Search in #evergreen

Results

Result pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148

Results for 2013-11-22

11:05	bshum	senator: I didn't backport bug 1233340 because when I reported the issue, I was only aware of it being unhappy in master/2.5. I think dbwells must have added backport targets after I had done my dance with master.
11:05	pinesol_green	Launchpad bug 1233340 in Evergreen 2.4 "2.5 serials receiving error" (affected: 2, heat: 12) [Undecided,Fix committed] https://launchpad.net/bugs/1233340
11:05	bshum	senator++
11:06	senator	cool. thanks for testing and all in the first place
11:06	bshum	As far as milestones go though
11:06	bshum	I think I'll go ahead and make new ones for you if you don't think the maintainers will repackage.
11:07	bshum	Because otherwise, it could easily get lost in shuffles
11:07	* senator	pings them about it real quick
11:07	bshum	Okay, I'll wait a bit then
11:07	dbwells	Yeah, I looked at it, and really couldn't remember. It seems odd that the series targetting doesn't show up in any of the listed change history. Sorry if I dropped the ball on it.
11:08	senator	no problem. didn't mean to make it a blame-y thing for anybody, just wanted to be sure i still have the right idea of what the test/merge/backport practice is
11:09	eeevil	I have something that I'd like to get into 2.4 sooner rather than later as well, and since we haven't updated the downloads page, or committed the tag versions, we could announce a delay and release early next week.
11:11	bshum	Sounds good to me.
11:11	eeevil	wrapping is a pain, but waiting another month on known bugs is worse
14:57	tsbere	jeff: Ditto that here. <_<
14:57	bshum	jeff: Sounds like what I end up doing every time I start from scratch too :)
14:57	jeff	scipted_installs++
15:05	bshum	berick: For the purposes of testing new bug 1254146 and if we already had an opac entry in actor.org_unit_custom_tree (ours is set to false) what should we expect to see or not see?
15:05	pinesol_green	Launchpad bug 1254146 in Evergreen 2.4 "Custom org tree with no org units causes TPAC server error" (affected: 1, heat: 6) [Undecided,New] https://launchpad.net/bugs/1254146
15:05	bshum	I'm firing up a test VM to try with a blank system
15:06	berick	bshum: if you have a non-active tree or a tree with entries, the bug will not manifest
15:06	bshum	berick: Gotcha.
15:06	bshum	So really I'll test with a clean system then.

Results for 2013-11-21

11:02	pinesol_green	kmlussier: But kmlussier already hates launchpad search!
11:02	kmlussier	Indeed I do.
11:06	csharp	@hates
11:06	pinesol_green	csharp hates dojo_hold_policies_interface; SIP; when libraries purchase third party products without testing and blame Evergreen for it not working; reports; the fact that the Base Filters is unnecessarily greyed out when applying an Aggregate Filter and vice versa; evil; reports more; reports even moar; details; reports even more; and the fact that the Base Filters is unnecessarily greyed (1 more message)
11:06	csharp	@more
11:06	pinesol_green	csharp: out when applying an Aggregate Filter and vice versa even more
11:07	bshum	@hates

Results for 2013-11-20

12:23	gmcharlt	/me offers up https://bugs.launchpad.net/evergreen/+bug/1253163 as both a warning and a plea for folks playing around with Pg 9.3 to comment
12:23	pinesol_green	Launchpad bug 1253163 in Evergreen "authority indexes can fail on Postgres 9.3.0" (affected: 1, heat: 6) [Undecided,New]
12:23		ericar joined #evergreen
12:28	csharp	gmcharlt: I was going to do a PG upgrade from 9.1 to 9.3 on a test server - I'll see if I can recreate
12:28	gmcharlt	csharp++
12:30	dbs	gmcharlt: you got past the xslt problem that dbwells and I were chewing on?
12:31	dbs	https://bugs.launchpad.net/evergreen/+bug/1243023

Results for 2013-11-19

08:36	csharp	I'm pretty sure that's set in PINES - lemme check
08:37	csharp	yes it is
08:37	csharp	hmm
08:37	kmlussier	csharp: I'm pretty sure a message displays. I kept forgetting to turn off that OU setting when testing the holds ratio, so I saw it a few times.
08:37		mmorgan joined #evergreen
08:38	csharp	yeah - that was my original assumption - then I started digging around in the template code and didn't see anything :-/
08:42		ericar joined #evergreen
11:09	jeff	thus, unless there are other overdue circs that are past the grace period, the fine generator will not trigger a calcualtion of system penalties.
11:10	jeff	that's probably what's causing what you're seeing.
11:10	Bmagic	jeff: That is the issue then
11:11	jeff	and it's a bit of an inconsistency, since a manual recalc will add the penalty. we should probably determine if the penalty should count overdues within the grace or not.
11:11	jeff	while this is based on a reading of the code and my current understanding of things, please keep in mind that i haven't tested this theory. :-)
11:12	Bmagic	jeff: right on, sound like an interesting debackle
11:12	Bmagic	jeff: right on, sound like an interesting debacle
11:19	Bmagic	jeff: For now, it sounds like I can use that script that you directed me to on a cronjob? Long term, perhaps the code for fine_generator.pl could be discussed in dev?
12:16		smyers_ joined #evergreen
12:17		dMiller__ joined #evergreen
12:18	jeff	anyone present using non-3M SIP-based self checkout terminals? i'm wondering how other vendors react to "charge privileges denied" in the patron status field.
12:19	jboyer-isl	JCPL was running Bibliotheca machines when I left. If you've some simple steps to test what you're interested in I could pass them on.
12:20	jeff	in evergreen, standing penalties 1 and 2, as well as any standing penalty with a non-empty block list will result in "charge privileges denied" being set to Y.
12:20	jeff	(this is for standing penalties with an org_unit based on the home_ou and ancestors of the patron in question)
12:21	jeff	so, if you define a max items out limit of 40, a patron with 40 items out has a standing penalty which blocks circ.
12:22	jeff	i'm interested in changing the behavior on the evergreen side of things, and would like to better understand how other SIP clients react to "charge privileges denied"
12:23	jeff	so, it would be interesting to know if patrons with the max number of items out can use a bibliotheca terminal, or if they're refused outright.
12:27	eeevil	jeff: sound primarily like the impedance mismatch between SIP2 and ILSen, where the latter has a separate idea of "renewal" but the former does not, and ties renewal to "charge privileges" ... some ILSs lie about the patron's status and deny circs at attempt time, which the EG SIPServer driver could do, I suppose
12:30	jeff	huge impedance mismatch, yes.
12:30	jeff	and eg does reject at attempt time, of course.
12:31	jeff	and (at least in our environment) we do have at least one case where we want to alert when the patron scans their card -- when they're expired.
12:31	jeff	i'm going to reach out to 3M to see if they can provide more details, but thought i'd seek out others with other vendors here. i don't see sal_, but maybe I'll e-mail her.
12:33	jeff	incidentally, there is also a "renewal privileges denied", which... oh hey, which evergreen sets to match the charge privileges denied value. i suppose it's time to test.
12:34	jeff	because if i can make evergreen "do the right thing" with regard to renewals being permitted as long as there isn't a RENEW-blocking ausp, that would be logical and would avoid a new config for "don't report these standing penalties" or similar.
12:35	jeff	semi-related, does anyone know if we've ever populated a csp block value of INET, or is that just something in our db?
12:37	jeff	_bott_: any idea about the INET block type on some standing penalties? was that a GRPL/ME thing?
12:38		linuxhiker1 joined #evergreen
12:39	tsbere	jeff: First I have heard of such a thing
12:45		hopkinsju_ joined #evergreen
14:39	bshum	I occasionally dabble looking at our yaz logs to see what people search against when they look at our z39.50 from EG
14:39	jeff	not a particularly focused or productive question, but something i've been thinking about a little bit.
14:40	tsbere	jeff: I find that answer varies. Including on "who you ask". Some people insist that no logic should be done in the DB other than what is required for things like foreign key relationships and not null constraints and such, after all.
14:41	jeff	sure, there's purists like that. i wasn't asking from that point of view. :-)
14:42	jeff	i suspect that things like in-db ingest are not cpu bound, but haven't tested. it might be.
14:55	mceraso	This may be a silly question, but I'm asking anyway :) Would it be okay to join the merchandising committee for Evergreen? Rogan sent out an email yesterday
14:55	mceraso	Apologies, wrong window :)
14:55		mrpeters joined #evergreen

Results for 2013-11-15

11:22	* dbs	will add something about that to the README
11:26	csharp	--help++
11:27	dbs	okay, --help and --load-all-sample are now documented
11:28	rfrasur	bshum: do you have a quick link to the set of test data?
11:28	pinesol_green	[evergreen\|Dan Scott] Document how to load concerto sample data - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=28915dd>
11:29	bshum	dbs++
11:29	ericar	Angela and I are unable to access the google hangout for the DIG hack. I will be working on the Serials doc that we signed up for.
12:19		mjingle joined #evergreen
12:20	gmcharlt	rfrasur: can we help?
12:20	dbs	to just generate html, asciidoc on windows only needs python; you only need cygwin if you're doing pdf / epub etc
12:20	bshum	gmcharlt: We're testing it.
12:20	rfrasur	I'm not sure :-), Ben is taking a look.
12:20	* Dyrcona	thought Windows was just for games.
12:21	csharp	dbs: oh - good to know
13:40		stevenyvr2 joined #evergreen
13:40	eeevil	dconnor: reducing queries is great. there are already several mechanisms for running arbitrary queries that don't suffer from injection attacks
13:47		RoganH_ joined #evergreen
13:56	jeff	dconnor: going back to my earlier suggestion -- a json query lets you construct a query at the perl layer that returns fieldmapper objects that you can then use without further need to parse/map them. i can't guarantee that you can do everything in a json query that you are doing in your asset.get_holdings_maintenance_page function, but at first glance I think it's do-able.
13:56	jeff	might be worth testing and comparing performance / complexity.
13:56		akilsdonk_ joined #evergreen
13:58	dconnor	jeff: Ok, I'll check it. thanks!
13:59		rjackson-isl joined #evergreen
14:11	eeevil	I don't think you can use filter syntax for classes, no
14:12	paxed	the requirements list we went through in the meeting had something like 1600 requirements... some were pretty interesting ones, which our current ils can do
14:12	paxed	the mathematical comparisons was one
14:13	kmlussier	rfrasur: I used to use http://andrewk.webfactional.com/asciidoc.php to do simple testing of my asciidoc. But it's deactivated now. :(
14:13	eeevil	paxed: checked, and you can't use the alias trick today ... adding filter aliases would be another not-too-hard thing
14:13	rfrasur	kmlussier++
14:13	rfrasur	gist.github is working pretty well.
14:14	rfrasur	I'm just testing snippets here and there.
14:27		alexlazar left #evergreen
14:34		kbutler joined #evergreen
14:37		rangi joined #evergreen
14:37		berick joined #evergreen
14:38		_bott_ joined #evergreen
14:38		rfrasur joined #evergreen
14:41	rfrasur	Btw, I can use both 2.2 and 2.5 on this computer. I just need to make sure that I'm using the right icon and change the servers. I forgot that'd worked before.
14:46		kbeswick joined #evergreen
14:48	rfrasur	o0(github is friendly. They sent me a welcome email with a dude and a guitar.)
14:48	bshum	dbs: The link to the asciidoc for "Introducing SQL to Evergreen administrators, round two" on your site isn't working.
14:48	bshum	It seems to be linked to the older epub file and not the raw text
14:49	bshum	Someone expressed interest in seeing v2 merged back into the core docs.
15:22	bshum	Or maybe it's label_sortkey
15:23	* bshum	always gets his variable wording mixed up
15:26		bwicksall joined #evergreen
15:36	yboston	what is a good combo of options to test parse the full docs. I usually use asciidoc -a data-uri -a icons -a toc -b html5 root.txt
15:36	yboston	but I am getting an error
15:36	yboston	asciidoc: WARNING: RELEASE_NOTES_2_5.txt: line 163: filter non-zero exit code: source-highlight -f xhtml -s bash: returned 127
15:36	yboston	asciidoc: WARNING: RELEASE_NOTES_2_5.txt: line 163: no output from filter: source-highlight -f xhtml -s bash
15:40		kmlussier joined #evergreen
15:42	mmorgan	Can someone help me understand user activity types?
15:42	mmorgan	"OPAC Login (tpac)" I get, but what is "Login via opensrf"? The users I'm looking at aren't staff.
15:54	senator	yboston: i would guess you just don't have the source-highlight package installed
15:56		kbeswick joined #evergreen
15:57	kmlussier	yboston: I don't have the source-highlight package installed and get a similar error. But it isn't a problem once the documents are added to the official repository. I've just learned to ignore them.
15:58	yboston	senator: that is what I suspected, but I still wodered what parameters were recomended, since I am also having issues with the "icons" parameter. I am trying to figure out how to install this filter
15:58	yboston	kmlussier: I just don't want to commit something that breaks the asciidoc build, so I wnated to test it first, and so far it won't build
15:59	kmlussier	yboston: Oh, then that's different. I can usually get it to build even with the source-highlight package warnings.
15:59	mmorgan	kmlussier: Looks like there are a number of users in the syrup tables for those students, so that could certainly be part of it. Thanks!
16:00	* kmlussier	heads out for her long commute.
17:03	jeff	hooray.
17:15		mmorgan left #evergreen
17:20	phasefx_	jeff: lp still sucks ;)
17:20	jeff	for those working with postgresql 9.3 and perl DBI, you'll likely hit this failed test at install time for DBD::Pg, known and safe to --force past: https://rt.cpan.org/Public/Bug/Display.html?id=88865
17:32		linuxhiker joined #evergreen
17:35	paxed	hmm ... the comments on get_ac_key in EGCatLoader/Record.pm seem to indicate it could be removed now?
17:37	berick	paxed: yes! good eyes. we can kill that now

Results for 2013-11-14

14:33		Phenol joined #evergreen
14:38		Polonel joined #evergreen
14:46		mjingle joined #evergreen
14:49	Dyrcona	Is there a minimum version of pgTAP that we should be using to run the tests?
14:50	Dyrcona	I ask because tsbere installed from source and the tests work for him.
14:51	Dyrcona	I installed from a packaged repo, probably Ubuntu's for 12.04, and the tests fail because I don't the isnt_empty() function.
14:51		ericar joined #evergreen
14:53	tsbere	I am assuming 0.93
14:53	tsbere	I could be wrong
14:53	Dyrcona	You're probably right. I have 0.90.
14:53	jeff	0.23 2009-12-18T23:03:39Z
14:53	jeff	* Added `is_empty()` to test that a query returns an empty set (no results).
14:53	jeff	oh.
14:53	jeff	isnt_empty
14:54	Dyrcona	Yep. the latter.
14:54	jeff	0.92.0 2013-01-16T00:41:07Z
14:54	jeff	* Added `isnt_empty()` to ensure that a query does not return the empty set.
15:00	csharp	jeff: once I'm out of the woods of our 2.5 upgrade (scheduled
15:00	kmlussier	jeff++ :)
15:00	jeff	csharp: step 1: log in to https://eg.example.com/jasperreports/ ;-)
15:00	csharp	...MLK weekend), I plan to set up jasperreports for PINES in a test instance
15:01	* csharp	would love that
15:01	jeff	csharp: great to hear!
15:01	jeff	(that you'll be experimenting)
15:01	csharp	yeah
15:37	paxed	dbwells: yeah, i just read the email
15:43	eeevil	dbwells: do you think you'll have a chance to look at the branch on https://bugs.launchpad.net/evergreen/+bug/1251353 soon?
15:43	pinesol_green	Launchpad bug 1251353 in Evergreen 2.4 "QP query generation is incorrect with two or more bool ops in a row" (affected: 1, heat: 6) [Undecided,New]
15:50	dbwells	eeevil: thanks for working on that bug. I have looked at the branch, but I have little hope of grasping what it is actually doing in the time I have for it right now. My plan is to push it into production and test it empirically, but not until after 5:00pm (in order to limit civilian casualties).
15:52	eeevil	dbwells: cool, and thanks
16:01		kbutler joined #evergreen
16:03	bshum	Callender: Can you describe the situation you encountered that led to bug 1251424 ?
16:05	Callender	but for some libraries, that button would still be greyed out and you couldn't make a hold for yourself
16:05	bshum	Callender: In the staff client, that's expected I would think... unless you click the radio button to say "Place this hold for me (insert my name here)"
16:05	Callender	but if you first clicked the radio button for making it for another patron, and then clicked back to yourself, without ever even entering anything, then the place hold would open up
16:05	bshum	Or you keyed in a barcode that actually worked
16:06	* bshum	tests
16:06	Callender	it doesn't always do that though.. my test system and other systems I tested, you can go in and place hold for yourself just fine
16:06	bshum	Oh I see what you're saying now.
16:06	Callender	it was just certain circumstances that I couldn't re-create.. but it always happened for certain libraries.. it was a very weird bug
16:07	Callender	that change I made on launchpad, that fixed it across the board on systems I tested it on, and didn't seem to harm systems that weren't broken. I need someone though familiar with the code to kind of double check what I'm doing there
16:09	bshum	Yeah, I was able to reproduce the issue now.
16:10	bshum	The radio button starts with the placing hold for someone else.
16:10	bshum	But if you click on the option for placing a hold for the staff account itself
16:10	bshum	Sometimes it wouldn't change the color right away
16:10	bshum	Callender++ # the change works as you described.
16:10	Callender	ok, I'm glad you were able to see it.. beacuse this one drove me crazy for a little while :)
16:10	bshum	If memory serves, this probably affects 2.4 and 2.5 only
16:15	bshum	Your description confused me at first
16:15	bshum	Cause I thought maybe your change made the submit button always available again.
16:15	bshum	Which it shouldn't be.
16:15	bshum	But testing seems to bear out no issues on my end so far.
16:16	Callender	ahh ya, I see what you're saying. It wasn't so easy for me to describe
16:18	bshum	I'll tweak the commit message a bit to be more concise on what we're doing here.
16:18	bshum	Then I'll commit it through
17:39		stevenyvr2 left #evergreen
17:44	dbs	alas, "[source,conf]" in the 2.5 release notes appears to be breaking PDF output?
18:15		BigRig_ joined #evergreen
18:22	pinesol_green	[evergreen\|Dan Scott] Add basic docs for testing with pgTAP - <http://git.evergreen-ils.org/?p=Evergreen.git;a=commit;h=3c86315>
18:26		edoceo joined #evergreen
19:06		hopkinsju joined #evergreen
19:06		hopkinsju left #evergreen

Results for 2013-11-13

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

Results for 2013-11-12

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	...

Results for 2013-11-11

07:44	csharp	@later tell hopkinsju from my investigation of holds, EG gathers a list of "potential" copies, and, all things being equal, selects the first one it finds
07:44	pinesol_green	csharp: The operation succeeded.
08:23	csharp	@hates
08:23	pinesol_green	csharp hates dojo_hold_policies_interface; SIP; when libraries purchase third party products without testing and blame Evergreen for it not working; reports; the fact that the Base Filters is unnecessarily greyed out when applying an Aggregate Filter and vice versa; evil; reports more; reports even moar; details; and reports even more
08:24	csharp	@hate the fact that the Base Filters is unnecessarily greyed out when applying an Aggregate Filter and vice versa even more
08:24	pinesol_green	csharp: The operation succeeded. csharp hates the fact that the Base Filters is unnecessarily greyed out when applying an Aggregate Filter and vice versa even more.
08:25		mrpeters joined #evergreen

Results for 2013-11-08

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

Results for 2013-11-07

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

Results for 2013-11-06

14:29	Dyrcona	Well, corroborating what dbs said: I created a table with a timestamp column and a plain text column. When I try to insert those "truncated" timestamps into the column, psql reports a syntax error.
14:33	Dyrcona	excel--
14:34	bshum	Well, I've had some really weird junk in my postgres timestamps before. I call that the curse of strange migrated data. Still not entirely sure how they ended up so screwy.
14:35	csharp	mmorgan: ah - interesting - thanks for testing
14:35	csharp	yeah my working assumption is that the SQL-to-spreadsheet transfer may be the culprit
14:36	csharp	they really need a competent support vendor
14:41	* Dyrcona	has a hunch who that is and a clue who the library might be, but isn't sure.
14:43	kmlussier	I'm thinking 2 p.m. Eastern Tuesday is probably the best bet for a dev meeting. http://doodle.com/n9avnst6nvs2342g
14:48		smyers__ joined #evergreen

Results for 2013-11-05

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)