2009-12-02T00:31:17 *** atz has quit IRC 2009-12-02T00:31:56 *** atz has joined #evergreen 2009-12-02T00:32:56 *** scott_r has joined #Evergreen 2009-12-02T00:35:01 *** [1]atz has joined #evergreen 2009-12-02T00:35:17 Evening. Anyone available for giving a few pointers? I'm attempting to add new properties to existing copies, without much success... 2009-12-02T00:42:01 *** atz has quit IRC 2009-12-02T00:42:02 *** [1]atz is now known as atz 2009-12-02T00:48:19 scott_r: hey, what sort of properties? 2009-12-02T00:50:40 Specifically, I'm trying to add a 'discount price'. I know the statistics column in the copy editor allows for custom fields (using dropdown boxes), however for this field (and possible future ones) I'm trying to support custom fields using a regular textinput box 2009-12-02T00:52:19 I could not find built-in support for that. I tried overriding/replacing some of the functionality used for the custom combobox fields, only to run into some stringent type checking server-side. Which likely indicates the DB wouldn't have supported it anyway. 2009-12-02T00:52:43 I think long term we're going to need to provide an option for uncontrolled stat cats for copies like we do for patrons. One option for you now is to add a dedicated field, but that's pretty heavy 2009-12-02T00:53:02 Yeah, I was just looking into that, and it seemed a bit much 2009-12-02T00:53:10 There are no documented instructions on that, eh? 2009-12-02T00:54:27 no. You'd need to edit a database schema file, create a database upgrade script, tweak the fm_IDL.xml file, tweak server/circ/util.js, and server/cat/copy_editor.js 2009-12-02T00:55:33 Yeah, well at least I was on the right track then. Was hoping I was making it harder than it needed to be though. 2009-12-02T00:55:34 upgrade script assuming you were going to submit a patch and get this into trunk 2009-12-02T00:56:31 a more quick & dirty but less "robust" way of doing this would be to use a copy note 2009-12-02T00:57:19 Hmm, where do the copy notes show up within the client? 2009-12-02T00:57:50 I've spent much more time looking through the code, than actually toying around with the client itself... 2009-12-02T00:58:32 they're very de-emphasized/buried. The push was to get folks to move away from abusing notes and use stat cats instead. You see them in the OPAC if you drill down to the item details, and you can also get to them through the Item Attribute Editor 2009-12-02T00:59:28 I've been thinking about making them more prominent in the Copy Editor and the Copy Details interfaces 2009-12-02T01:00:25 freetext notes tend to be troublesome because there's no forced consistency, and folks inevitably want to report on them and run head-on to inconsistencies and data entry errors 2009-12-02T01:00:27 Ah yes, found them. That may be a hack solution for the moment. However, We're also hoping to have the information in newly added fields be more prominent/accessible in other windows as well. I imagine that would probably be a bit dirtier using the copy notes vs a stat cat? 2009-12-02T01:01:45 eg. columns in the item status window, etc 2009-12-02T01:02:08 which I don't believe is currently supported with any stat cat. Would that be correct? 2009-12-02T01:03:37 for item status, it'd definitely be easier to go with a true/dedicated field. The thing about notes and stat cats is that you can have many of them for any given item, so they don't lend themselves to lists like that 2009-12-02T01:05:55 though you could create a virtual field for items and have the middle layer populate it with data from other tables (like notes or stat cats), but since stat cat definitions can be changed by end-user staff with the right privileges, and such code would have to be hard-coded, that's not really a good strategy 2009-12-02T01:07:10 I think there may already be development involving "discount price" though, at least in the realm of acquisitions. I don't know how all that works, alas 2009-12-02T01:07:50 I think the acq system has a flexible way of creating arbitrary attributes/fields for various objects 2009-12-02T01:08:55 True. Alright, I'll attempt the process of creating a stat cat proper, and document that. Probably the best balance between speed/efficiency. That may be good enough, and then I can manually add fields as needed. If I can get a sign-off, I'll continue looking into supporting the custom text. 2009-12-02T01:09:44 Thanks for the help! If I get to delve into supporting that functionality, I'll likely be back sooner than later. 2009-12-02T01:09:55 well, stat cat has a major disadvantage in this case for copies, and that's it can only use defined entries. Depending on how many entries you'd have to define for your dataset, it may not be viable 2009-12-02T01:10:19 couldn't have $0 to $infinity in a drop-down menu, for example 2009-12-02T01:10:55 oh my bad. Meant a dedicated field. 2009-12-02T01:11:07 could may have a stat cat providing percentages, though. Percentage applied to the dedicated price field 2009-12-02T01:11:45 That could be an interesting implentation. I'll have to find out exactly how our discount prices are to be calculated... 2009-12-02T01:12:19 may be worth shooting an email to the list too 2009-12-02T01:12:35 see what the acq folks are doing 2009-12-02T01:13:36 Probably a good idea. Either way, I'd also like to see some instructions on the wiki for adding dedicated fields. 2009-12-02T01:14:47 I'll throw something together on that once I've got a field successfully added 2009-12-02T01:14:57 the main problem I forsee there is surviving upgrades if such changes don't make it into the source 2009-12-02T01:15:00 scott_r++ 2009-12-02T01:16:33 True. Always something in the way... 2009-12-02T01:17:53 just need more brainz 2009-12-02T01:18:06 Hopefully I can get a few of my changes into the source yet. Waiting on testing a few of them on our live server. Then I'll see what I can submitted. 2009-12-02T01:18:22 Hah, brainz and time/money :) 2009-12-02T01:19:10 *** levani_el has joined #evergreen 2009-12-02T01:19:12 cool deal. I'd definitely pick the brains of another developer or two before you invest much time in it. I'm the quick&dirty&horrify-others developer :) 2009-12-02T01:19:23 *** levani_el has quit IRC 2009-12-02T01:20:13 Yeah, I should do that. I've only been working on Evergreen for a short time, and I'm sure I've taken a shortcut or two. Unintentionally, of coures :D 2009-12-02T01:20:59 that said, a dedicated field isn't really hard to figure out.. my only concern would be resistance to a proliferation of such fields if they really should be something more local. My bet is enhancing stat cats to be controlled or uncontrolled for copies is the best bet in the long run 2009-12-02T01:22:26 Yep, I definitely agree. Of course my library wants this fast tracked, and I'm only part-time, so I'll have to take the short-term fix to start 2009-12-02T01:23:01 well, the data will be easy to migrate however you do it :) could turn notes into stat cats later, or dedicated fields, etc. 2009-12-02T01:23:43 Exactly. The benefit of the digital age. :) 2009-12-02T01:23:52 could even create child tables that inherit from asset.copy and are exposed to the reporting system via fm_IDL.xml 2009-12-02T01:24:27 ah, hadn't thought of that yet. 2009-12-02T01:25:04 *** levani_el has joined #evergreen 2009-12-02T01:25:21 hello 2009-12-02T01:25:40 how run LONG_MAX.diff ? this is a bug solve 2009-12-02T01:25:42 wouldn't be easy to get that data to display elsewhere like that, but it'd be very self-contained and easy to manipulate later 2009-12-02T01:26:09 levani_el: the LONG_MAX bug with perl 5.10 is cosmetic. You can safely ignore it. It's harmless 2009-12-02T01:26:24 Indeed, and getting the data to display elsewhere wasn't a huge need at the moment anyway. 2009-12-02T01:26:39 where found it to ignor? 2009-12-02T01:27:11 levani_el: just mentally ignore it 2009-12-02T01:27:15 --- /usr/lib/perl/5.10/_h2ph_pre.ph.old 2008-12-05 16:22:43.000000000 -0500 +++ /usr/lib/perl/5.10/_h2ph_pre.ph 2008-12-05 16:22:49.000000000 -0500 2009-12-02T01:29:03 Well, that's definitely given me some options to think about. I'll have to hang around this channel more often. Sure beats all the guessing I was doing, even though they were good guesses :). 2009-12-02T01:29:03 delete this and save? 2009-12-02T01:29:08 # gross hack unless (defined &__LONG_MAX__) { sub __LONG_MAX__ { 2147483647 } } 2009-12-02T01:29:46 Time to give my eyes a break though, thanks again phasefx! I'll be back with updates at some point soon I'm sure. Night all! 2009-12-02T01:29:52 levani_el: oh, I don't know about that. You should be able to use your original file and not mess with it. The only bug is spurious warnings to console 2009-12-02T01:30:09 scott_r: nighters 2009-12-02T01:30:10 *** pmplett has quit IRC 2009-12-02T01:30:30 *** scott_r has left #Evergreen 2009-12-02T01:45:12 ooh... someone else with a need for a dollar value per copy other than "price" 2009-12-02T01:46:09 delete this lines and don`t get error 2009-12-02T01:46:20 byt get error in autogen Exception: OpenSRF::EX::ERROR 2009-12-01T22:43:41 OpenSRF::AppRequest /usr/local/share/perl/5.10.0/OpenSRF/AppSession.pm:1012 System ERROR: Exception: OpenSRF::DomainObject::oilsMethodException 2009-12-01T22:43:41 OpenSRF::AppRequest /usr/local/share/perl/5.10.0/OpenSRF/AppSession.pm:1012 <500> Severe query error -- see error log for more details 2009-12-02T01:51:52 error login in staff client? 2009-12-02T02:00:03 *** levani_el has quit IRC 2009-12-02T03:28:04 *** eguest309 has joined #Evergreen 2009-12-02T06:17:54 *** eguest309_ has joined #Evergreen 2009-12-02T06:17:54 *** eguest309 has quit IRC 2009-12-02T06:17:56 *** eguest309_ is now known as eguest309 2009-12-02T06:34:16 *** eguest309 has left #Evergreen 2009-12-02T07:51:23 *** mck9 has joined #evergreen 2009-12-02T08:04:04 *** rickd has quit IRC 2009-12-02T08:34:08 mrpeters-isl pasted "loans.sql" at http://paste.lisp.org/display/91429 2009-12-02T08:52:45 *** jenny has joined #evergreen 2009-12-02T09:03:47 *** Meliss has joined #Evergreen 2009-12-02T09:13:01 *** bshum has joined #evergreen 2009-12-02T09:17:31 *** dbs has joined #evergreen 2009-12-02T09:29:14 *** phase_bb has joined #evergreen 2009-12-02T09:34:23 *** bshum has quit IRC 2009-12-02T09:36:28 *** bshum_ has joined #evergreen 2009-12-02T09:37:07 *** bshum_ has quit IRC 2009-12-02T09:37:27 *** bshum_ has joined #evergreen 2009-12-02T09:38:18 *** bshum_ has left #evergreen 2009-12-02T09:39:13 *** bshum_ has joined #evergreen 2009-12-02T09:41:48 *** bshum_ is now known as bshum 2009-12-02T09:45:15 *** atz has quit IRC 2009-12-02T09:45:58 *** atz has joined #evergreen 2009-12-02T09:47:43 Has anyone written anything about their experience or steps in upgrading from postgres 8.2 to 8.3 or beyond? Or is that an unnecessary step when moving from EG 1.4 to 1.6? 2009-12-02T09:48:17 *** atz has quit IRC 2009-12-02T09:48:57 *** atz has joined #evergreen 2009-12-02T09:50:41 *** matt__ has joined #evergreen 2009-12-02T09:54:06 *** mrpeters-isl has quit IRC 2009-12-02T09:54:10 *** mrpeters-isl has joined #evergreen 2009-12-02T09:57:17 *** atz has quit IRC 2009-12-02T09:57:58 *** atz has joined #evergreen 2009-12-02T10:09:44 *** r123 has joined #evergreen 2009-12-02T10:18:42 random question: In the Z30.50 screen when I have a lot of results, my screen displays 10 rows of results initially, and when I click Fetch More Results, it fetches 10 at a time. In some other locations it's as few as 2. Is this configurable somewhere, or is this number set based on something like screen resolution? 2009-12-02T10:23:10 mrpeters-isl: there's a hard coded limit in the source (currently set to 10). That number is divided by the number of active/selected z-targets, and that new number is sent as a limit along with the query to each service 2009-12-02T10:23:30 *** [1]atz has joined #evergreen 2009-12-02T10:23:54 so if you have two targets selected, each will get limit: 5 as part of the underlying search query, and return up to 5 results 2009-12-02T10:24:56 bshum: upgrading from postgresql 8.2 to 8.3 isn't a necessary part of evergreen 1.4 to 1.6 2009-12-02T10:25:15 phasefx_ so you are saying its impossible for someone to only see 2 records at a time? 2009-12-02T10:25:17 bshum: however, I believe branflakes (brandon uhlmann) from BC has been working on that 2009-12-02T10:25:21 dbs: Thanks, I'll bear that in mind when I'm working on it later today. 2009-12-02T10:25:47 dbs: I've been trying to migrate our test server materials from postgres 8.2 to 8.3 but haven't been able to get my metarecord searching working for some reason. 2009-12-02T10:25:47 mrpeters-isl: if only one of the 5 targets matches a given query, then I would expect only 2 records 2009-12-02T10:26:06 ahh true 2009-12-02T10:26:32 mrpeters-isl: it's not particularly intuitive, but it is what it is :) 2009-12-02T10:26:46 mrpeters-isl: you can change the limit from 10 to 2, but if you have more than 2 sources selected, each will potentially return 1 result 2009-12-02T10:27:33 no no, we dont want to limit it...was just looking for explination as to why it appeared different on different machines 2009-12-02T10:27:50 ah 2009-12-02T10:28:08 on another note, yahoo.com is now blocking all emails from Evergreen Indiana! horray! 2009-12-02T10:28:31 they did that to other Evergreen sites as well recently 2009-12-02T10:28:37 we're such spammers, you know, since the patrons providing their emails is opting in to it! 2009-12-02T10:28:48 bradl: good to know! 2009-12-02T10:29:05 sent a note, hopefully they will respond/resolve it 2009-12-02T10:29:17 you'll need to provide yahoo with a privacy policy for your site 2009-12-02T10:29:17 sboyette: I think the opensrf bug wranglers should be a sub-team of the evergreen bug wranglers, rather than vice-versa - not that it matters too much right now 2009-12-02T10:29:24 I hope your experience is better than ours :) 2009-12-02T10:29:29 uh oh! 2009-12-02T10:29:32 mrpeters-isl bradl: that's really good to know 2009-12-02T10:29:50 pita 2009-12-02T10:30:13 I really enjoyed having email sent from my denials@gmail.com address blocked by hotmail.com 2009-12-02T10:30:24 bread, of course 2009-12-02T10:30:42 dbs: i was just trying to get things linked such that bugs could actually start being filed in either place. i'll gladly hand the group over and let you do whatever you want, if you want 2009-12-02T10:31:00 sboyette: no, I understand, it's expedient 2009-12-02T10:31:24 bradl: how did you eventually get it resolved? 2009-12-02T10:32:45 or, have you? 2009-12-02T10:33:53 we posted a privacy notice in an orphaned page 2009-12-02T10:34:12 assured them we weren't spammers... and eventually the email started to flow again 2009-12-02T10:34:28 interesting 2009-12-02T10:34:52 any chance we could have a look at your privacy notice as a base in drafting one for Indiana? 2009-12-02T10:34:59 just to see the items that Yahoo apparently wants to see 2009-12-02T10:35:20 http://www.esilibrary.com/esi/privacy.php 2009-12-02T10:35:46 thanks 2009-12-02T10:39:46 *** atz has quit IRC 2009-12-02T10:39:47 *** [1]atz is now known as atz 2009-12-02T11:09:28 *** eguest309_ has joined #Evergreen 2009-12-02T11:17:20 *** brendan_ga_ has quit IRC 2009-12-02T11:24:28 *** [1]atz has joined #evergreen 2009-12-02T11:42:49 *** atz has quit IRC 2009-12-02T11:42:50 *** [1]atz is now known as atz 2009-12-02T11:47:40 *** brendan_ga has joined #evergreen 2009-12-02T11:51:06 *** mrpeters-isl has quit IRC 2009-12-02T11:51:11 *** mrpeters-isl has joined #evergreen 2009-12-02T11:53:28 *** [1]atz has joined #evergreen 2009-12-02T11:57:45 *** Stompro has joined #evergreen 2009-12-02T12:11:34 *** atz has quit IRC 2009-12-02T12:11:34 *** [1]atz is now known as atz 2009-12-02T12:12:20 mrpeters-isl pasted "hold.sql" at http://paste.lisp.org/display/91444 2009-12-02T12:25:52 *** jmeeuwen has quit IRC 2009-12-02T12:34:23 *** rickd has joined #evergreen 2009-12-02T12:41:29 hmm. "can we prevent a patron from placing a hold on a book they already have checked out?" 2009-12-02T12:46:05 * _bott_ looks at dbs and tilts my head like a confused dog 2009-12-02T12:46:06 there were recent commits along the lines of preventing hold on an available item... shouldn't be too difficult... don't know if that would be in the permit scripts or if it would involve "actual" code changes 2009-12-02T12:46:08 dunno... I don't see why you would. 2009-12-02T12:46:40 as for use-case, we've had patrons place holds on all 18 overdue items they have checked out, then return them and they get placed right back on the holdshelf for that patron (assuming no other holds) 2009-12-02T12:46:42 the person is just getting "back in line"... it doesn't mean they have a stranglehold on the item 2009-12-02T12:47:28 so your procedure would otherwise make them wait in the library for the item to get reshelved, but otherwise accomplish the exact same outcome (user gets the books) 2009-12-02T12:47:34 a patron doing this forces other patrons to place a hold in order to get the items in question. in some environments, that's not an issue. others with more of a "browsing library" focus care greatly 2009-12-02T12:47:50 atz: yeah, I think the idea is that most people aren't savvy enough to place a hold in the first place. so evil dude gets the checkout, then places a hold to be first in line to get it again 2009-12-02T12:48:12 dbs: and while he's at it, a metarecord hold to get in line twice. 2009-12-02T12:48:19 jeff: hah, you're evil :) 2009-12-02T12:49:01 but they aren't proposing to prevent the same user from getting the just-returned item off the shelf and checking it back out 2009-12-02T12:49:02 <_bott_> You could prevent it if they're the only hold by using fetchBestHold() and comparing hold.usr and patron.id ...but that only works if they'd already be next in line. 2009-12-02T12:49:41 guess you could add an adjustment to push the current holder of an item to the bottom of a holds queue for the same item 2009-12-02T12:50:09 that would still get the user the item, dbs 2009-12-02T12:50:12 so if anyone else places a hold subsequently, they immediately usurp evil dude 2009-12-02T12:50:15 when no other holds exist 2009-12-02T12:50:45 atz: right, the idea is to try to enable the item to get into other hands 2009-12-02T12:51:05 teach other users to place holds? 2009-12-02T12:51:10 if no other hands want the item, then it can immediately go back to evil dude 2009-12-02T12:51:44 atz: thanks for showing such understanding. it really makes my day. 2009-12-02T12:51:54 heh, sorry 2009-12-02T12:52:58 yeah. it's a bit of an outlier, most people aren't used to using holds, so in that environment the savvier user gets an edge for hotly contested items 2009-12-02T12:52:59 i mean the mechanism is there and available to everybody, with priority everybody can agree on 2009-12-02T12:53:37 Apparently not everybody agrees on the priority, eh? 2009-12-02T12:54:02 ... if they filed a hold, they would agree that first come first served is fine (enough) 2009-12-02T12:54:43 i don't see why the user that knows the precise shelving location and when the afternoon reshelving round is scheduled should get the advantage... 2009-12-02T12:54:55 especially when the system does support holds 2009-12-02T12:55:42 is there a max number of holds ? 2009-12-02T12:57:00 there's a bit of a further complication, I suppose, in that we've effectively disabled placing holds on available items. So the holds race starts once the item is first checked out. 2009-12-02T12:57:20 and guess who has the advantage in that contest. 2009-12-02T12:58:02 ah, yeah that tilts the scale to one who knows the moment-of-checkout 2009-12-02T12:58:38 So, the more I think about it, the better I like the philosophy of not preventing holds by the current borrower, but ensuring that they move to the back of the line (if there is any line) 2009-12-02T12:59:29 that way, they get the item for one borrowing period, and it becomes available to another borrower - and, if there are a number of holds piling up, we can move it to a reserve shelf or a shorter borrowing period until things cool down 2009-12-02T13:00:24 It still requires other people to place holds to get in line, so you should still be happy atz :) 2009-12-02T13:01:18 hey, not my dog/horse/fight... whatever works for y'all 2009-12-02T13:04:39 (all assuming no other holds) if current patron places a hold, at checkin the book gets captured for the hold. if current patron is not permitted to place a hold, patron must wait for item to be checked in and reshelved, then must be first to hands-on the book. 2009-12-02T13:08:14 jeff: right, although in our library if there are no holds, then the staff would just sign the item right back out to the borrower. So I concur with atz that other borrowers need to place holds, damnit. 2009-12-02T13:08:51 I just think that there needs to be some queue manipulation in our case, due to not allowing holds on available items 2009-12-02T13:09:16 s/other borrowers/other would-be borrowers/ 2009-12-02T13:09:20 dbs: where is your current dev tutorial living? 2009-12-02T13:09:54 jeff: still at http://evergreen-ils.org/~denials/ as it hasn't really changed since October 2009-12-02T13:10:09 thanks 2009-12-02T13:11:24 guess I could add a link to it from http://evergreen-ils.org/dokuwiki/doku.php?id=community_tutorials if it's going to live there for a while 2009-12-02T13:12:38 * dbs does so 2009-12-02T13:12:41 * dbs goes to get food 2009-12-02T13:39:14 so ... massive leak in my brand new office. may be evacuating perfect timing, eh? 2009-12-02T13:41:16 miker_: I'm sure it can be fixed in less than 20 minutes 2009-12-02T13:45:14 i've almost got eg working! :) I roughly followed the 1.4.0 install page with a copy from svn, and it starts up...a few issues though... 2009-12-02T13:45:49 *** pmplett has joined #evergreen 2009-12-02T13:46:45 settings-tester.pl tells me to copy oils_web.xml.example to oils_web.xml in /openils/conf...i have, but it still complains 2009-12-02T13:48:44 *** djfiander has joined #Evergreen 2009-12-02T13:50:36 *** wlayton has joined #evergreen 2009-12-02T13:56:26 * phasefx_ really dislikes high latency net connections 2009-12-02T13:56:32 *** pmplett is now known as pmp_coffee 2009-12-02T13:57:59 miker_: hopefully before you moved everything into it? 2009-12-02T13:58:12 *tap* is this thing on 2009-12-02T13:59:04 *** jenny1 has joined #evergreen 2009-12-02T13:59:28 phasefx_: cable or dsl? 2009-12-02T14:00:08 blackberry 2009-12-02T14:00:26 phasefx_: no, i mean, why are you even having to use your BB? ;) 2009-12-02T14:00:29 what died 2009-12-02T14:00:32 haha, cable 2009-12-02T14:00:47 *** ChanServ sets mode: +o dbs 2009-12-02T14:00:49 * berick thinks it might be time for dsl ;) 2009-12-02T14:00:50 *** dbs changes topic to "Meeting agenda: http://evergreen-ils.org/dokuwiki/doku.php?id=dev:meetings:2009-12 | Welcome to the #Evergreen library system channel! | We are publicly logged. | Large pastes at http://paste.lisp.org/new/evergreen" 2009-12-02T14:00:55 apparently they can't handle a little rain 2009-12-02T14:00:58 Huzzah 2009-12-02T14:01:36 the laser link my last employer installed between two buildings didn't cope well with freezing rain. This is a problem in Ontario 2009-12-02T14:01:42 *rap* *rap* *rap* 2009-12-02T14:01:51 all rise 2009-12-02T14:02:01 * djfiander tries to find a seat at the back of the room 2009-12-02T14:02:32 * berick hums Night Court theme 2009-12-02T14:02:38 I hereby call this here meeting to order; all you smart-alecks can use #evergreen as a back-channel 2009-12-02T14:03:06 Wanna run through the agenda? Any other exciting things to add? 2009-12-02T14:03:39 * dbs takes silence as assent 2009-12-02T14:03:53 * djfiander attempts to type louder 2009-12-02T14:04:03 let's test this netbook's battery! 2009-12-02T14:04:23 miker_++ # failover to your android if necessary 2009-12-02T14:04:54 berick: did we put George's acq-bugs into trac or launchpad? or did you just fix them? 2009-12-02T14:05:31 dbs: not yet. 2009-12-02T14:05:31 *** jenny1 has left #evergreen 2009-12-02T14:05:49 I will before the next meeting 2009-12-02T14:06:19 along w/ some other feature requests (if that's where we decide to put them) 2009-12-02T14:06:19 I'll carry that Action Item over to the upcoming minutes, then 2009-12-02T14:06:25 berick: record bugs? or fix'em? 2009-12-02T14:06:31 wlayton++ 2009-12-02T14:06:34 dbs: neither, but will record 2009-12-02T14:06:39 hah, okay 2009-12-02T14:06:39 then fix 2009-12-02T14:06:48 hooray. made it. 2009-12-02T14:06:58 once you finish web self-checkout part deux 2009-12-02T14:07:11 and a few other odds-n-ends, yes 2009-12-02T14:07:39 looks like the authority bug and user editor bug were fixed, berick++ 2009-12-02T14:08:04 (well, with the caveat about the new user editor vs. legacy but that's cool) 2009-12-02T14:08:14 yes, i didn't mark it out for that reason 2009-12-02T14:08:27 and trunk represents code that is some time off from release 2009-12-02T14:08:52 we can poke gmcharlt about it then, make it an action item for dbs 2009-12-02T14:09:18 trunk is always somewhere off in the horizon by definition :) 2009-12-02T14:09:32 fair enough ;) 2009-12-02T14:09:51 phasefx_: I think you backported those two fixes to 1.6.0.0 - yeah? (more relaxed “key” for oils_persist; bug fix for fleshing un-rendered list rows when scrolled into visibility) 2009-12-02T14:09:58 I did 2009-12-02T14:10:18 so after someone is on 1.6.0.0, they'll no longer lose such things as their spine label settings upon upgrade 2009-12-02T14:11:30 yayz 2009-12-02T14:11:43 the scroll into visiblity thing, I'm not sure if I made it work with both xulrunner 1.8 and 1.9, but 1.9 for sure 2009-12-02T14:12:09 1.8 is supposed to be deprecated for EG 1.6, but I know Indiana is using a custom build with xr1.8 for printer issues 2009-12-02T14:12:48 perhaps we can ask mrpeters-isl or someone else to try it out with 1.8 then 2009-12-02T14:13:12 if I can get more info on those printer issues, I want to file a bug with the mozilla folks (it affects Firefox as well) 2009-12-02T14:14:06 perhaps we could post a launchpad bug to collect printer issue info from affected sites? 2009-12-02T14:14:20 sounds good to me 2009-12-02T14:14:51 I'll make one and go soliciting for info 2009-12-02T14:14:57 phasefx_++ 2009-12-02T14:15:06 jeff: I see http://evergreen-ils.org/dokuwiki/doku.php?id=dev:start is coming along 2009-12-02T14:15:43 slowly but surely. 2009-12-02T14:15:45 jeff++ 2009-12-02T14:15:48 any and all help would be welcome. 2009-12-02T14:16:35 I plan/need to keep revising and enhancing the dev workshop, and nail it down to a specific release (right now it's OpenSRF-next + EG 1.6 or thereabouts) 2009-12-02T14:16:56 i did not post anything to list after last time, but i will do so soon (say, within the next 7 days) 2009-12-02T14:17:09 one question... who drafted the original "contributing" document? 2009-12-02T14:17:32 *** jenny has quit IRC 2009-12-02T14:17:38 I think it was stolen from PostgreSQL.org 2009-12-02T14:17:39 jeff: that was me 2009-12-02T14:17:49 parts were indeed stolen from pg 2009-12-02T14:18:30 we should wiki it for easier to modification 2009-12-02T14:18:37 miker_: ah, gracias. i really like it, but wanted to know who to seek forgiveness from if i started hacking it up. 2009-12-02T14:18:46 phasefx_: i placed it in the wiki. care to change the link from the website? 2009-12-02T14:18:57 s/stolen/inspired by/ :) 2009-12-02T14:18:58 phasefx_: http://evergreen-ils.org/dokuwiki/doku.php?id=contributing 2009-12-02T14:19:00 I can do that. jeff++ 2009-12-02T14:19:16 now the "This is a living document" is true! 2009-12-02T14:19:26 "It's Aliiiive!" 2009-12-02T14:19:34 plagiarized 2009-12-02T14:19:51 "took license with, per the terms of said license" 2009-12-02T14:20:16 So to change the commit-bit granting process to handing out commit bits to anyone who asks just takes a wiki edit? Cool! 2009-12-02T14:20:28 heh 2009-12-02T14:21:31 dbs: or a $100 check made out to me 2009-12-02T14:21:40 * phasefx_ wants to write up a simple bazaar howto, btw 2009-12-02T14:22:02 More seriously, can I suggest that we discuss any substantive changes on-list, possibly with a +/-1 voting system (which itself - a decision-making process - would be a substantive change) 2009-12-02T14:22:22 +1 2009-12-02T14:22:23 dbs: agreed 2009-12-02T14:22:27 substantive code-changes? +1 2009-12-02T14:22:53 I read that as substantive process change 2009-12-02T14:23:03 are there reasons (busines or otherwise) not to list current sponsored development work somewhere, be it wiki or launchpad? 2009-12-02T14:23:04 jeff: I meant substantive changes to the "contributing" document 2009-12-02T14:23:05 "major hacking up" of the doc 2009-12-02T14:23:10 dbs: aha. 2009-12-02T14:23:29 jeff: some folks don't want to be identified 2009-12-02T14:23:33 +1 also, and i'll not make major changes without discussion 2009-12-02T14:23:46 miker_: any reason not to say "being sponsored by an anonymous sponsor"? 2009-12-02T14:23:55 there's a difference between listing the work and identifying the sponsor 2009-12-02T14:23:58 that said, I would love to see a list of current sponsored development work, with the sponsors (yeah, I know it's all coming from Microsoft) redacted 2009-12-02T14:24:00 jeff: that's a thought 2009-12-02T14:24:04 miker_: at least, ensuring that "SOMEONE is working on this for SOMEONE" is public info somewhere? 2009-12-02T14:24:05 what jeff said 2009-12-02T14:24:13 +1 2009-12-02T14:24:25 imho, the document is too long 2009-12-02T14:25:14 I'm going to be soliciting some stakeholders for dev info for the newsletter; I could also bring up the thought of a more persistent document 2009-12-02T14:25:16 atz: cool - now that we have a process for making substantive changes, you could post your suggested revisions to the list and try out the voting system once things get worked out 2009-12-02T14:25:28 atz: noted. i don't want to drag discussion of that doc into this meeting, so i'll start discussion on the list for "this document needs some attention, here are some ideas, anyone else have thoughts?" 2009-12-02T14:25:36 jeff++ 2009-12-02T14:26:08 * dbs carefully notes the lack of denial on miker_'s part that Microsoft is sponsoring the bulk of new work 2009-12-02T14:26:18 dbs++ 2009-12-02T14:26:34 heh 2009-12-02T14:26:42 * miker_ whistles inocently ... ALL THE WAY TO THE BANK 2009-12-02T14:27:00 Can we move back on follow-up? sboyette did post about the buildbot several times (yay!) 2009-12-02T14:27:12 yay buildbot! 2009-12-02T14:27:18 sboyette++ 2009-12-02T14:27:22 sboyette++ 2009-12-02T14:27:23 sboyette++ 2009-12-02T14:28:13 *** error_23 has joined #Evergreen 2009-12-02T14:28:14 I think we're still looking for a QA team proposal, though 2009-12-02T14:28:46 I think so, too 2009-12-02T14:29:03 yes. I will go get my sharp pokin' stick 2009-12-02T14:29:22 (e.g. scope of team's responsibilities, some initial concrete directions, how the team works with Launchpad or Trac or whatever) 2009-12-02T14:29:24 his office is also flooding, btw 2009-12-02T14:29:46 like with water? 2009-12-02T14:29:49 so, we move that to action items? 2009-12-02T14:29:51 flooding-- 2009-12-02T14:29:59 djfiander: yes. you missed my earlier whining, I guess 2009-12-02T14:30:04 miker_: Yep. I'll carry that one forward, too. 2009-12-02T14:30:05 miker_: and it would be fair to open that up to someone else, wouldn't necessarily have to be sboyette to put the proposal together, if anyone else wants to volunteer 2009-12-02T14:30:06 sorry ... back to biz 2009-12-02T14:30:10 right-o 2009-12-02T14:30:42 1.6.0.0 post-release post-mortem - dbwells found a long-standing bug that I introduced (augh) 2009-12-02T14:31:20 otherwise I'm not sure if there's much else cued up for 1.6.0.1 yet 2009-12-02T14:31:33 dbs: the direct_ingest stuff? 2009-12-02T14:31:45 there are a couple minor things, I think 2009-12-02T14:32:11 yeah, likely some small fixes that didn't get forcabley shoved into the tag 2009-12-02T14:32:45 berick's wide-character fix, too 2009-12-02T14:32:57 yep, that's one 2009-12-02T14:33:06 is it a known/reported issue that in 1.6 that direct_ingest is significantly slower? 2009-12-02T14:33:11 miker_: direct_ingest fix would be nice, if wlayton's approach works 2009-12-02T14:33:29 well, I'd prefer we move EVERYTHING to nfd 2009-12-02T14:33:41 which would also address the issue, afaict 2009-12-02T14:33:54 jamesrf: yep - that's localized URIs, with flags available to prevent it - not sure we did document that anywhere though 2009-12-02T14:34:11 miker_: not according to dbwells, who found issues with some records using NFD and some with NFC 2009-12-02T14:34:18 ahh ... well, then 2009-12-02T14:34:20 nevermind 2009-12-02T14:34:22 heh 2009-12-02T14:34:47 that said, we don't have a record in hand identified as causing issue with NFD that we can inspect. I'll bug him to see if we can get one. 2009-12-02T14:34:58 * dbs likes real test data 2009-12-02T14:35:06 I saw you asked for that, now 2009-12-02T14:35:13 I'd be happy with artificial test data 2009-12-02T14:35:24 For 1.6 buglets, I think there's a few dojo interfaces that have "..." as translated strings. 2009-12-02T14:35:36 George pointed it out to me but I haven't had a proper look yet. 2009-12-02T14:35:49 (i'm paying attention again. my apologies; was working on i18n issues in build and lost track of time.) 2009-12-02T14:35:57 wlayton: yes ... conify interfaces 2009-12-02T14:35:58 wlayton: right, I saw that report but wondered if rebuilding the translations in 1.6.0.1 will resolve that problem 2009-12-02T14:36:15 dbs: That's what I was wondering, too. 2009-12-02T14:36:45 well, conify dojo stuff exhibits issues similar to that 2009-12-02T14:37:00 miker_: I think it something like copy statuses, in the case I saw. 2009-12-02T14:37:14 specifically, using the translation-addition widget inside a grid cell 2009-12-02T14:37:22 dojo grid cell 2009-12-02T14:37:41 *shudder* 2009-12-02T14:37:44 that sounds like it 2009-12-02T14:38:02 it would be good to get all of that into a bug 2009-12-02T14:38:03 all, trying to delete all bibs from a particular library and cant seem to diable the delete rule 2009-12-02T14:38:05 have tried ALTER TABLE asset.copy DISABLE RULE protect_copy_delete; 2009-12-02T14:38:07 so ... there's a way to address it, but it hasn't been yet 2009-12-02T14:38:13 but i get a syntax error, what am i doing wrong here? 2009-12-02T14:38:46 mrpeters-isl: you will probably need to DROP the rule, then recreate it 2009-12-02T14:39:02 mrpeters-isl: DROP RULE foobar ON tablename; 2009-12-02T14:39:12 dbs: I'll add a bug for it in LP. We might already have one filed with ESI that I can use. 2009-12-02T14:39:37 *** HitScan has joined #evergreen 2009-12-02T14:39:42 dbs: thanks! 2009-12-02T14:39:51 wlayton: you have a conglomorate ticket that includes mention of it, IIRC ... but LP please! 2009-12-02T14:40:14 so to bounce ahead in the agenda, I'll put out a call for translations now that 1.6.0.0 is out, so that we can start packaging up 1.6.0.1 soonish 2009-12-02T14:40:47 (and I'll point people at Launchpad as a possible means of translating new or existing languages, subject to it being a beta-ish experiment) 2009-12-02T14:41:01 dbs: which means ... are we going to use LP for that, and if so, do we target trunk, or "current" or ... all? 2009-12-02T14:41:12 does work done in launchpad translate back and forth to things like pootle? 2009-12-02T14:41:13 miker_: that's part of the beta-ishness 2009-12-02T14:41:41 k 2009-12-02T14:41:43 jeff: it's all done on top of .po files 2009-12-02T14:41:45 I'm targeting trunk atm, with the thought that the vast majority of strings will backport cleanly to 1.6.0.0 2009-12-02T14:41:53 right 2009-12-02T14:41:55 they're both tools that generate the same output, right? yeah, okay. good! 2009-12-02T14:42:12 dbs: how to add the rule back? 2009-12-02T14:42:43 but there were lots of changes I had to make to trunk to adopt a standard GNU gettext layout (which makes Launchpad and other tools like Pootle happier) 2009-12-02T14:42:49 CREATE RULE protect_copy_delete AS ON DELETE TO asset.copy DO INSTEAD UPDATE asset.copy SET deleted = TRUE WHERE OLD.id = asset.copy.id; 2009-12-02T14:42:55 mrpeters-isl: --^ 2009-12-02T14:42:58 thanks! 2009-12-02T14:43:11 jeff: but yes - and POedit, etc 2009-12-02T14:44:04 OpenSRF release: I'm advocating for just packaging up current trunk as 1.4, given that it's been 4 months since 1.2 and lots of change has gone in 2009-12-02T14:44:32 no objections here 2009-12-02T14:44:56 will need to make sure things are internally consistent, given some recent json parser changes 2009-12-02T14:45:09 i.e. make sure it's always being used, etc. 2009-12-02T14:45:25 but other than that, it should be more or less release-able 2009-12-02T14:45:26 berick: aye. mck9 seems to be pretty good about that 2009-12-02T14:45:30 indeed 2009-12-02T14:45:33 as it happens, nightly packaging of opensrf started up last night. http://testing.esilibrary.com/buildbot/tarballs/opensrf-latest.tar.gz 2009-12-02T14:45:40 sboyette++ 2009-12-02T14:45:42 openils is having some issues -- the ones which made me late to the meeting :) 2009-12-02T14:45:51 i18n, yes, dbs-- 2009-12-02T14:46:04 (I'm going to throw a polib-- in there for good measure) 2009-12-02T14:46:51 I think it's mostly 2009-12-02T14:46:55 polib-- 2009-12-02T14:47:21 that tarball WorksForMe, and it's accompanied by http://testing.esilibrary.com/buildbot/tarballs/opensrf-latest-is which contains the build number of the contents of the tarball 2009-12-02T14:47:34 so you can use that to automate downloads, since the tarball only generates when there's a new build 2009-12-02T14:47:44 (likely every night for openils, less so for opensrf) 2009-12-02T14:47:49 let me know if there are issues 2009-12-02T14:47:57 heh, cool (latest-is) 2009-12-02T14:48:12 scripting++ 2009-12-02T14:48:43 so if we decide we're happy with a given build, how hard is it to say "Cut build XXYY as 1.4.0"? 2009-12-02T14:48:53 (i'll write all this up once openils is working, too) 2009-12-02T14:49:52 groovy 2009-12-02T14:50:05 I always think of chainsaws when I say groovy 2009-12-02T14:50:12 "woodshed" 2009-12-02T14:50:35 * dbs chuckles demonically 2009-12-02T14:50:38 i don't know, exactly. at a minimum, version numbers need changing and a changelog needs generating. 2009-12-02T14:50:44 better than "bikeshed" 2009-12-02T14:50:56 but that could be done by unpacking the tarball everyone likes, putting that stuff in, adn repacking it 2009-12-02T14:51:00 ok, this all seriously went off the rails 2009-12-02T14:51:18 jeff: very significantly better 2009-12-02T14:52:06 thoughts on the LaunchPad experiment? 2009-12-02T14:52:35 dbs: the i18n experiment? 2009-12-02T14:52:51 well, translations are still sort of getting sorted out; bugs are getting some traffic, although some bugs are getting created and just being dormant 2009-12-02T14:52:58 i still think their tagging implementation sucks, but other than that it's working for me 2009-12-02T14:52:59 19 bugs opened, 8 have fixes committed 2009-12-02T14:53:01 I entered my first bug on it during this meeting (*shamefaced*). I like how it searches existing bugs first 2009-12-02T14:53:01 interface and performance of the site are both good so far... 2009-12-02T14:53:05 it's shiny, and i appreciate being able to open a bug. it never got trac integration, so it seems some of the promise of the experiment isn't there. 2009-12-02T14:53:06 oh, just in general 2009-12-02T14:53:08 dugh 2009-12-02T14:53:14 i've been trying to pay attention to bugs. at least tagging and stuff. 2009-12-02T14:53:16 lp++ 2009-12-02T14:53:31 "it's better than a sharp stick in the eye!" 2009-12-02T14:53:32 dbs: Positive so far. It needs a link from evergreen-ils.org, though 2009-12-02T14:53:40 and at dan's suggestion, there's https://launchpad.net/opensrf now 2009-12-02T14:53:42 wlayton: that's a good idea 2009-12-02T14:53:47 well, in conjunction with dan 2009-12-02T14:53:51 don't want to blame him entirely 2009-12-02T14:54:24 since LP bugs are global, that should aid in sorting a little bit 2009-12-02T14:54:26 I'm wanting to try pushing a personal bazaar branch to lp. May start working with topic branches instead of trunk all the time 2009-12-02T14:54:32 I'm wondering if dbwells or somebody else wants to push their own branch on lp 2009-12-02T14:54:44 oh, there goes phasefx beating me to the punch :) 2009-12-02T14:54:47 does it work with git? 2009-12-02T14:55:15 jamesrf: bzr only, I'm afraid; I think it can import a single non-bzr branch (thus we imported SVN trunk) 2009-12-02T14:55:32 * phasefx_ got trac-bzr working too, fun fun 2009-12-02T14:55:36 jamesrf: not directly... but via a bzr-git extension 2009-12-02T14:55:45 I miss the days when your choices were "RCS" or "SCCS" and everybody picked RCS. 2009-12-02T14:56:07 i had a question about the bug wranglers 2009-12-02T14:56:13 is that open to anyone, or just comitters 2009-12-02T14:56:24 jamesrf: atz and senator have stuff on github I think, I think it would make sense to at least post links to where people are keeping public repos 2009-12-02T14:56:31 Open-ILS/src/perlmods/OpenILS/Application/Trigger/Reactor/AstCall.pm 2009-12-02T14:56:38 er.... wrong pastepad 2009-12-02T14:56:43 heh 2009-12-02T14:56:47 http://github.com/senator/OpenILS-Evergreen 2009-12-02T14:57:04 jamesrf: that's subject to the QA team proposal, but the idea was supposed to be open to anyone who wants the responsibility of being a wrangler 2009-12-02T14:57:17 i don't think bug wranglers would necessarily have to be committers 2009-12-02T14:57:48 in a busy system, there is a lot more front line QA on a bug... and the bug could be against translation or documentation 2009-12-02T14:57:52 just folk what know the code good (that's technical talkin') 2009-12-02T14:58:16 or even who know how to ask the right questions to elicit what the real problem is 2009-12-02T14:58:25 sure 2009-12-02T14:58:52 Resolution: YouBeTripping. 2009-12-02T14:59:08 In the absence of a full QA team proposal, I would be happy to add anyone who wants to wrangle bugs to the wrangler team 2009-12-02T14:59:31 we have the diggers and the wranglers :) 2009-12-02T14:59:48 hippies vs. the cowboys 2009-12-02T15:00:21 better than neats vs scruffies 2009-12-02T15:00:26 i don't have a QA proposal because i had no idea what a QA team should really *do* when I said I'd take care of that. it's been given a lot of thought lately, and work stuff has been falling into place. i shall get that out shortly. 2009-12-02T15:00:37 i'm certainly not opposed to letting people in to triage adn work on bugs 2009-12-02T15:00:54 * moodaepo describe diggers/wranglers por favor 2009-12-02T15:00:54 that "Contributing" document might need to grow to encompass the diggers and the wranglers... hrmm 2009-12-02T15:01:01 dbs had the group set to moderated, so if someone disruptive shows up, they can be kicked out 2009-12-02T15:01:05 s/had/has/ 2009-12-02T15:01:16 moodaepo: diggers = Documentation Interest Group members 2009-12-02T15:01:16 worry about that when it happens 2009-12-02T15:01:19 is my position 2009-12-02T15:01:33 agreed - let's not overdo the gatekeeping on potential contributors 2009-12-02T15:01:34 dbs: it currently has a "documentation" section that needs to be updated to reflect DIG 2009-12-02T15:01:45 dbs Aaah right 2009-12-02T15:01:47 i would like to volunteer for QA but i don't know enough about QA to write a proposal 2009-12-02T15:01:54 part of the reason why i left "contributing" outside of the "dev:" namespace. 2009-12-02T15:02:17 "contributing" is also users reporting bugs, etc, etc, etc. 2009-12-02T15:02:25 moodaepo: bug wranglers = people who triage bugs, potentially write patches to fix bugs, create test cases to reproduce bugs and prevent further regressions, expand test coverage (in my one-line proposal) 2009-12-02T15:02:30 i'm drafting a message on the contributing thing right now, honest... :-) 2009-12-02T15:02:51 Well, that's a QA team proposal, I guess bug wranglers would do the first three parts 2009-12-02T15:05:20 https://launchpad.net/~evergreen-bugs is now an open team. Join us! 2009-12-02T15:06:13 Enough on LaunchPad for now? Does someone else want to take a whack at Trac-LaunchPad integration? I bailed once I started working out the dependencies for Trac 2009-12-02T15:06:24 "you will need to be approved by an admin" -- "You have successfully joined Evergreen Bug Wranglers." <-- timing 2009-12-02T15:06:31 hah 2009-12-02T15:07:07 i am not stepping forward, but has the intended goal of trac<->lp integration been stated previously? 2009-12-02T15:07:25 (I don't even know if new bugs have been opened in trac since launchpad launched, fwiw) 2009-12-02T15:07:32 goal/scope? is it "see what's possible, let that drive the scope of the experiment"? 2009-12-02T15:07:54 jeff: there was a fair bit of discussion about it at the last meeting, as part of the experiment 2009-12-02T15:08:06 from what I can tell, LP is becoming authoritative ... if we advertise it more it will gain more power, IMO 2009-12-02T15:08:08 k. "see previous logs". willdo. 2009-12-02T15:08:39 Trac would be the fully-triaged bug repo, launchpad would be the wild west, but I'm happy living in the wild west (to stretch that cowboy metaphor even further) 2009-12-02T15:08:49 what miker_ said 2009-12-02T15:09:14 I think we should go ahead and link to the lp bug page from the community site 2009-12-02T15:09:23 +1 2009-12-02T15:09:25 yeah, 2009-12-02T15:09:33 make a top-level Bug Report (or similar) nav header 2009-12-02T15:09:48 +3 2009-12-02T15:09:51 and if developers need a quiter bug holding pen, we have trac for that 2009-12-02T15:10:06 quieter 2009-12-02T15:10:26 +1 again 2009-12-02T15:10:36 or a psychological crutch (filters don't always work in the head) 2009-12-02T15:11:19 okay, so moving on, albeit still on launchpad "Feature requests - use of Launchpad or other vehicle for tracking?" I think only dbwells has tried blueprints.launchpad.net/evergreen so far 2009-12-02T15:12:06 dbwells: so .. you like it? 2009-12-02T15:12:32 https://blueprints.launchpad.net/evergreen/+spec/serials-basic-predictions-receiving is the example - pretty basic description with a link to a more complete spec 2009-12-02T15:12:45 in any case, recording requests there is better than what we do today, so ... put up a link on the comm site, and announce it to the lists, and see if it takes off? 2009-12-02T15:12:52 hey, it has a whiteboard, but you can't draw on it :) 2009-12-02T15:13:10 mebbe we could throw a bunch of blueprints up there and use that as the basis for the EG 2010 hackfest 2009-12-02T15:13:19 so would it be a good experiment to put some sponsored dev requests into blueprint? 2009-12-02T15:13:23 miker_: I like your thinking 2009-12-02T15:13:33 dbs: ooh, i like the hackfest thinking. 2009-12-02T15:13:58 dbs: that would be "propose for sprint", right? :) 2009-12-02T15:14:07 jeff++ 2009-12-02T15:14:20 jeff: aye. or crawl, in my case 2009-12-02T15:14:37 translations, I think we've covered-ish 2009-12-02T15:14:54 "can we proceed w/ telephony (re patch to open-ils-dev)?" 2009-12-02T15:15:19 what exactly is that asking? :) 2009-12-02T15:15:35 jeff: haha, you like that? 2009-12-02T15:15:38 miker_: about the blueprint, without things to link to like bugs or branches, it doesn't really do much at this point 2009-12-02T15:15:49 that was my hastily scrawled "i meant to ask about that" item 2009-12-02T15:15:57 berick: aha :) 2009-12-02T15:16:34 we have an approach for telephony and are ina holding pattern before we commit anything, pending.. some sign from the heavens 2009-12-02T15:16:50 message was "telephony building blocks (patch review)" 2009-12-02T15:16:57 mainly wondering if there are any objections to proceeding 2009-12-02T15:16:59 * miker_ descends, placard in hand 2009-12-02T15:16:59 thanks atz 2009-12-02T15:17:03 so, i have thoughts and feedback that would best go on-list, not in-meeting. i will do that. 2009-12-02T15:17:13 jeff++ 2009-12-02T15:17:38 jeff: see current state at http://github.com/senator/OpenILS-Evergreen/commits/asterisk 2009-12-02T15:17:55 atz: thanks. has it changed since the original message to -dev? 2009-12-02T15:18:21 nevermind. i see. 2009-12-02T15:18:25 dbs: sorry for the late reply, but concerning records which fail with decomposed characters, see here (also posted to the list): http://list.georgialibraries.org/pipermail/open-ils-dev/2009-May/004558.html 2009-12-02T15:18:42 no, not substantially. but I'm trying to add to it today 2009-12-02T15:19:03 dbwells: right - if you still have those MARC records, that would be really useful 2009-12-02T15:19:23 dbs: they are in the attachment 2009-12-02T15:19:42 dbwells: oh duh, missed that link 2009-12-02T15:19:49 dbwells++ 2009-12-02T15:20:19 \x{8} is backspace ... that's just bad data, no? (assuming it's not all mangled beyond comprehension at that point) 2009-12-02T15:20:35 dbwells++ #indeed 2009-12-02T15:20:52 that's probably post-mangling, but we can dissect later 2009-12-02T15:21:34 did anyone else present have thoughts on telephony? 2009-12-02T15:22:47 * crickets * two weeks is a long time to wait 2009-12-02T15:23:09 indeed. well, i'll get my two cents in on the list, then. 2009-12-02T15:23:14 jeff++ 2009-12-02T15:23:49 pre-emptive post-incrementing will get you nowhere. 2009-12-02T15:23:52 on the outstanding patch front, there was also a patch from art rhyno to enable client-side paging for records with hundreds or thousands of attached items - did anyone peek at that? 2009-12-02T15:24:36 I'll push it into a bug and ask for feedback on the list again 2009-12-02T15:24:49 that would be impressive.... 2009-12-02T15:25:15 well, so's art. 2009-12-02T15:25:27 oh hey, crap atz - I skipped over (conveniently) our craftsman action item 2009-12-02T15:25:39 shhh! 2009-12-02T15:25:46 we're supposed to open 100 bugs or something like that... 2009-12-02T15:25:58 let 1000 bugs bloom 2009-12-02T15:26:12 re that paging thing ... I have lots of feedback on that ;) 2009-12-02T15:27:19 dbs: I'll wait on the LP bug for providing feedback 2009-12-02T15:27:31 miker_: great 2009-12-02T15:27:45 one derailment averted! 2009-12-02T15:27:47 so - we have now fallen off of the edge of the agenda 2009-12-02T15:27:56 we're done? 2009-12-02T15:28:55 I think we should increment people like dbwells++ and wlayton++ and djfiander++ and jamesrf++ and others for providing patches and bug fixes and code reviews and whatnot, but I don't have a complete list and don't want to offend anyone by omission 2009-12-02T15:29:40 wait till you've had a chance to look at my code ;-) 2009-12-02T15:29:43 i am going to resubmit my opt-in patch via lauchpad when things quiet down in our shop 2009-12-02T15:30:01 the perl6 guys have a bot which auto-karmas anyone who makes a commit. if there's an LP API, someone could handle that similarly :) 2009-12-02T15:30:04 so, just to confirm... craftsman issues/bugs is no-action-yet, nothing-to-report, moving on? 2009-12-02T15:30:05 #evergreen++ 2009-12-02T15:31:09 *** pmp_coffee is now known as pmplett 2009-12-02T15:31:24 miker_: would you object to my attaching the craftsman bug document to a single launchpad bug for now, so that we can start dividing it up into child bugs? 2009-12-02T15:32:02 that makes sense to me 2009-12-02T15:32:38 dbs: if you svn-rm'd craftsman I wouldn't object 2009-12-02T15:32:44 heh 2009-12-02T15:32:48 hah 2009-12-02T15:32:55 but it has rounded corners! 2009-12-02T15:32:55 mrpeters-isl might, though 2009-12-02T15:33:13 *** eguest309_ has quit IRC 2009-12-02T15:34:22 jamesrf: sounds good (on the opt-in patch) 2009-12-02T15:34:50 So I think we're pretty much officially done, you can all go back to your paying jobs 2009-12-02T15:35:13 (I take miker_'s comment on craftsman as an "okay") 2009-12-02T15:35:22 dbs++ 2009-12-02T15:35:30 s/paying jobs/flooded offices/ 2009-12-02T15:35:38 dbs: thanks again for leading us toward the promised land 2009-12-02T15:35:43 dbs++ 2009-12-02T15:35:45 dbs++ 2009-12-02T15:35:48 thanks, all! 2009-12-02T15:36:05 craftsman++ # it has rounded corners 2009-12-02T15:36:21 wlayton++ # for lining up this meeting and agenda and everything else 2009-12-02T15:36:23 I'll write up some minutes tonight 2009-12-02T15:36:29 like that :) 2009-12-02T15:36:32 wjr++ # (i lol'd) 2009-12-02T15:37:01 *** wlayton has quit IRC 2009-12-02T15:37:04 *** djfiander has left #Evergreen 2009-12-02T15:37:29 * _bott_ needs to get in his plug asking for Program Proposals for EG2010! 2009-12-02T15:37:42 <_bott_> We've got some good ones, but could use more! 2009-12-02T15:37:59 oh hell, I meant to bring that up as an agenda item 2009-12-02T15:38:04 _bott_: when's the deadline? 2009-12-02T15:38:15 wondering whether we wanted to try to coordinate some developer-oriented sessions 2009-12-02T15:38:23 <_bott_> Hmm... wondering if I have a public one. 2009-12-02T15:38:35 dec 15 2009-12-02T15:38:36 deadline is december 15th I believe 2009-12-02T15:38:41 * berick needs to know how much to procrastinate 2009-12-02T15:38:52 <_bott_> Worth noting, we've pushed the "developer's update" to the opening session, to spur discussion througout 2009-12-02T15:39:08 _bott_: make sense 2009-12-02T15:39:10 / avoid hungover developers 2009-12-02T15:39:10 <_bott_> jeff++ dbs++ 2009-12-02T15:39:30 "Making your OPAC dance to the tune of MARC a la http://laurentian.concat.ca/opac/en-CA/skin/lul/xml/rdetail.xml?r=888287&l=105" 2009-12-02T15:39:43 <_bott_> encourage creative development ...over beers 2009-12-02T15:40:02 you think I won't drink after the hackfest? 2009-12-02T15:40:20 miker_: I think we should be drinking pre-hackfest as a defensive measure 2009-12-02T15:40:31 dbs: agreed 2009-12-02T15:40:51 * berick cancels hackfast, goes to smoky dive bar 2009-12-02T15:41:01 I'm going to go see if my offices has stopped floating ... biab 2009-12-02T15:41:09 <_bott_> We do have the #3 beer bar on planet earth http://www.hopcat.com/main/About.aspx 2009-12-02T15:41:12 bibtemplate++ 2009-12-02T15:42:43 *** pmplett has quit IRC 2009-12-02T15:50:01 is chopchop still maintained/viable/usable? 2009-12-02T15:50:45 jamesrf: probably ... 2009-12-02T15:53:13 just curious wrt there being so much documentation about it and it having been around forever but rarely spoken of 2009-12-02T15:53:33 might give it a whirl 2009-12-02T15:54:37 jamesrf: I made chopchop an opt-in part of the ./configure process, but the code is still there 2009-12-02T15:55:06 jamesrf: it's known to fail under extended/heavy load 2009-12-02T15:55:46 but that was 3 yrs ago 2009-12-02T15:55:50 a lot could have happened since then 2009-12-02T15:56:42 need to write it in erlang.. oh, wait 2009-12-02T15:57:56 http://svn.open-ils.org/trac/OpenSRF/browser/trunk/src/jserver says there hasn't been an awful lot happening :) 2009-12-02T16:00:15 *** Meliss has quit IRC 2009-12-02T16:03:42 *** dmcmorris_esi has quit IRC 2009-12-02T16:05:45 Is there a reason to use chopchop instead of ejabberd? 2009-12-02T16:06:31 mck9: it's just easier to configure. 2009-12-02T16:07:11 but it's unstable, has limited throughput, and, IIR, has no security ;) 2009-12-02T16:07:27 I ask because a lot of the complications in socket_bundle.c appear to be there solely to support chopchop. 2009-12-02T16:07:44 Eliminate chopchop and we could simplify socket_bundle.c quite a bit. 2009-12-02T16:08:05 *** HitScan has quit IRC 2009-12-02T16:10:37 wonder if anyone is using it 2009-12-02T16:11:37 * dbs nominates that we announce that chopchop is, like, totally deprecated as of OpenSRF 1.4, then immediately excise it post OpenSRF-1.4 2009-12-02T16:12:00 i was just typing words about deprecation/removal scheduling, but i'm slow 2009-12-02T16:12:12 Specifically chopchop is the only thing I know of that needs to open a listener socket. Everything else acts as a client to Jabber (of whatever implementation). 2009-12-02T16:12:23 *** pmplett has joined #evergreen 2009-12-02T16:12:39 although having it might make it easier to create an OpenSRF package on Fedora et al without having to worry about ramming conflicting ejabberd settings onto a system 2009-12-02T16:13:41 *** matt__ has quit IRC 2009-12-02T16:15:28 *** bshum has quit IRC 2009-12-02T16:15:52 But from what berick says, you wouldn't want to use chopchop in a production system anyway. It might be useful on a toy system for testing or the like. 2009-12-02T16:17:58 mck9: yes, I'm much more serious about excising it than i am about keepin git. 2009-12-02T16:18:22 *** rickd has quit IRC 2009-12-02T16:19:31 I'm not campaigning to get rid of chopchop, but I think it's worth asking what the costs and benefits are. 2009-12-02T16:20:12 The cost: otherwise needless complexity in socket_bundle.c. The benefit: I really don't know. 2009-12-02T16:33:21 *** bshum has joined #evergreen 2009-12-02T16:39:35 *** giordano_fatello has joined #evergreen 2009-12-02T16:40:50 *** dcage has joined #Evergreen 2009-12-02T16:41:37 *** jenny has joined #evergreen 2009-12-02T16:44:39 *** bshum has left #evergreen 2009-12-02T16:51:39 *** giordano_fatello has quit IRC 2009-12-02T16:54:21 *** Kulps has joined #Evergreen 2009-12-02T16:56:02 Question: Does evergreen have any type of ldap integration? 2009-12-02T16:57:22 Kulps: nothing in evergreen core 2009-12-02T16:57:52 Thats unfortuante 2009-12-02T16:58:00 has anybody else asked about it 2009-12-02T16:59:07 Kulps: yes, it has come up a few times, and I think there have been some patches posted to the list 2009-12-02T16:59:37 http://markmail.org/message/ztjridwvg2aidscj 2009-12-02T17:00:19 * dbs notes as an aside that that would be a nice feature to add 2009-12-02T17:00:34 thats a recent post too 2009-12-02T17:00:36 interesting 2009-12-02T17:00:57 I dont mean to troll, do you know of any library suites that come with LDAP out of the box? 2009-12-02T17:02:03 koha perhaps: http://wiki.koha.org/doku.php?id=en:development:ldap 2009-12-02T17:02:23 Koha is purely hosted though 2009-12-02T17:02:24 isnt it 2009-12-02T17:02:40 (To be honest I read it was hosted but never checked for myself) 2009-12-02T17:02:42 Kulps: no, you can download and install it yourself 2009-12-02T17:02:43 no, koha is open source like anything 2009-12-02T17:03:20 I'll state my scenario, I sit on the board of directors for a newly formed hackerspace 2009-12-02T17:03:39 we've got enough books to warrant a library and we're looking for a way to manage them 2009-12-02T17:04:02 *** jenny has left #evergreen 2009-12-02T17:04:38 we have no previous installation so migration is a non issue, but we would like to tie it into our open ldap 2009-12-02T17:04:44 to keep everything unified 2009-12-02T17:04:57 sounds like koha is your best bet at the moment 2009-12-02T17:05:38 or I'm sure some evergreen users would like that added to evergreen 2009-12-02T17:06:18 We've lots of programmers in our mix but I dont know if I want to let them at a key service on the network 2009-12-02T17:09:59 i think dbs is right... koha is probably your best bet for now 2009-12-02T17:10:22 atz would certainly know what it takes to get ldap properly integrated :) 2009-12-02T17:10:38 fair enough 2009-12-02T17:10:42 LDAP has been kicked around in EG circles for while... probably will make it's way into trunk eventually 2009-12-02T17:11:06 What should I watch out for when setting up? Common problems etc. 2009-12-02T17:11:36 there are different expectations for MS (active dir.) and openldap 2009-12-02T17:11:55 default (last I checked) was openldap style 2009-12-02T17:12:03 *** r123 has quit IRC 2009-12-02T17:12:11 I'd love to use MS AD, because thats what I know, however for us openldap is the only option, due to cost 2009-12-02T17:18:06 *** dcage has quit IRC 2009-12-02T17:21:54 *** dbs has quit IRC 2009-12-02T17:32:20 *** jamesrf is now known as jamesrf-afk 2009-12-02T17:38:20 *** Kulps has left #Evergreen 2009-12-02T17:39:36 *** branflakes has joined #Evergreen 2009-12-02T17:40:18 *** jmeeuwen has joined #evergreen 2009-12-02T17:48:54 In Evergreen 1.2, OpenSRF perl files installed by default in /openils/lib/perl5/OpenSRF; in 1.6.0.0, it looks like they install in /usr/local/share/perl/{and so} by default. 2009-12-02T17:49:52 Is this an expected behaviour, or does it mean I'm installing OpenSRF and its dependencies wrong? 2009-12-02T17:51:18 [In previous versions, some folks used NFS to share an /openils/ folder across multiple machines in a cluster, and this breaks that functionality.] 2009-12-02T18:11:08 *** branflakes is now known as branflakes-afk 2009-12-02T18:19:17 *** pmplett has quit IRC 2009-12-02T19:03:08 *** branflakes-afk has quit IRC 2009-12-02T19:06:46 just wanted to followup on branflakes 2009-12-02T19:07:12 i solved the problem by making just the perl bits on the seondary machine on the brick 2009-12-02T19:07:34 and running perl Makefile.PL && make && make test && make install in the perl folder of the OpenSRF src 2009-12-02T19:07:58 i would presume something similar would need to be done for the python if we were using python 2009-12-02T19:08:14 anyway... for the record there. 2009-12-02T19:24:04 *** pmplett has joined #evergreen 2009-12-02T19:32:02 *** jamesrf-afk is now known as jamesrf 2009-12-02T19:38:07 *** pmplett has quit IRC 2009-12-02T19:58:42 *** atz has quit IRC 2009-12-02T19:59:24 *** atz has joined #evergreen 2009-12-02T20:04:16 *** branflakes has joined #Evergreen 2009-12-02T20:04:25 @later tell branflakes in 1.4+, you need to install opensrf on all servers, and installing in the system perl dirs is indeed the expected behavior 2009-12-02T20:04:25 miker_: The operation succeeded. 2009-12-02T20:11:45 miker_: Thanks for the tip. 2009-12-02T20:16:49 *** brendan_ga has quit IRC 2009-12-02T20:28:31 *** brendan_ga has joined #evergreen 2009-12-02T20:40:13 branflakes: heh ... you must have appeared just as I was typing that 2009-12-02T20:40:27 *** wlayton has joined #evergreen 2009-12-02T20:41:24 Yeah guess so. So is the idea that you can install opensrf, then just wipe out the parts it creates in /openils? 2009-12-02T20:41:31 Or can you not use the NFS trick at all anymore? 2009-12-02T20:41:55 s/can/should 2009-12-02T20:44:48 you can use NFS 2009-12-02T20:45:01 as long as it's the same version everywhere, overwriting is fine 2009-12-02T20:45:14 (or hiding, I guess, via nfs mount) 2009-12-02T20:50:22 is just installing the perl as above not sufficient? does it need to put other stuff someplace? 2009-12-02T21:03:45 *** jamesrf is now known as jamesrf-afk 2009-12-02T21:38:29 *** jamesrf-afk is now known as jamesrf 2009-12-02T21:39:06 *** dbs has joined #evergreen 2009-12-02T21:57:04 sboyette: after all that, it looks like the unicode thing was just something I had failed to forward-port to trunk 2009-12-02T22:01:52 so, buildbot is failing on opensrf... cp: cannot create regular file `/usr/lib/apache2/modules/osrf_json_gateway.so': Permission denied 2009-12-02T22:02:10 if folk were not already aware. unclear if it's a buildbot problem, or a valid code issue. 2009-12-02T22:02:51 oh, and then a different permission-related error on the next build. looks like buildbot issue. 2009-12-02T22:09:00 "make install" not running as root? 2009-12-02T22:09:14 or not sudo'ing 2009-12-02T22:09:17 but yeah 2009-12-02T22:09:32 or filesystem broke and went read-only, etc 2009-12-02T22:09:44 likely something that sboyette will need to investigate. 2009-12-02T22:10:05 maybe already fixed, and just needs to see a new opensrf revision number? ;-) 2009-12-02T22:10:44 * phasefx_ thinks make install should work as non-root if the right --prefix is given, but I don't think that's the case 2009-12-02T22:20:16 phasefx_: that's a fine thing to think, but it requires changing how the apache modules get built and installed 2009-12-02T22:20:56 which is also required for packaging purposes, albeit that's via DESTDIR 2009-12-02T22:21:14 lots and lots of work to do! 2009-12-02T22:27:50 fun fun 2009-12-02T23:03:46 *** atheos has quit IRC 2009-12-02T23:12:53 dbs: just for you: gopher://2400.8n1.us/0/egbb 2009-12-02T23:13:31 jeff: that's awesome :) 2009-12-02T23:13:48 definitely OK 2009-12-02T23:16:04 *** wlayton has quit IRC 2009-12-02T23:16:20 minor bug meant OK was showing as NOT_OK -- fixed. 2009-12-02T23:18:26 hah 2009-12-02T23:21:37 jeff: you just made our 1.6 upgrade evening in BC :) 2009-12-02T23:23:21 branflakes: glad to help! :) 2009-12-02T23:23:45 i had been meaning to do that since dbs asked the other day, just needed to find free time for something "fun" 2009-12-02T23:24:13 jeff++ 2009-12-02T23:29:38 *** error_23 has quit IRC 2009-12-02T23:30:32 *** error_23 has joined #Evergreen