:40 |
bshum |
gmcharlt: I'm going to rename that page from "sfc" to "conservancy" since we're not linking to it yet anyways, but also because of the name changes from ages back |
14:41 |
gmcharlt |
sounds good |
14:46 |
|
gsams joined #evergreen |
15:02 |
gmcharlt |
bshum++ |
15:03 |
bshum |
And done. |
15:03 |
RoganH |
So Google Checkout is finally dead (or at least officially dying). |
15:04 |
RoganH |
I've had the tragedy of spending all day taste testing BBQ for the catering for our south east conference. A horrible work day. |
15:05 |
bshum |
RoganH: That just makes me want to book a flight south. |
15:05 |
gmcharlt |
RoganH: we all would have been happy to share some of the load |
15:05 |
gmcharlt |
all you had to do was ask |
09:30 |
jeff |
paxed: that probably wouldn't have any effect on how expert search on 084a works for you now. |
09:30 |
paxed |
jeff: hm. :/ |
09:31 |
bshum |
Expert search is special. |
09:33 |
paxed |
another question: afaics, setting OUS ui.patron.edit.au.*.show to false doesn't actually do what i thought it would do, that is, it still shows the field in patron reg. |
09:34 |
paxed |
at least when it comes to the prefix/title and suffix fields, haven't tested with others. |
09:37 |
bshum |
I thought there were different settings for registration vs. editing |
09:39 |
paxed |
i don't think so. |
09:40 |
paxed |
register.js has orgSettings['ui.patron.edit.' + fmcls + '.' + fmfield + '.show'] |
11:01 |
jeff |
grr |
11:05 |
Dyrcona |
$rid = $rid->id() if (ref($rid)); #easy peasy... |
11:08 |
Dyrcona |
"They're off and running at Rockingham!" |
11:11 |
* tsbere |
finds it very difficult to install and test a router when the internet connection it is supposed to connect to isn't there |
11:11 |
|
mceraso joined #evergreen |
11:16 |
csharp |
could someone test something for me? log into the staff client, search the catalog, click on a record to bring up details, then hover on Add to My List and select an existing list |
11:17 |
csharp |
for me that creates an internal server error and a log message very similar to that reported in bug |
11:17 |
csharp |
bug 1060953 |
11:17 |
pinesol_green |
Launchpad bug 1060953 in Evergreen 2.3 "Internal Server Errors with "large" bookbags in TPAC" (affected: 2, heat: 14) [High,Fix released] https://launchpad.net/bugs/1060953 |
11:17 |
* bshum |
is happily showing how to IRC for mceraso (his new coworker) |
11:17 |
csharp |
mceraso: welcome! |
15:03 |
jeff |
(er, the message before last now) |
15:04 |
jammin |
Yep, got that. |
15:08 |
jammin |
Basically the paypal payflow pro "transparent redirect" looks like a method where the merchant server (our Evergreen) generates the form for the customer's (patron's) browser, but the form posts directly to paypal, and their data never touches our server. Sounds a little bit like Stripe from what I can tell, actually. |
15:09 |
jboyer-isl |
jammin: I've spent some time looking at the CC options, and currently only the last 4 of the card number are stored in the database (though that can be changed by adding a trigger to always null it out, etc.) |
15:09 |
jboyer-isl |
As you mentioned, I don't believe they hit the logs anymore. |
15:12 |
jboyer-isl |
Having worked with the stripe code (but not having time to test it properly, or sign off on it. :,( ) I'd say that's definitely a great option, once it's in. |
15:13 |
csharp |
okay - I'm having a srfsh syntax problem - and I know a lot of it is that I never use JSON for anything so I don't know the syntax... I'm trying 'request open-ils.actor open-ils.actor.user.transaction.fleshed.retrieve.authoritative {"authtoken", "xact_id"}' |
15:13 |
csharp |
the error is "Expected colon after hash key; didn't find it" |
15:13 |
jboyer-isl |
Also, as for the "transparent redirect" form, Stripe doesn't exactly work that way. The sensitive data is transfered by a javascript file downloaded from Stripe, then a token is inserted into your form, and the non-sensitive data is posted to Evergreen to actually do the billing. |
16:27 |
dbwells |
eeevil: I had an older version of MARC::File::XML installed, so I was getting errors in the old SAX.pm, but I just updated to latest, and still get errors, now in XML.pm. Just trying to determine now if the error is in our code, or in MARC::File::XML. |
16:27 |
csharp |
yessir |
16:28 |
eeevil |
dbwells: not a situation where you can replace the whole field? (no subfild list in the rule, I mean) |
16:28 |
bshum |
csharp: Hmm, my offline-blocked-list.pl works fine... |
16:28 |
bshum |
csharp: What version of opensrf are you testing with? |
16:28 |
* Dyrcona |
used the wrong file the other day and got a message, but forgot what it was exactly. |
16:28 |
* dbs |
hopes someone will test out bug #1242999 given the effort he put into rudimentary pgTAP tests |
16:28 |
pinesol_green |
Launchpad bug 1242999 in Evergreen "Encode.pm 2.54 breaks database functions (naco_normalize, maintain_control_numbers, others)" (affected: 1, heat: 6) [High,Confirmed] https://launchpad.net/bugs/1242999 |
16:28 |
csharp |
may be a genasys bug |
16:29 |
dbs |
Given that the community has invested in the QA effort, we should really consider requiring pgTAP tests for db-changing bugs |
16:30 |
bshum |
csharp: Oh, that I couldn't comment on :) |
16:30 |
csharp |
bshum: yeah - it's okay - if I can rule out genasys, I'll be back ;-) |
16:30 |
csharp |
thanks for confirming |
16:30 |
bshum |
csharp: Well, fwiw, I'm running on OpenSRF master and the command to run offline block list is still going without issues |
16:31 |
bshum |
Just finished, no errors. |
16:31 |
Dyrcona |
This reminds me that I had an idea for a branch to make --localhost unnecessary with osrf_control. |
16:31 |
dbwells |
eeevil: I was obviously just turning knobs, so to speak, and seeing what happened. I make no promises for the realistic-ness of my tests :) That said, there are certainly cases where you might want to replace one or two subfields, but you can't necessarily predict whether those will be the *only* subfields. |
16:34 |
eeevil |
dbwells: gotcha. worth an lp bug, then, certainly |
16:37 |
csharp |
duh - it helps if opensrf is running |
16:38 |
csharp |
okay - I'll take that as a sign that I need to quit for the day :-) |
20:18 |
gsams |
I'm her for about 2 more hours myself, down in Texas |
20:18 |
hbrennan |
Ah good. I'm not alone |
20:19 |
gsams |
Let me know if you have any other issues with that, I'm currently creating reports like a mad man down here. |
20:20 |
hbrennan |
I'll be sure to get it tested soon and I'll get back to you to take advantage of that report brain if I need to |
20:28 |
bshum |
hbrennan: I'm pondering your question to the general list about getting some sort of hold report filtered to copy locations and I don't think this would be possible with the staff client reporting. |
20:28 |
bshum |
hbrennan: My thinking is that most holds generally target bib records (title holds), so there isn't an exact and tight match to any given copy. At least not 100% of the time. |
20:29 |
hbrennan |
We usually only have one copy of something so it doesn't really matter |
20:31 |
hbrennan |
right after I get CCNA certified... |
20:31 |
hbrennan |
It looks useful! |
20:31 |
hbrennan |
I just need to find an extra few dozen hours a week for studying |
20:32 |
gsams |
heh, I'm currently studying for CIW database design specialist cert for school right now, I understand trying to find a few dozen extra hours |
20:32 |
gsams |
I'm basically up against the wall for testing at this point |
20:34 |
bshum |
hbrennan: http://pastie.org/8444266 for fun |
20:35 |
bshum |
I just wrote that from memory, so I haven't actually tested it to see how it would fare... might need additional tweaking |
20:35 |
hbrennan |
I am a psychology major, now diving into It |
20:36 |
hbrennan |
I'm 400 pages into Part 1 of the CCNA book |
20:36 |
hbrennan |
Thanks bshum++ |
12:49 |
rfrasur |
Dyrcona: I only have experience with C ++ and it was only a semester...so enough to say I learned something for a month or so...and then never use again. |
12:49 |
rfrasur |
(it was a fun elective though) |
12:50 |
|
Sato`kun joined #evergreen |
12:52 |
bshum |
dbwells: Tested browse with my test server and production data and you were right that it was pretty broken. |
12:52 |
bshum |
Applied those alter functions and things went smoothly after that. |
12:52 |
csharp |
csharp## |
12:53 |
bshum |
I'll do some more testing after lunch. |
12:53 |
rfrasur |
lol, csharp++ |
12:53 |
csharp |
@karma c++ |
12:53 |
pinesol_green |
csharp: c++ has neutral karma. |
14:10 |
csharp |
leonard_: have you checked your logs for messages about holds? |
14:11 |
jeff |
leonard_: holds will not be targeted if the library associated with your staff client session is not considered "open" today in Evergreen. |
14:12 |
jeff |
leonard_: are these new holds, migrated holds, existing holds that were placed in evergreen and not migrated? |
14:13 |
leonard_ |
jeff: I am testing how holds works, I am placing a hold in TPAC, inmediatly I run the hold targeter cron, and then I try Pull List for hold requests |
14:14 |
jeff |
if you look at the patron account, do the holds show up, and do they have a status and/or a value in "current copy"? |
14:14 |
leonard_ |
no, the status is in waiting for copy |
14:15 |
leonard_ |
the currente copy value is No Copy |
15:00 |
|
mjingle joined #evergreen |
15:03 |
|
RoganH joined #evergreen |
15:09 |
bshum |
dbwells++ |
15:23 |
csharp |
okay testing the fix for bug 1244432 - before the fix, I saw some browse queries running for 1.5+ days |
15:23 |
pinesol_green |
Launchpad bug 1244432 in Evergreen "Browse search functions need to be 'stable'" (affected: 1, heat: 8) [Critical,New] https://launchpad.net/bugs/1244432 |
15:24 |
csharp |
now I see 11 seconds - still not great |
15:24 |
csharp |
but that may be may environment |