2008-05-12T04:13:03 *** ibbo has joined #OpenILS-Evergreen 2008-05-12T08:08:55 *** kbeswick has joined #openils-evergreen 2008-05-12T08:34:29 *** Slazer has joined #openils-evergreen 2008-05-12T08:59:42 *** dbs has joined #openils-evergreen 2008-05-12T09:17:20 *** dbs has quit IRC 2008-05-12T09:18:19 *** rsinger_ has joined #OpenILS-Evergreen 2008-05-12T09:24:58 *** gmcharlt has joined #OpenILS-Evergreen 2008-05-12T09:25:50 *** dbs has joined #openils-evergreen 2008-05-12T09:29:33 I hate screwing up 1M+ record imports 2008-05-12T09:29:50 "oops" ;) 2008-05-12T09:30:02 dmcmorris_esi: yeah :) 2008-05-12T09:30:04 You know what I hate more than that? 2008-05-12T09:30:10 whassat? 2008-05-12T09:30:15 dbs: how long does 1M take to import? 2008-05-12T09:30:21 screwing up 1M+ production records ;) 2008-05-12T09:30:34 I haven't done it... thankfully ;) 2008-05-12T09:36:25 berick: on our 8-core machines, 780K has been running for 820 minutes and is probably a little more than halfway through 2008-05-12T09:37:11 13+2/3 hours... 2008-05-12T09:37:13 (that's just the MARC21XML -> SQL portion; actually loading the SQL and then attaching the copies takes lots more time) 2008-05-12T09:37:45 well, not more time than the marcxml -> sql bit, but significant 2008-05-12T09:37:51 wow, that seems like a long time 2008-05-12T09:38:06 yeah 2008-05-12T09:38:26 dbs: just curious, what's your opensrf logging level set to? 2008-05-12T09:39:21 2 - and the logs are tiny 2008-05-12T09:39:26 k 2008-05-12T09:39:33 good thought though! 2008-05-12T09:46:07 hmm. something seems weird - looks like the record numbers have spiked up past 1000000 even though I know there are actually 780K records in our legacy system 2008-05-12T09:46:43 looks like there are exactly twice as many records in our marcxml. uh-oh. 2008-05-12T09:47:19 yup: grep '"001">' ../LU_records.marc.xml | sort | uniq | wc -l returns 785179 2008-05-12T09:47:31 I've concatenated two files together. ARGH! 2008-05-12T09:47:49 (insert mocking laughter from the internets here) 2008-05-12T09:48:01 *** Karen__ has joined #OpenILS-Evergreen 2008-05-12T09:51:38 yeah - unlink wasn't actually deleting any of the input files, so it was just concatenating wildly. ugh. 2008-05-12T10:06:21 *** djfiander has joined #openils-evergreen 2008-05-12T10:21:37 *** rsinger_ is now known as rsinger 2008-05-12T10:27:30 academic politics suck 2008-05-12T10:31:30 *** atheos has joined #OpenILS-Evergreen 2008-05-12T10:31:30 *** atheos_ has quit IRC 2008-05-12T10:33:36 wth, i have FF 2.0.0.14. the debian firefox-2-dom-inspector is installed but gives me no menu option for the dom inspector. if I try to install from the FF site, it tells me i need to update FF to .. 2.0.0.14 2008-05-12T10:33:43 * berick blames Hardy 2008-05-12T10:34:20 *** rsinger has quit IRC 2008-05-12T10:51:41 *** rsinger has joined #OpenILS-Evergreen 2008-05-12T11:41:26 *** pinesol has joined #openils-evergreen 2008-05-12T11:41:31 *** phasefx2 has quit IRC 2008-05-12T11:41:42 *** phasefx2 has joined #openils-evergreen 2008-05-12T11:47:03 djfiander: ok with you if i create a dir oilsweb/public/oils/media/ui_js/ ? We still need somewhere to put page-drawing JS that's specific to the Pylons app. I'm thinking the direclty layout within the dir will just match the template directory layout. if a given .html file needs a companion JS file, just plop it into this new dir. 2008-05-12T11:50:02 And maybe someday somebody will be assigned the task of creating an accessible version of the acq interface in pure Pylons 2008-05-12T11:50:18 just to throw randomness in there 2008-05-12T11:52:26 yeah, the infrastructure is there 2008-05-12T11:54:28 for more randomness, I'm also wondering about the navigation tree.. we probably need a better navigation tool than that. 2008-05-12T11:54:40 dojo menus, maybe 2008-05-12T11:58:41 an accordian container, perhaps? 2008-05-12T12:00:08 perhaps 2008-05-12T12:00:25 berick: ui_js works for me, I think. 2008-05-12T12:01:24 djfiander: ok, i'll set up some of my current views to do that.. see how it goes 2008-05-12T12:01:33 if it seems sane, i'll commit 2008-05-12T12:01:42 it won't break any existing code 2008-05-12T12:01:44 people actually fly into Windsor airport?!? 2008-05-12T12:01:54 berick: of course it won't 2008-05-12T12:01:57 For a loose definition of "people" 2008-05-12T12:02:04 and "fly" 2008-05-12T12:02:17 dbs: how long is that flight for you? 2008-05-12T12:03:53 *** sylvar has joined #openils-evergreen 2008-05-12T12:03:56 i see it would be quite a drive... 2008-05-12T12:05:23 berick: yeah, two days, easily. 2008-05-12T12:06:45 berick: the flights are something like 45 minutes each (sudbury to toronto, toronto to windsor) 2008-05-12T12:07:55 ah, just over an hour for each flight, actually 2008-05-12T12:07:57 dbs: sudbury's pretty much a "worst case" for driving to windsor. would you cut through the states, or go around the canadian side? 2008-05-12T12:08:54 huh, never considered going through the states 2008-05-12T12:09:19 given the stress related to border crossings, i would default to canada 2008-05-12T12:09:23 went to thunder bay for xmas a couple of years ago, driving through the states is a no-brainer 2008-05-12T12:09:37 (remember, i is st00pid) 2008-05-12T12:10:15 drive might be a lot more pleasant on the interstate, though, rather than 401... 2008-05-12T12:10:17 dbs: but for sudbury, it's a wash whether you go to North Bay or the Sault before heading south 2008-05-12T12:10:38 going to north bay would be a bad idea 2008-05-12T12:10:40 dbs: yup: upstate michigan is pretty rural. 2008-05-12T12:10:57 http://maps.google.ca/maps?f=q&hl=en&geocode=&q=from:sudbury,+on+to:windsor,+on&sll=49.891235,-97.15369&sspn=35.779204,82.177734&ie=UTF8&ll=44.363133,-80.639648&spn=4.940003,10.272217&t=h&z=7 2008-05-12T12:11:02 dbs: I always felt that way ;-) 2008-05-12T12:11:19 unless I felt like adding an hour to my drive! 2008-05-12T12:11:49 right. I've never been to Sudbury. always get it wrong 2008-05-12T12:12:18 that actually looks doable. 2008-05-12T12:14:20 yeah, it's only about 3 hours more to drive than it is to fly 2008-05-12T12:15:25 heh... long layover 2008-05-12T12:16:19 well - it takes half an hour to drive to the sudbury airport, and I'll guesstimate that for windsor airport -> hotel as well 2008-05-12T12:16:41 travel sucks, except for the fun bits in between 2008-05-12T12:17:59 from the time you get into the car in windsor to the hotel is about 15 - 30 minutes 2008-05-12T12:18:44 but how long from the hotel to the casino? _that's_ the critical info. 2008-05-12T12:19:23 new hotel: gotta drive. half hour or so. 2008-05-12T12:19:29 original hotel: walk. 2008-05-12T12:19:37 oh right. boo. 2008-05-12T12:20:17 there was another hotel down town, quality suites something. stayed there for a management workshop couple years ago, it might have been good. 2008-05-12T12:20:24 but we're all booked. 2008-05-12T12:24:34 *** ibbo has left #OpenILS-Evergreen 2008-05-12T12:24:57 * dbs groans as he realizes his import process just stepped all over the existing TCN values 2008-05-12T12:25:18 I guess I'll dump a list of the existing TCNs and use that as a keyfile, as god/eeevil intended 2008-05-12T12:25:32 heh 2008-05-12T12:25:41 well, the dontuse file, actually 2008-05-12T12:27:55 I was all proud of myself for working around the duplicate IDs (--startid didn't actually work) but forgot about the TCNs. 2008-05-12T12:27:57 Is it easy to display scope notes for an authority? Or is that not something eg does at the moment? 2008-05-12T12:29:00 I'm editing a MARC record and I don't see an obvious way to get to any sort of authority maintenance... 2008-05-12T12:29:45 I'm working with a major sleep deficit after getting back from taking my wife to the doctor, so forgive me if my questions today are a bit dumb. 2008-05-12T12:30:12 This is going to be ugly. two different legacy systems, each having natural TCNs starting from 1. gah. 2008-05-12T12:30:46 sylvar: you mean BT/NT scope stuff? I don't think that that level of authority support exists yet 2008-05-12T12:31:27 Could be wrong, but I think validating authorities just means "yes, this is in our list of authorities" or "no, this is not in our list of authorities" 2008-05-12T12:31:28 dbs: yes, or "Scamander, Newt: USE Rowling, J. K." 2008-05-12T12:32:38 I haven't got to the point of trying to implement authorities in our test instance yet, though, so I'm no authority on the subject 2008-05-12T12:33:00 sylvar: the 1xx has to match, so if the value in the MARC matches a 4xx or 5xx, but not a 1xx, it will be flagged 2008-05-12T12:33:44 eeevil: the 1xx has to match what? 2008-05-12T12:34:10 the controlled field in the MARC 2008-05-12T12:35:09 so, if you have an authority record with a 100a=Scamander, Newt then it won't be flagged, but if your 100a=Rowling, J. K. and it has a 400a=Scamander, Newt then it will be 2008-05-12T12:39:29 I think sylvar is interested in having authority data displayed in a certain way 2008-05-12T12:40:01 sylvar: is that right? 2008-05-12T12:42:13 I want to know about scope notes -- it's one of the university's concerns about migrating. Whether we can display them in the staff catalog. 2008-05-12T12:43:59 sylvar: what's the marc tag? 2008-05-12T12:44:52 *** dbs has quit IRC 2008-05-12T12:44:58 They didn't specify, but if the answer is "yes, for these tags", that's fine. 2008-05-12T12:45:13 well, it's not as simple as that... 2008-05-12T12:45:37 the display, today, is based on a marc->mods transformation 2008-05-12T12:45:56 *** dbs has joined #openils-evergreen 2008-05-12T12:46:44 ahh... well, what I'm seeing is /authority/ scope notes 2008-05-12T12:47:40 so .. extra-bib info 2008-05-12T12:48:50 sylvar: do you have an understanding of where they want these scope notes displayed? 2008-05-12T12:49:23 eeevil: Not precisely, no. 2008-05-12T12:58:09 eeevil: hey, error message for you. on the acq box postgres.. when I do: \d acq.fund_allocation 2008-05-12T12:58:12 ERROR: column t.tgconstraint does not exist 2008-05-12T12:58:15 LINE 3: WHERE t.tgrelid = '59796' AND t.tgconstraint = 0 2008-05-12T12:58:32 wow ... that's pg internal 2008-05-12T12:58:49 do other tables work? 2008-05-12T12:59:11 \d seems to result in the same error regardless of the table 2008-05-12T12:59:48 apart from \d not working, i can't find any other problems 2008-05-12T13:01:27 berick: phasefx just had a thought... version mismatch between the psql shell and the server version 2008-05-12T13:01:32 is the shell newer than the server? 2008-05-12T13:01:49 probably 2008-05-12T13:01:59 yep, 8.3.1 2008-05-12T13:02:02 for the shell 2008-05-12T13:02:18 there you go 2008-05-12T13:02:30 get that 8.3 sh...stuff off of there, man ;) 2008-05-12T13:02:46 heh, will do 2008-05-12T13:02:47 thx 2008-05-12T13:47:45 *** kbeswick has quit IRC 2008-05-12T13:51:41 so a nice little dontuse file generated by foreach my $x (1..1000000) { print FH "$x\n"; } and we're off again 2008-05-12T13:57:25 *** kbeswick has joined #openils-evergreen 2008-05-12T14:03:15 hahahaha 2008-05-12T14:03:24 dbs++ 2008-05-12T14:04:24 -= THIS MESSAGE NOT LOGGED =- 2008-05-12T14:09:24 I ... I ... I'm pretty sure I don't want to know... 2008-05-12T14:13:56 *** dbs has quit IRC 2008-05-12T14:16:17 *** dbs has joined #openils-evergreen 2008-05-12T14:35:02 Alright, I'm running into a little problem. I'm trying to run the staff client but I get an error about and XML parsing error. 2008-05-12T14:38:17 XML Parsing Error: undefined entity 2008-05-12T14:38:17 Location: chrome://open_ils_staff_client/content/main/main.xul 2008-05-12T14:38:17 Line Number 21, Column 1: ^ 2008-05-12T14:39:33 so what entity shows up at line 21 on content/main/main.xul ? 2008-05-12T14:39:50 Open-ILS/xul/staff_client/chrome/content/main/main.xul 2008-05-12T14:42:00 21: 22: onload="try { main_init(); } catch(E) { alert(E); }" 2008-05-12T14:42:01 23: onunload="try { G.auth.logoff(); } catch(E) { alert(E); }" 2008-05-12T14:42:01 24: title="&staff.auth.title;" 2008-05-12T14:42:01 25: width="640" height="480" 2008-05-12T14:42:01 26: xmlns="http://www.mozilla.org/keymaster/gatekeeper/there.is.only.xul"> 2008-05-12T14:42:24 missing entity, I bet 2008-05-12T14:43:00 &staff.auth.title; isn't set in whatever dtd the page is loading 2008-05-12T14:45:22 yep, and it would need to be loaded from a DTD in chrome 2008-05-12T14:45:35 (sorry, in a meeting so can't be as useful as normal) 2008-05-12T14:46:30 So how would I know what DTD it is trying to use? Or is it just the default one as I haven't specified any? 2008-05-12T14:48:09 Slazer: it would be at the top of that same file, as a processing instruction ... IIRC 2008-05-12T14:48:11 DTD is defined up at the top of the XUL file 2008-05-12T14:48:35 Ah, right. 2008-05-12T14:53:27 I've looked at two lang.dtd files (not sure which file "chrome://open_ils_staff_client/locale/lang.dtd" actually refers to) and both have the staff.auth.title entity. 2008-05-12T14:54:42 Slazer: what version are you building the SC from 2008-05-12T14:55:13 I ...am not sure. 2008-05-12T14:55:29 which branch did you check out from svn? 2008-05-12T14:55:34 trunk 2008-05-12T14:56:06 k ... we'll have to point phasefx or dbs at that ... whichever is available first 2008-05-12T14:56:33 dbs: i am sending you a patch to test whenever you get a chance. it appears to be working (the autoconf with the hardcoded make files so far). 2008-05-12T15:01:10 I'll be out of this meeting soon 2008-05-12T15:02:24 chrome://open_ils_staff_client/locale/lang.dtd would mean it is in Open-ILS/xul/staff_client/chrome/locale/en-US right? If it does... I don't have a lang.dtd there at all. 2008-05-12T15:11:00 Ok, got it. Never mind. I was just ...doing something wrong. *bows head in shame* dbs helped me. 2008-05-12T15:18:30 time to visit the accountant 2008-05-12T15:18:37 *** djfiander has quit IRC 2008-05-12T15:19:46 Ok, now something else I need help with. I still have my first build (20080505) and not today (20080512) so the staff client runs but gives me an error about the server not supporting the version of the client. How do I build it properly so I get 20080512? 2008-05-12T15:20:19 just add a symlink on the server side from the new build to the old on 2008-05-12T15:20:41 ln -s 20080512 20080505 2008-05-12T15:20:51 er 2008-05-12T15:20:55 swap those around 2008-05-12T15:21:41 Oh, heh, that works. But where? What directory do I make the symlink in? 2008-05-12T15:23:30 /openils/var/web/xul/ 2008-05-12T15:23:36 by default, anyway 2008-05-12T15:24:14 Ah yes, there it is. Thanks. 2008-05-12T15:43:47 *** dbs has quit IRC 2008-05-12T15:46:04 *** dbs has joined #openils-evergreen 2008-05-12T15:53:42 *** kbeswick has quit IRC 2008-05-12T15:59:34 *** pmurray_away is now known as pmurray 2008-05-12T16:00:11 miker_: Ping. 2008-05-12T16:00:45 *** Slazer has quit IRC 2008-05-12T16:01:43 eeevil: ping ^^ 2008-05-12T16:02:39 phasefx: Thanks -- I forgot about that handle. 2008-05-12T16:02:44 eeevil: Ping. 2008-05-12T16:05:37 pmurray: pong 2008-05-12T16:06:24 eeevil: thanks for taking care of slazer while I was otherwise occupied 2008-05-12T17:03:33 *** pmurray is now known as pmurray_away 2008-05-12T18:01:51 *** Karen__ has left #OpenILS-Evergreen 2008-05-12T18:07:28 *** dbs has quit IRC 2008-05-12T18:45:49 Are there ways other than Z39.50 that a cataloger might use to download MARC records from a vendor? 2008-05-12T18:46:10 I recall using FTP in Dynix Classic many years ago, but... 2008-05-12T19:02:52 *** djfiander has joined #OpenILS-Evergreen 2008-05-12T19:56:29 berick: did 'flesh_li_details' paramter for open-ils.acq.lineitem.retrieve change a few days ago? 2008-05-12T19:56:56 I'm not getting fleshed lineitem-details any more (that is, fund and location aren't there) 2008-05-12T20:01:23 *** asmodai has quit IRC 2008-05-12T20:01:30 big bang theory! back in 30 2008-05-12T20:39:37 I am being ignored 2008-05-12T20:45:57 djfiander: i haven't touched the perl code is some time 2008-05-12T20:46:07 hmm. 2008-05-12T20:47:37 djfiander pasted "lineitem.retrieve" at http://paste.lisp.org/display/60658 2008-05-12T20:48:27 it used to work. really. 2008-05-12T20:49:11 but now the alert is showing me that the li_details have a bunch of nulls with a jub_id and a lineitem_detail id being the only things set 2008-05-12T20:49:48 what lineitem id are you retrieveing? 2008-05-12T20:49:51 i'm gonna try srfsh 2008-05-12T20:50:02 25 2008-05-12T20:50:04 I think 2008-05-12T20:50:43 yup, 25. "Vernon God Little" 2008-05-12T20:55:25 looking at the db, the objects are pretty much all empty, except for ID, lineitem, and occaisonally the fund id 2008-05-12T20:55:55 berick: hmm... there were funds there. maybe I scrambled them with my fund editing code :) 2008-05-12T20:56:20 I see 3 with funds for lineitem 25 (fwiw) 2008-05-12T20:56:34 ok. that's right. 2008-05-12T20:56:40 but I'm not getting them in the JS 2008-05-12T20:58:13 woops 2008-05-12T20:58:17 no, I am seeing them 2008-05-12T20:58:24 never mind. PEBKAC 2008-05-12T20:59:05 well, I'm seeing them in the alert... 2008-05-12T21:01:31 hmm. 2008-05-12T21:04:51 *** dbs has joined #openils-evergreen 2008-05-12T21:10:46 ok, the right data is being retrieved, but only the id and lineitem are being stored in the store. 2008-05-12T21:11:18 a problem with acqlid.toStoreData? fmall.js is right. 2008-05-12T21:12:35 are the other fields being stored (with null values)? 2008-05-12T21:12:44 nope. 2008-05-12T21:13:21 wait, let me confirm, but I'm pretty sure not. 2008-05-12T21:13:29 of course, i'm not sure if it usually stores null values or not 2008-05-12T21:15:01 nope. there's an 'id' field and a 'lineitem' field, but no other fields. 2008-05-12T21:15:17 even when the value coming across the wire is non-null (the fund) 2008-05-12T21:16:08 is the fund fleshed? 2008-05-12T21:16:48 in the li_detail, it's coming across the wire, yeah, as a fund id. 2008-05-12T21:17:01 I get the fund name elsewhere. this did work, once upon a time. 2008-05-12T21:20:35 well, objects don't collapse to their ID in toStoreData, and the store just throws away objects in fields, I think 2008-05-12T21:20:47 so, you need an unfleshed lid, I think 2008-05-12T21:21:36 all that's in the fund field in the lid is the id. 2008-05-12T21:22:00 yeah, that call's not fleshing the fund 2008-05-12T21:22:32 I get the fund name elsewhere. this did work, once upon a time. 2008-05-12T21:24:04 sorry, right. read "yeah" backwards 2008-05-12T21:25:34 djfiander: what field is the details array in.. 'details'? 2008-05-12T21:26:36 ah, lineitem_details 2008-05-12T21:27:53 ok, the fund is surving the trip through acqlid.toStoreData() 2008-05-12T21:28:22 on to check ItemFileWriteStore() 2008-05-12T21:37:09 and the McMaster SQL load commences... 2008-05-12T21:37:14 hmm 2008-05-12T21:39:19 dbs: I'll be interested in the speed of staged search after you get those in ... any idea on the record count? 2008-05-12T21:41:55 should be around 4M records after LU + Mac 2008-05-12T21:43:46 hmm... the right data is in the store, and I can see it in the model (in the JS console), but model.getRow() is returning incomplete info 2008-05-12T21:43:47 augh. cancel that. looks like the import process still isn't finished. 2008-05-12T21:44:36 silly me, it hasn't even been 7 hours yet :) 2008-05-12T21:46:24 heh 2008-05-12T21:46:36 dbs: should'a removed the indexes ;) 2008-05-12T21:46:57 I guess /I/ should'a mentioned that this morning :P 2008-05-12T21:47:04 no no, I'm talking about the marc2bre / direct_ingest / parallel_pg_loader 2008-05-12T21:47:12 ahhh... 2008-05-12T21:47:32 are you doing the full pipe? 2008-05-12T21:47:41 yeah, the indexes will definitely be turfed for the load 2008-05-12T21:47:49 yep 2008-05-12T21:48:52 ahh... so no idea of recs/sec 2008-05-12T21:50:22 one of the processes doesn't care about --quiet, and is showing 2056 2008-05-12T21:50:49 that's parallel_pg_loader 2008-05-12T21:51:05 not quite bibs / sec :) 2008-05-12T22:05:00 *** lisppaste6 has quit IRC 2008-05-12T22:05:16 ah! 2008-05-12T22:05:28 it's in the store, but the model doesn't see the fields? 2008-05-12T22:05:56 dbs: that's rows/s ... I'd estimate that to be 50 r/s or there abouts 2008-05-12T22:06:36 which is in line with direct_ingest on a fast-ish CPU 2008-05-12T22:07:12 this is with a remote db, fwiw 2008-05-12T22:07:38 I don't think it's worth anything 2008-05-12T22:08:09 oh. 2008-05-12T22:08:11 hmm. 2008-05-12T22:12:02 fuck me 2008-05-12T22:12:30 the store/model(?) base the names of the fields in the store on the FIRST item in the array we build the store from 2008-05-12T22:12:44 so, if the first record has a fund, then you can query the fund field. 2008-05-12T22:12:50 if it doesn't, then no funds for you 2008-05-12T22:13:17 no wonder it used to work and then stopped. the order of the lids changed so the fund 'disappeared' from the model. 2008-05-12T22:14:09 so, toStoreData needs to create fields for all fields in the records, even things that are null. 2008-05-12T22:14:52 I figured this out by very carefully shifting lids with defined funds onto the beginning of the list and voila! funds were displayed. 2008-05-12T22:14:55 POS 2008-05-12T22:15:51 djfiander: well, now, there is another way ... you can add a field datatype map 2008-05-12T22:16:19 but that's sensitive to fmall changing 2008-05-12T22:17:01 not if I give you a storeFieldMap() method ;) 2008-05-12T22:18:08 and wrap that and toStoreData in a toStoreStructure method 2008-05-12T22:18:33 wouldn't it just make all of the rest of code simpler to create the fields in the store in the first place? 2008-05-12T22:18:57 you mean for dojo in general? 2008-05-12T22:19:32 I mean inside the .toStoreDate() method 2008-05-12T22:19:54 lemme look at it 2008-05-12T22:20:11 yeah, I'm looking at it, but it's a bit arcane to figure out on the fly 2008-05-12T22:20:35 ok .. the problem is actually in toHash 2008-05-12T22:21:23 ok .. a 2-line change should fix it in toHash 2008-05-12T22:21:29 right, "if there's a field, and it's non-null..." 2008-05-12T22:21:31 cool 2008-05-12T22:21:33 even better 2008-05-12T22:22:02 so, this time it was miker_ 's fault 2008-05-12T22:22:04 ;-) 2008-05-12T22:22:09 committing to trunk ... don't want to give svnmerge fits ... 2008-05-12T22:22:11 indeed 2008-05-12T22:22:13 miker-- 2008-05-12T22:22:43 *** lisppaste6 has joined #openils-evergreen 2008-05-12T22:22:52 i can merge.. 2008-05-12T22:22:56 cool 2008-05-12T22:23:04 i have my clean merge tree waiting for trouble ;) 2008-05-12T22:23:06 djfiander: ok .. one more change 2008-05-12T22:23:52 merged 2008-05-12T22:24:05 berick: heh ... merge again, please :) 2008-05-12T22:24:14 k 2008-05-12T22:28:44 back 2008-05-12T22:28:49 so now I can svn up hash? 2008-05-12T22:29:37 and dojoData.js 2008-05-12T22:29:53 it has to tell toHash to allow nulls 2008-05-12T22:30:05 djfiander: yeah, svnup 2008-05-12T22:30:06 so as not to break other stuff... 2008-05-12T22:30:12 ok 2008-05-12T22:32:00 heh. ok, working. sorta 2008-05-12T22:33:30 my grid is now full of 'null' 2008-05-12T22:35:35 good news: all the fields are defined. 2008-05-12T22:35:46 bad news: the fields all have the string "null" in them 2008-05-12T22:35:49 are the fund IDs there? 2008-05-12T22:35:59 when they're defined, yeah 2008-05-12T22:36:07 ok, good 2008-05-12T22:36:21 yeah, the store will stringify whatever it's given.. so it will turn the nulls into 'null' 2008-05-12T22:36:53 so, there's no way to create fields that don't have a value? 2008-05-12T22:37:17 well, there is, because if the first lid has a fund, the rest have a null fund field 2008-05-12T22:37:47 IIR, if the field just isn't defined, it will put a '?' into the grid cell. other option, of course, is to replace nulls with "" 2008-05-12T22:38:22 yes, but if I define a 'get' function, it can check to see if the field is defined and return nothing, which I do 2008-05-12T22:38:37 but now, it is defined 2008-05-12T22:38:58 well, I can fix that too... sec 2008-05-12T22:39:09 *hack*hack*hack* 2008-05-12T22:41:46 ok ... merge and try that :) 2008-05-12T22:41:52 also, going to watch House 2008-05-12T22:42:15 merge 2008-05-12T22:42:15 d 2008-05-12T22:43:16 great. 2008-05-12T22:43:26 hash.js failed to load: syntax error :P 2008-05-12T22:44:25 hah! 2008-05-12T22:44:36 '{' are not optional in JS!!! :P 2008-05-12T22:44:52 i'll fix 2008-05-12T22:45:10 I feel better now :) 2008-05-12T22:46:08 perfect. 2008-05-12T22:47:53 should be updated 2008-05-12T22:48:12 yup 2008-05-12T22:50:13 btw folks - I think we should try to catch a Detroit hockey game next week 2008-05-12T22:50:42 that would be awesome 2008-05-12T22:52:48 so, one night: hockey, one night: casino, one night: essex 2008-05-12T22:53:45 you know, I just reverted Lineitem.js to the last checked in version and everything worked fine *sigh* 2008-05-12T22:55:27 yeah, games haven't been selling out so there's a chance we could get tickets 2008-05-12T22:55:43 man, I am one harsh patch reviewer 2008-05-12T22:58:04 ok, bedtime 2008-05-12T22:58:07 *** djfiander has quit IRC 2008-05-12T23:00:13 *** sylvar has quit IRC 2008-05-12T23:23:42 *** dbs has quit IRC