| # | Time | Nick | Message |
|---|
| # | 00:22:05 | atz | yeah, but what's that in latin? |
| # | 00:22:33 | dbs | vidi, vici, nom nom nom |
| # | 00:22:44 | atz | well said |
| # | 00:24:47 | jeffdavis | technically that should be "veni, vidi, nom nom nom" |
| # | 00:25:49 | jeffdavis | (when I die I will be in the circle of hell reserved for pedants, where the demons deliberately mangle spelling and grammar and say things like "irregardless") |
| # | 00:35:44 | dbs | jeffdavis: its unevitable |
| # | 00:41:32 | dbs | jeff: is http://svn.open-ils.org/trac/ILS-Contrib/changeset/809/conifer/branches/rel_1_6_0/web/opac/skin/default/js/rdetail.js a valid patch? |
| # | 00:41:55 | jeff | looking |
| # | 00:42:07 | dbs | don't have an unpatched system ready at hand to test "search by barcode as logged-in user and get no bookbag" problem |
| # | 00:42:28 | dbs | I committed that so long ago - shocked that I didn't push it into core if it's real |
| # | 00:42:38 | jeff | oh |
| # | 00:42:47 | jeff tests |
| # | 00:43:39 | phasefx | night owls |
| # | 00:43:50 | jeff | who, ho |
| # | 00:43:53 | jeff | who, who |
| # | 00:44:21 | phasefx | dbs++ # re: wiki edits |
| # | 00:45:04 | dbs | jeff++ # moving that document to the wiki in the first place |
| # | 00:46:05 | jeff | log in, advanced search, search by barcode, i get the add to bookbag option and it works. rdetail.js does not appear to have your change. |
| # | 00:46:11 | jeff | in production rel_1_6 |
| # | 00:46:16 | jeff | er, rel_1_6_0 |
| # | 00:47:26 | jeff | why did it break on a barcode search? |
| # | 00:49:30 | dbs | Dunno. I guess, if it really did affect us, it's because of some of the wonky skin customizations |
| # | 00:49:49 | dbs | When I upgrade, I'll try dropping that patch and see if it affects anything |
| # | 00:49:54 | dbs | jeff++ thanks |
| # | 03:06:30 | dbs | @later tell phasefx ugh, 1.6.1.1 client got packaged up with that stupid busted "about" label definition in the build/chrome/locale/lang.dtd files for everything except en-US, I think |
| # | 03:06:30 | pinesol | dbs: The operation succeeded. |
| # | 03:25:21 | gdunbar has quit IRC |
| # | 03:27:09 | Callender has quit IRC |
| # | 04:07:14 | dbs | whee. live-ish on 1.6.1.1+. needed to add perms for booking stuff (%RESERVATION%) and adjust lots of skin customizations. and add all of the password reset stuff, along with UUID::Tiny |
| # | 04:07:47 | dbs | add the open-ils.booking clause to opensrf.xml... |
| # | 04:09:09 | dbs | still need to adjust action-trigger-runner to run more than once a day, now |
| # | 04:10:48 | dbs | symptom of missing UUID::Tiny; autogen.sh fails because open-ils.actor.org_tree.retrieve method can't be found. because open-ils.actor fails to start (heh) |
| # | 04:12:44 | dbs | Also need to figure out how booking stuff is supposed to work, as there are no docs to my knowledge, and we didn't catch that webinar |
| # | 06:12:59 | cerpy has joined #evergreen |
| # | 06:16:39 | cerpy | my debian filesystem got errors, after resolving that, apache won't start due to segmentation faults. I heard there is an Evergreen script that starts services that might help...? |
| # | 07:08:21 | dbs | cerpy: seg faults? that's not good. I don't think a startup script could help with that |
| # | 07:13:58 | dbs | if you shut everything down, then reboot, that's about as clean as you can get |
| # | 07:14:30 | dbs | everything = osrf_ctl.sh -l -a stop_all (with or without the -l) |
| # | 07:15:05 | cerpy | dbs: yes, I did that. I'm trouble shooting now. |
| # | 07:16:45 | dbs | ugh, not good |
| # | 07:17:34 | dbs | in the worst case scenario, you might need to get another copy of the virtual image and start fresh. hopefully any changes you made are checked into a local SVN! |
| # | 07:19:43 | miker_ | dbs: http://open-ils.org/dokuwiki/doku.php?id=booking (plenty of booking docs) |
| # | 07:19:52 | miker_ hits the road |
| # | 07:37:41 | dbs | thanks miker_! had no idea that existed |
| # | 07:39:04 | dbs | search in the wiki finds it, but http://open-ils.org/dokuwiki/doku.php?id=booking&do=backlink says there are no links to it :) |
| # | 07:41:12 | dbs | @later miker_ are those booking docs under the CC-BY-SA license that the DIG is using? didn't see a license in the docs themselves, but would be awesome if they could be rolled into the doc project |
| # | 07:41:12 | pinesol | dbs: Error: The "Later" plugin is loaded, but there is no command named "miker_" in it. Try "list Later" to see the commands in the "Later" plugin. |
| # | 07:41:18 | dbs | @later tell miker_ are those booking docs under the CC-BY-SA license that the DIG is using? didn't see a license in the docs themselves, but would be awesome if they could be rolled into the doc project |
| # | 07:41:18 | pinesol | dbs: The operation succeeded. |
| # | 07:59:11 | alxp has joined #evergreen |
| # | 08:04:15 | mck9 has left #evergreen |
| # | 08:04:21 | mck9 has joined #evergreen |
| # | 08:29:49 | gmcharlt | dbs: FWIW, I killed that comment spam |
| # | 08:31:16 | gdunbar has joined #evergreen |
| # | 08:47:52 | dbs | gmcharlt: yay |
| # | 08:59:19 | tsbere has joined #evergreen |
| # | 09:06:27 | miker_ | dbs: I'll make sure a license statement gets added to that. the DIG can use it, of course |
| # | 09:08:49 | Callender has joined #evergreen |
| # | 09:12:18 | Dyrcona has joined #evergreen |
| # | 09:13:04 | Dyrcona | When I insert something that is supposed to be a JSON string in the database, in actor.org_unit_setting for instance, do I need to put "" around it? |
| # | 09:16:01 | senator | Dyrcona: yes |
| # | 09:16:39 | Dyrcona | tyvm. I thought so, because something was complaining about my lib.courier_code entries. |
| # | 09:16:52 | collum has joined #evergreen |
| # | 09:16:57 | Dyrcona | just thought i'd ask before updating my entries. |
| # | 09:17:06 | senator has banged his head against that before |
| # | 09:17:09 | miker_ | Dyrcona: OpenSRF::Util::JSON::perl2JSON() is your friend |
| # | 09:17:50 | miker_ | just give it your perl $thingy an shove what it gives you in directly |
| # | 09:18:10 | Dyrcona | ok. but i did this with insert statements. :) |
| # | 09:18:12 | miker_ | of course, you're probably talking about doing it directly in the db |
| # | 09:18:13 | miker_ | right |
| # | 09:18:20 | Dyrcona | yep |
| # | 09:18:39 | miker_ should add an is_json CHECK constraint on that if it's not already there |
| # | 09:22:15 | bshum has joined #evergreen |
| # | 09:24:28 | dbs | heh - good to have that clarification: I'll go back to requiring a JSON list of field/subfields for the callnumber stuff and use JSON2perl in the middle-layer to parse |
| # | 09:25:41 | Melissa has joined #evergreen |
| # | 09:25:53 | dbs | been mulling that over for a few days thinking that a GUI for manipulating the array shouldn't be /that/ hard (and better than asking people to hand-code valid JSON arrays in the ou setting) |
| # | 09:26:47 | dbs | thanks for the doc clarification, miker_ |
| # | 09:26:49 | dbs hits the road |
| # | 09:27:14 | cerpy | aptitude install sshd |
| # | 09:27:25 | cerpy | sorry folks :) |
| # | 09:31:34 | dbs has quit IRC |
| # | 09:36:32 | Dyrcona | what's up with org_unit lassos? Shae just mentioned them in training. |
| # | 09:38:13 | berick | arbitrary groupings of org units for opac searches |
| # | 09:39:05 | Dyrcona | yes. she said there was a hold up on the UI. she said she check and let us know, but I'm here, so I thought I'd ask. |
| # | 09:39:20 | berick | right, there's no UI yet for defining the groups |
| # | 09:39:29 | tsbere is sitting next to Dyrcona and also curious |
| # | 09:40:16 | Dyrcona | tsbere: you can develop the ui, then. :) |
| # | 09:40:57 | berick | the opac supports them, once the groups are defined, but i'd be surprised if there wasn't still some testing/cleanup needed there as well. |
| # | 09:40:58 | tsbere | Can't be that hard....if anyone has documented how the backend needs things to look/work. |
| # | 09:41:36 | berick | actor.org_lasso |
| # | 09:41:45 | berick | in the db |
| # | 09:41:51 | Dyrcona | cool. we'll look. we use something like this in our current ils. |
| # | 09:42:02 | berick | once there, autogen will push the lassos out to the opac |
| # | 09:42:31 | berick | cool |
| # | 09:42:53 | jenny has joined #evergreen |
| # | 09:42:54 | berick | and.. actor.org_lasso_map |
| # | 09:43:26 | tsbere just noticed that part |
| # | 09:43:45 | tsbere | Looks like you name it, then create a pile of "this org unit is part of that" entries |
| # | 09:45:31 | miker_ | tsbere: that's correct |
| # | 09:45:43 | yboston has joined #evergreen |
| # | 09:46:25 | miker_ | I have plans for user (as in patron, instead of administrator) defined lassos, too |
| # | 09:47:00 | tsbere | As in "check these boxes to select where you are searching" stuff? |
| # | 09:48:21 | miker_ | tsbere: not on the fly, no |
| # | 09:48:43 | miker_ | "named groups of locations that I like to search at the same time" |
| # | 09:48:58 | miker_ | a "my account" interface to set them up |
| # | 09:49:37 | tsbere | Are there technical issues with on the fly variants? |
| # | 09:50:09 | miker_ | not other than it hasn't been designed :) |
| # | 09:50:44 | miker_ | and it's a totally different thing from stored groups in terms of backend implementation |
| # | 09:51:30 | tsbere | I could see implementing the patron level named groups as a "select the on the fly boxes for me" function, saved from "save this set" button |
| # | 09:51:51 | miker_ | sure, but the backend will still look different |
| # | 09:53:06 | miker_ | I'm not saying we shouldn't consider doing it, just that user stored sets are the part that are generally designed -- they are nearly identical to admin-defined lassos |
| # | 09:56:08 | dbs has joined #evergreen |
| # | 10:00:35 | dbs | org_unit_lassos++ # handy for the one place we use them |
| # | 10:01:04 | tsbere | the one place you use them? What place is that? |
| # | 10:01:29 | dbs | http://hrsrh.concat.ca/ - group of health libraries pulled together from across a couple of different branches of the org tree |
| # | 10:01:38 | tsbere | ahhh |
| # | 10:02:14 | dbs | note the sg=1 CGI param, that's about all there is to it. thanks again, miker_++ |
| # | 10:03:37 | tsbere | It doesn't show in the "choose a library to search" part though |
| # | 10:03:58 | bshum | Hmm, once entered, is there a special variable we have to enter in the URL to search an org lasso? Cause I'm not seeing any... yes, what tsbere said |
| # | 10:05:15 | bshum | Aha, sg is it |
| # | 10:05:53 | dbs | tsbere: right, that's part of the missing UI |
| # | 10:09:47 | tsbere | So I might have a new project. Funny, I just finished one, and the day after I submit the patch I get a new potential one. |
| # | 10:10:03 | tsbere has other things from the back burner unreleated to evergreen to take care of first, though |
| # | 10:11:00 | miker_ | tsbere: "it's not over until the fat lady ... er ... commit message is sent" |
| # | 10:11:02 | miker_ | ;) |
| # | 10:11:15 | dbs | tsbere++ # more, more! |
| # | 10:11:58 | tsbere | on my task list includes going out to a library and teaching them more about dreamweaver, attending a meeting or two, and finishing populating data for a site I built. >_> |
| # | 10:12:18 | tsbere has been neglecting some of that in the fight with automatic updating and other enhancements |
| # | 10:13:29 | dbs is happy just being awake after finishing off the upgrade to rel_1_6_1 last night, now reading booking docs (esi++) and figuring out how to get the 1.6.1.1 client on people's desks without having to individually edit each installed lang.dtd |
| # | 10:14:25 | dbs | I sure wish there was a staff client auto-update mechanism... :) |
| # | 10:14:27 | tsbere | individually edit wha? Wouldn't you just overwrite them? |
| # | 10:15:55 | dbs | the current 1.6.1.1 client has corrupted en-CA/lang.dtd (and probably others) |
| # | 10:16:35 | tsbere | ahh |
| # | 10:16:46 | dbs | the *@&@# "about button label" glitch prevents the staff client from even opening if you switched to a non-en-US locale |
| # | 10:17:24 | tsbere | That sounds annoying |
| # | 10:19:30 | tsbere | not sure the stuff my update code turns on will run if the client drops out to quickly, for the record. |
| # | 10:44:09 | phasefx | dbs: you need me to package a new 1.6.1.1? the process is also documented if you want to give it a try |
| # | 10:44:53 | dbs | yeah, we've created our own packages in the past but I thought you would be interested in knowing yet another reason 1.6.1.1 is still on the shelf :) |
| # | 10:45:15 | phasefx | ha |
| # | 10:46:17 | miker_ | I think it's 1.6.0/1.6.1 release time again :( |
| # | 10:46:43 | youdonotexist has joined #evergreen |
| # | 10:46:58 | tsbere | quick, see if my patch is compatible! :P |
| # | 10:47:08 | tsbere goes back to paying attention to the training/consulting session |
| # | 10:47:11 | dbs | miker_: on the bright side, at least the 1.6.1.1-1.6.1.2 upgrade script is basically there |
| # | 10:47:21 | miker_ | dbs++ # indeed |
| # | 10:47:39 | miker_ | dbs: just tracking commits, or have you confirmed with a diff? |
| # | 10:47:52 | dbs | tracking commits |
| # | 10:47:58 | miker_ | k |
| # | 10:48:13 | dbs | independent confirmation would be good + is probably a good general practice |
| # | 10:48:51 | dbs | I think I copied an extra function that didn't actually change into the mix, but it's a create or replace so shouldn't matter. |
| # | 10:49:18 | cerpy has quit IRC |
| # | 10:52:53 | mrpeters-isl has quit IRC |
| # | 10:55:12 | dbs | hrm Error: Transport error method=open-ils.booking.resources.create_from_copies, status=404 Source File: http://laurentian-test.concat.ca/js/dojo/dojo/dojo.js Line: 123 |
| # | 10:56:02 | phase_bb has joined #evergreen |
| # | 10:57:50 | phasefx | dbs: open-ils.booking in opensrf*.xml? |
| # | 10:58:17 | dbs | yep |
| # | 10:58:32 | dbs | and running |
| # | 10:59:46 | dbs | maybe restart apache... one sec |
| # | 11:02:32 | dbs | i can introspect open-ils.booking via srfsh |
| # | 11:03:03 | dbs | ahhh... opensrf*.xml includes opensrf_core.xml! |
| # | 11:04:11 | tsbere | yea, wildcards are like that. Or did you proc it as a regex? |
| # | 11:04:11 | dbs | that's better. phasefx++ |
| # | 11:04:19 | phasefx | sorry, taking keyboard shortcuts with stuttering ISP :) |
| # | 11:04:57 | dbs | phasefx: no, not your fault, I cavalierly dismissed opensrf_core.xml without thinking "web page is calling this service, needs to be a public service" |
| # | 11:07:06 | phasefx | for those following trunk, postgres 8.3 will no longer work |
| # | 11:07:50 | tsbere | It won't? |
| # | 11:07:59 | dbs | oh really? it was working just a couple of days ago |
| # | 11:08:19 | tsbere | I would be concerned....but we have been using 8.4 for months. |
| # | 11:08:39 | phasefx points at miker_ |
| # | 11:09:06 | miker_ | dbs: there are things that would break before (right-trunc search), but now there are 8.4 features in direct use in the schema |
| # | 11:09:25 | miker_ | CREATE FUNCTION ... RETURNS TABLE (...) and RETURN QUERY ... |
| # | 11:09:36 | dbs | ah, didn't catch those ones |
| # | 11:09:47 | dbs is totally in favour of moving to 8.4 as a minimum |
| # | 11:09:58 | miker_ | for 2.0, me too |
| # | 11:11:03 | tsbere | Why limit to 2.0? 8.4 min for 1.6.x releases! |
| # | 11:11:14 | dbs | for rel_1_6_1 - is it okay to run action_trigger_runner every 2 minutes or so? |
| # | 11:11:52 | dbs | we've been running it once a day, just using it for overdue / predue notices, but now that we're rocking the password reset we need some snappier response times |
| # | 11:11:53 | miker_ | dbs: I see no reason not to ... it'll block concurrent runners, so you're really just saying "run it constantly" |
| # | 11:12:26 | dbs | okay, thanks for the reassurance to my sleep-deprived brain |
| # | 11:13:24 | miker_ | just make sure you're running it from only one box, and preferrably your "utility" server :) |
| # | 11:14:28 | dbs is going to try to put the 1.6.0 - 1.6.1 upgrade notes together, there are plenty of gotchas to stumble through that some good upgrade notes will hopefully keep stupid questions like mine to a minimum |
| # | 11:14:40 | berick | dbs: if you don't want a steady stream of overdue notices, you should still have a nightly run that does --process-hooks and leave that out of the 2-minute runs. |
| # | 11:16:04 | berick | --process-hooks being the passive hook event creation |
| # | 11:19:02 | dbs | done and done. went with every minute, just for fun |
| # | 11:19:12 | dbs | thanks, miker_++ and berick++ |
| # | 11:19:15 | dbs | @karma |
| # | 11:19:15 | pinesol | dbs: Highest karma: "dbs" (218), "miker_" (130), and "berick" (103). Lowest karma: "----------------------------------" (-16), "my_laptop" (-15), and "rhel" (-3). You (dbs) are ranked 1 out of 682. |
| # | 11:19:28 | tsbere | will a non --process-hooks one block the --process-hooks one? |
| # | 11:19:29 | dbs | bah, that's so wrong |
| # | 11:20:03 | phasefx | it's your good looks |
| # | 11:20:30 | miker_ | and amazing singing voice |
| # | 11:21:21 | dbs | tsbere: I set the non --process-hooks one to not run during the hour that the --process-hooks one runs |
| # | 11:21:35 | dbs | * 01-05,07-23 * * * |
| # | 11:22:44 | tsbere | how far into the hour does the process-hooks one run? |
| # | 11:23:31 | berick | tsbere: looks like --process-hooks will run even if the lockfile is present |
| # | 11:23:46 | berick | well, wait, sorry |
| # | 11:23:57 | miker_ | berick: that's just trunk, IIRC |
| # | 11:24:00 | dbs | Hopefully the people desperately wanting password resets at 6:00 am can wait until 7:00 for their password |
| # | 11:24:21 | berick | if not in process-hooks mode, the process will go away. if in process-hooks mode, it will block and wait. and, yes, that's only in trunk |
| # | 11:24:52 | dbs | tsbere++ # asking the hard questions |
| # | 11:25:25 | berick | that appears to be the same in 1.6.1 |
| # | 11:28:02 | tsbere | alternatively, add something like $([ $(date +%H) == "06" ] && echo "--process-hooks") to the command line or something and have it still run every minute. |
| # | 11:31:38 | tsbere used tricks like that in his staff client makefile changes |
| # | 11:48:33 | tsbere | with a little more work I suppose you could use a secondary file. "If it was touched today, don't include the parameter. If it wasn't, and it is hour x, touch it and include the parameter" or something |
| # | 11:53:55 | _bott_ | Any good reason not to clean out old rows in action.hold_copy_map for holds already fulfilled, canceled, or otherwise dead? |
| # | 11:54:54 | phasefx | folks asking why such and such was or was not captured? |
| # | 11:55:33 | miker_ | _bott_: they should get cleared out when the item is captured or the hold canceled (if it happens via actual EG codepaths) |
| # | 11:56:00 | miker_ | if it's done magically under the covers, then clear it out, yes |
| # | 11:56:33 | _bott_ | hmm... wonder why it's not. Nearly 68% of our rows are related to already fulfilled requests |
| # | 11:56:37 | jeff | i've seen instances of it not being cleared out. |
| # | 11:57:10 | jeff | of course, we're talking about the same system. :) |
| # | 11:57:20 | miker_ | well, clear away. and if you find a codepath that should, but doesn't, please point it out or send a patch! :) |
| # | 11:57:49 | _bott_ | will do. have to dig more later, just wanted to know if there was any good reason they wouldn't/shouldn't |
| # | 11:58:10 | miker_ | no good reason that I can think of |
| # | 11:58:41 | tsbere | do they hang around until the item is checked back in? |
| # | 11:58:47 | jeff | tsbere: they should not. |
| # | 11:58:51 | miker_ | phasefx: btw, that's ostensibly what the unfilled_hold table is for |
| # | 11:59:00 | tsbere | I didn't ask if they should, I asked if that is what is happening |
| # | 11:59:01 | miker_ | tsbere: shouldn't |
| # | 11:59:06 | miker_ | heh |
| # | 11:59:34 | jeff | i've seen them hang around for longer, and while i don't know the cause, i also don't know what at checkin would remove 'em. |
| # | 12:00:11 | miker_ | once the hold is "done" we don't care about the potentials map, so we don't ever look at it ... IOW, checkin should not go near that (for completed/dead holds) |
| # | 12:01:06 | tsbere just knows that what should be happening and what is actually happening may not even be close to the same thing |
| # | 12:01:16 | jeff | tsbere: *nod* |
| # | 12:02:03 | Dyrcona wants lunch. |
| # | 12:02:18 | tsbere agrees with Dyrcona on that one |
| # | 12:03:12 | jeff | there's a crosstab/pivot-table implementation in clark-kent.pl, and i've seen some suggested functions in postgresql. does anyone here have a favorite implementation? |
| # | 12:05:29 | Dyrcona | so, my understanding is circ modifier started out as a modifier, but has now morphed into an item type (because of how people use it). |
| # | 12:05:52 | Dyrcona | i also don't really know why I just typed that in. ignore me. |
| # | 12:06:48 | jeff | Dyrcona: some people use shelving location as a kind of "item type" |
| # | 12:07:12 | Dyrcona | jeff: that's what someone else in this training session is planning to do. |
| # | 12:07:39 | Dyrcona | i was hoping to use circ modifier just as a modifier, but I'm not sure it would really "help" to do that. |
| # | 12:08:11 | jeff | Dyrcona: due to the way that many libraries treat different item types with different circ rules, sometimes circ modifier does end up resembling item type -- but we have some things that are circ mod BOOK that are most certainly not books, but circ with the default/normal/same-as-BOOK rules |
| # | 12:09:22 | jeff | if you catalog all of your cake pans with circ modifier BOOK and then decide you're going to allow a different number of renewals later, you might mass-update the circ modifier on the cake pan copies (based on criteria like shelving location or call number) |
| # | 12:09:30 | jamesrf has joined #evergreen |
| # | 12:09:34 | phasefx | the intention was to give folks enough knobs and dials that they wouldn't have to have "item types" with inevitably overloaded meanings |
| # | 12:09:35 | Dyrcona | mmm, cake. |
| # | 12:09:43 | tsbere | the cake is a lie |
| # | 12:09:44 | jeff | or you might just make a CAKE-PAN circ modifier to begin with, and have BOOK and CAKE-PAN get the same rules. |
| # | 12:11:09 | dbs | which is easy to do with scripts :) |
| # | 12:11:42 | Dyrcona | i like the idea of scripts, but i'm going to be expected to try in-db rules, first. |
| # | 12:12:31 | Dyrcona | i played with scripts in the past when we were first looking at evergreen, before in-db rules. i liked that i could make it do whatever I wanted it to. |
| # | 12:12:36 | phasefx | or folks could, you know, have circ modifiers be very normalized based on behavior, and use item editor templates to replicate item type mentalities |
| # | 12:12:58 | dbs would like to move to in-db rules to distribute circ maintenance responsibility closer to the circ desk, but we just ain't there yet |
| # | 12:13:01 | Dyrcona | well, ttfn. lunch at last. |
| # | 12:13:11 | Dyrcona has left #evergreen |
| # | 12:29:25 | phase_bb has quit IRC |
| # | 12:48:55 | Granitize1 has joined #evergreen |
| # | 12:57:19 | Granitize1 has quit IRC |
| # | 12:57:26 | Granitize1 has joined #evergreen |
| # | 12:57:29 | eby has quit IRC |
| # | 13:01:07 | eby has joined #evergreen |
| # | 13:03:38 | dbs | Grr. Still can't get the "Delete address" button to work in the Dojo-based patron editor in rel_1_6_1; looks like type="addr-template" on the rows is getting stripped onLoad, and the X button doesn't get the onclick attribute set (according to firebug) |
| # | 13:06:45 | jamesrf has quit IRC |
| # | 13:07:47 | jamesrf has joined #evergreen |
| # | 13:12:30 | Dyrcona has joined #evergreen |
| # | 13:22:35 | bshum | When looking at the xact_start timestamp in the action.circulation table, why is it that sometimes it goes beyond seconds and gets drilled down to milliseconds? Whereas in other places it's only to the nearest second? |
| # | 13:23:12 | jeff | when it defaults to now() it will gain milliseconds. |
| # | 13:23:35 | jeff | there are various paths that lead to the default, others are specified elsewhere at a lower resolution. |
| # | 13:24:41 | phasefx | dbs: did you see changeset 17023? |
| # | 13:25:06 | dbs | phasefx: yep. that was already part of http://svn.open-ils.org/trac/ILS/browser/branches/rel_1_6_1/Open-ILS/web/js/ui/default/actor/user/register.js |
| # | 13:27:54 | bshum | jeff: So you're suggesting that it only defaults to "now" with milliseconds if it was checked out differently somehow than the other paths? |
| # | 13:28:38 | tsbere | I think that it only defaults to "now" with milliseconds when the database does "now" compared to any other code. |
| # | 13:28:43 | tsbere has not actually looked, though |
| # | 13:28:50 | jeff | bshum: correct |
| # | 13:29:06 | jeff | bshum: checkout vs renewal might be one difference. |
| # | 13:29:07 | dbs | phasefx: looks like it never hits that final "else" to add the onclick attribute |
| # | 13:30:15 | dbs | I think there are also triggers to adjust the due time of a given item - so rather than now() + 24 hours, it becomes now + 24 hours + whatever it takes to get to 23:59:59 |
| # | 13:30:41 | jeff | yeah. that was fun at our 1.6 upgrade. |
| # | 13:30:45 | dbs | also possibly if you override the due date in the checkin window |
| # | 13:31:00 | jeff | i'm still not sure there are good 1.4 to 1.6 upgrade scripts to fix existing circs that didn't get the 23:59 from the trigger. |
| # | 13:31:17 | jeff | dbs: i don't think overriding due date would change the xact_start, though. maybe. |
| # | 13:31:36 | bshum | jeff: The main reason we're checking is because we have three "missing" items that were checked out to a patron that claims they never checked them out. |
| # | 13:31:37 | dbs | oh, xact_start - duh. not working on enough sleep |
| # | 13:31:57 | bshum | And the xact_start times are "different" in this particular way vs. other items we've seen checked out. |
| # | 13:32:10 | miker_ | bshum: offline checkout, perhaps? |
| # | 13:32:14 | bshum | Could be as simple as the patron not telling the truth to the circ staff, but it could also be bad cic staff procedure. |
| # | 13:32:15 | jeff | bshum: and the three circs in question have what precision on their xact_start times? milliseconds or seconds? |
| # | 13:32:30 | jeff | miker_++ # offline suggestion |
| # | 13:32:39 | bshum | jeff: an example circ xact_start timestamp: "2010-07-16 16:34:22.238104-04" |
| # | 13:32:40 | tsbere | other thoughts: any sip clients/self checks kicking around? |
| # | 13:33:08 | jeff | bshum: is that example one of the three 'odd' circs in dispute? |
| # | 13:33:19 | dbs | hah. deleting that } else { clause in rel_1_6_1 enables the X to delete addresses |
| # | 13:33:29 | bshum | jeff: Yes, other timestamps for that particular patron look like this instead: "2010-07-16 16:33:00-04" |
| # | 13:33:40 | bshum | jeff: course those items checked back in properly. |
| # | 13:33:57 | bshum | tsbere: As far as we know, there shouldn't be any selfchecks in use at the library in question. |
| # | 13:34:08 | jeff | bshum: and are any of the circs renewals? |
| # | 13:34:23 | miker_ | bshum: I'd suggest scanning the logs around that time for the barcodes of the user and item in question |
| # | 13:35:07 | miker_ | the opensrf logs ... start with activity and gateway |
| # | 13:40:40 | bshum | miker_: Would it appear as barcodes or associated IDs? (no entries found in the logs specified thus far) |
| # | 13:40:51 | bshum | jeff: They don't appear to be renewals |
| # | 13:41:56 | miker_ | bshum: if the barcodes were scanned, then you'll see them in there at least for object retrieval |
| # | 13:42:27 | tsbere | bshum: You said no self checks now, but what about a self check demo that might have been run? #digging for options |
| # | 13:43:07 | bshum | miker_: Aha, nevermind, I found them, they were in a different activity.log file; split by dates and I had the wrong date open |
| # | 13:44:30 | bshum | miker_: It says that the item was checked out with open-ils.circ.checkout.full |
| # | 13:44:44 | bshum | miker_: Or something like that. I can probably get the entire line |
| # | 13:45:04 | bshum | The patron ID number matches up to the patron who's saying they never got it. |
| # | 13:45:23 | miker_ | bshum: find the auth token (first param) and grep for that. the /first/ instance of that in the activity log will tell you the staff that checked the item out to the patron |
| # | 13:48:35 | bshum | miker_: Would that be the part that comes after the IP address or before it buried in the ACT:**** portion? |
| # | 13:49:17 | bshum | Ah, nevermind, after the IP address must be. |
| # | 13:49:46 | tsbere | With static ips, that ip address might tell you where they were standing too. |
| # | 13:50:27 | bshum | tsbere: That's a pretty neat idea, ha. |
| # | 13:50:48 | bshum | Well, it looks like it's the correct circ staff account that we gave the library for circulation. |
| # | 13:51:01 | bshum | Which matches up with what got stored in the action.circulation record entry. |
| # | 13:52:08 | tsbere assumes that workstation id probably identifies what computer, but a laptop you might want the IP too |
| # | 13:52:46 | miker_ | workstation will identify the computer. can only register them once |
| # | 13:53:06 | tsbere | Yes. But a laptop may be able to log in from, oh, anywhere it has a net connection. |
| # | 13:53:34 | tsbere | So if the workstation was a laptop, the IP might indicate where it was when it did the circ |
| # | 13:54:05 | phasefx | you can re-register if you try hard enough. I think the override prompt will re-use the existing workstation id rather than delete/re-create the workstation |
| # | 13:54:07 | dbs | bshum's home, no doubt |
| # | 13:54:07 | tsbere | (Yes, our libraries have a laptop or two that run our ILS) |
| # | 13:55:13 | dbs | hmm. concerns about making "Active" a required field for the Dojo patron registration? |
| # | 13:55:49 | tsbere | How is "Active" defined? A checkbox? A dropdown? A date? |
| # | 13:55:52 | miker_ | dbs: in trunk via the IDL required setting? |
| # | 13:56:00 | dbs | (I was surprised when I registered a patron using "Show only required fields" then couldn't search for them in crazy search - 'cause Active is not required) |
| # | 13:56:05 | dbs | in 1.6.1. it's a checkbox |
| # | 13:56:45 | tsbere | How does "required" handle a checkbox that wasn't checked? (or is the required list just a "show these" check?) |
| # | 13:56:47 | phasefx | it's not defaulting to true either? |
| # | 13:56:49 | dbs | I would daresay we might even want it to default to true, if you're going to the trouble of registering a patron |
| # | 13:56:54 | dbs | not in 1.6.1 |
| # | 13:58:08 | tsbere | If you are going to make it required, it might be best to change to an option box or set of radio buttons, as "required" usually means "has a value" and empty checkboxes don't have values |
| # | 13:58:42 | dbs | "Required" in this context means "show or do not show" |
| # | 13:59:18 | tsbere | Concern would be "is it being used for something else too, like in a validator check" |
| # | 13:59:59 | dbs | Feel free to be concerned |
| # | 14:00:24 | jeff | hrm. anyone remember the "circ_lib on renewals is set to a different lib than the original circ" thing? Am I imagining it? |
| # | 14:01:28 | dbs | yeah, rings a bell. if the renewal is done via the OPAC it was using the user's home_ou I think? |
| # | 14:01:54 | jeff | okay, that matches what i remember. going to try and confirm here. do you remember if there's been a fix/patch? |
| # | 14:02:52 | dbs | I /think/ we fixed it in the renew script to just use the item's circ_lib |
| # | 14:04:36 | tsbere | dbs: You talking in register_table.tt2, or did I miss another interface you are referring to? (on the active being required thing) |
| # | 14:04:51 | dbs | correct |
| # | 14:05:43 | tsbere | web/js/ui/default/actor/user/register.js uses that to determine if it should be required too, then, doesn't it? |
| # | 14:07:30 | dbs | That's kind of the point, yes |
| # | 14:07:51 | dbs | So when you click "Show only required fields", only the required fields are shown. |
| # | 14:07:51 | tsbere | Does any kind of validator then use that information to determine what has to be filled in? |
| # | 14:08:14 | tsbere is splitting his attention and may have missed something |
| # | 14:08:19 | dbs | Isn't that the same question you asked at 13:58? |
| # | 14:08:53 | tsbere | possibly. I re-opened the table file 8 times thinking it was a different file each time due to paying attention to the training session here. >_> |
| # | 14:09:08 | miker_ | dbs: I have no concerns about either (making "required" and setting the default to checked) |
| # | 14:09:10 | Dyrcona | pay attention to Shae.... |
| # | 14:09:36 | dbs | So, the answer is - no. You can leave the checkbox unchecked ("inactive") and save the user. |
| # | 14:10:00 | dbs | No validation function says "Checkbox is null, oh no, stop the presses." |
| # | 14:10:16 | yboston has quit IRC |
| # | 14:10:31 | dbs | Now pay attention to Shae so you can fill me in on everything I don't understand about holds :) |
| # | 14:11:09 | tsbere | actually, dbs, just noticed something. |
| # | 14:11:34 | tsbere | don't make active "required", make it "show". As in instead of required='required' use required='show'? |
| # | 14:11:50 | tsbere noted some address fields using that |
| # | 14:12:52 | dbs | ah, that's a good suggestion |
| # | 14:15:17 | tsbere | I was wondering why the check was "required is exactly required" when the "not required" entries omitted it. Quick search, found a show. |
| # | 14:15:51 | dbs | Yep, I should have done a quick search. |
| # | 14:16:46 | jenny1 has joined #evergreen |
| # | 14:19:27 | jenny has quit IRC |
| # | 14:21:16 | tsbere thinks that required='duh' could also work for showing it, but not making it trigger the "marked as required" code |
| # | 14:21:27 | jenny1 has left #evergreen |
| # | 14:22:39 | dbs is now known as duhbs |
| # | 14:23:48 | jeff | bah. search.cpan.org seems down. |
| # | 14:24:57 | tsbere | define down? Not searchable? |
| # | 14:25:02 | tsbere loaded the homepage ok |
| # | 14:25:49 | tsbere searched for marc, got results |
| # | 14:26:01 | tsbere | need me to search for something for you jeff? |
| # | 14:26:14 | collum has left #evergreen |
| # | 14:27:09 | jeff | tsbere: no thanks. seems to have been a temporary localized issue. |
| # | 14:30:33 | duhbs | gah. a brace too far |
| # | 14:30:59 | tsbere | a brace too far? Remove an extra one when adjusting code? |
| # | 14:32:01 | tsbere | Oh, the closing brace you removed with the else? *just looking at the commit messages* |
| # | 14:34:02 | duhbs | correct |
| # | 14:34:07 | duhbs | and corrected |
| # | 14:44:11 | duhbs is now known as dbs |
| # | 14:44:51 | dbs stops obsessing about 1.6.1 stuff and goes off to set up some printers on a colleague's workstation. PARTY TIME |
| # | 14:45:28 | jamesrf has quit IRC |
| # | 15:04:45 | jeff | contrib/tablefunc.sql crosstab success! |
| # | 15:06:10 | tsbere | yay! What are you using it for? |
| # | 15:07:00 | pmplett is now known as pmpafk |
| # | 15:07:57 | jeff | tsbere: simple crosstab/pivot-table output of patron "home locations" in column 1, then columns for each library with uses during a time period as the data. |
| # | 15:08:16 | tsbere | ahh |
| # | 15:08:31 | jeff | automating in postgresql means i don't have to automate it in openoffice/excel |
| # | 15:08:51 | tsbere | very true. |
| # | 15:18:01 | mck9 has left #evergreen |
| # | 15:19:06 | tsbere wonders if there would be an objection to adding a new column to the lasso table there |
| # | 15:23:28 | mck9 has joined #evergreen |
| # | 15:24:34 | jeff | tsbere: what do you have in mind? |
| # | 15:34:19 | dbs | berick: good relabelling, that had confused the hell out of me |
| # | 15:36:04 | tsbere | jeff: Was thinking an opac visible flag |
| # | 15:36:40 | berick | dbs: you were not alone |
| # | 15:39:20 | Dyrcona | time to go. |
| # | 15:39:24 | Dyrcona has quit IRC |
| # | 15:43:38 | Granitize1 has quit IRC |
| # | 15:48:09 | jamesrf has joined #evergreen |
| # | 15:54:30 | dbs | "disclusion": I like it! less negative than "exclusion" |
| # | 15:56:53 | dbs discludes himself from work |
| # | 15:56:55 | dbs has quit IRC |
| # | 15:57:19 | phasefx | this ends this disclussion |
| # | 15:58:24 | miker_ | unfortunately, it means both "to exclude" and "to make known" ;) |
| # | 16:03:35 | jenny has joined #evergreen |
| # | 16:03:56 | Granitize1 has joined #evergreen |
| # | 16:03:59 | jenny has left #evergreen |
| # | 16:13:52 | Granitize1 has quit IRC |
| # | 16:18:46 | alxp has quit IRC |
| # | 16:21:55 | bshum has quit IRC |
| # | 16:39:20 | Melissa has left #evergreen |
| # | 16:39:54 | agJohn | Is there a tarball of the 1.6.1.1 branch available somewhere or is an SVN checkout the only option (this is for the migration install for NC, gmcharlt)... |
| # | 16:41:07 | gmcharlt | agJohn: unofficially, http://www.evergreen-ils.org/downloads/Evergreen-ILS-1.6.1.1.tar.gz |
| # | 16:41:36 | agJohn | Unofficial is great. Any patches that should be applied to this? |
| # | 16:43:05 | phasefx | I think the I18N build for it may be broken |
| # | 16:44:35 | agJohn | Ah; well, I guess I'll live w/o that for now.... |
| # | 16:44:41 | phasefx | that might just apply to staff clients built with it, don't know |
| # | 16:44:56 | jeff | phasefx: is there any current way to remote the staff client into navigating to a record? |
| # | 16:45:25 | jeff | phasefx: i'm thinking that there is not, but i know you're more familiar with some of the little experiments |
| # | 16:46:02 | phasefx | jeff: I'm not sure what you mean |
| # | 16:47:18 | phasefx | if you embed a telnet server into the client (I've been tempted), with a bit of effort you can do almost anything |
| # | 16:47:43 | atz | haha... still working the ncurses client angle |
| # | 16:48:05 | phasefx | I've been wanting to refactor things so that they work with the javascript shell and tab completion |
| # | 16:48:46 | jenny1 has joined #evergreen |
| # | 16:49:26 | phasefx | the control+shift+f7 trick was also a step in that direction.. would love for a javascript shell to pull down and share context with the window from which it was invoked |
| # | 16:52:07 | gmcharlt | atz: phasefx: I have a special fondness of terminal-based UIs for circ functions |
| # | 16:52:20 | jenny1 has left #evergreen |
| # | 16:52:47 | phasefx | gmcharlt: I think it would rock :) |
| # | 16:53:13 | atz | what was it in multilis?... gold-button+F12 or something? |
| # | 16:54:29 | phasefx | with an ncurses client, I'd want to make a Quake-Console that could pull down over the GUI client ;) |
| # | 16:59:43 | jeff | phasefx: from the command line, you can tell your running web browser to "open this url". i was wondering if similar function existed in the staff client, for "go to this record" |
| # | 17:00:12 | jeff | after thinking on it a bit, i'd need a helper app on the workstation to bridge the gap, so it's less useful. |
| # | 17:00:23 | tsbere | why would it need a helper app? |
| # | 17:01:08 | jeff | intended use case was for a browser-based report not running in the staff client could allow navigation to a bib record in the staff client. |
| # | 17:01:35 | tsbere | Ok |
| # | 17:01:51 | tsbere | Custom url handler in the browser combined with information from https://developer.mozilla.org/en/XULRunner/CommandLine#For_Single_Instance_Application and bam, no helper app required! |
| # | 17:01:52 | jeff | probably better to view the report in the staff client if that function is desired. |
| # | 17:02:15 | jeff | tsbere: no, the staff client just becomes the helper app ;-) |
| # | 17:03:14 | tsbere | jeff: That depends. The staff client as an extension might be part of the browser, and grabbing the URL handler directly. Or are you calling the browser itself a helper app then? |
| # | 17:03:49 | phasefx | yeah, I never tried anything like that. If we wanted a poor man's IPC, we could poll the contents of a file and execute it :) |
| # | 17:04:36 | jeff | just observing that browser(meta file with special mime type) -> helper app -> staff client becomes browser(custom url scheme/handler) -> staff client |
| # | 17:04:54 | jeff | and that in the second scenario, the "staff client just becomes the helper app" |
| # | 17:05:32 | phasefx imagines the fun delay of instant gratification if you have to login to the client |
| # | 17:06:23 | tsbere | Well, in that case, throw security out the window and embed a username/password into the special url! #badideadon'tdoit |
| # | 17:24:46 | pmpafk is now known as pmplett |
| # | 17:28:28 | Callender_ has joined #evergreen |
| # | 17:29:17 | berick | dojo-ified dom-node text hilighter http://dev198.esilibrary.com/~berick/hilite.html. it borrows heavily from http://johannburkard.de/blog/programming/javascript/highlight-javascript-text-higlighting-jquery-plugin.html |
| # | 17:29:41 | berick | will move into openils.something JS lib if it passes muster |
| # | 17:31:20 | atz | yeah, that jquery plugin is the bomb |
| # | 17:31:29 | atz | fast, simple and powerful |
| # | 17:32:28 | Dyrcona has joined #evergreen |
| # | 17:32:36 | Dyrcona | guess everyone is gone. |
| # | 17:33:12 | Dyrcona | however, i think i found a bug in biblio.map_authority_linking |
| # | 17:33:51 | Dyrcona | It looks for any field that has a subfield code="0", and that is incorrect since most fields are not supposed to have a code="0" |
| # | 17:34:18 | Callender has quit IRC |
| # | 17:34:18 | moodaepo has quit IRC |
| # | 17:34:20 | Callender_ is now known as Callender |
| # | 17:34:41 | Dyrcona | I think it should only look in the 1xx datafields, so I'll post a bug with a patch when I get some changed code that works. |
| # | 17:34:52 | gmcharlt | Dyrcona: subfield $0 has a specific meaning - it's meant to contain a link to the authority record controlling the heading |
| # | 17:35:10 | Dyrcona | right, but only in 1xx, according to marc |
| # | 17:36:01 | Dyrcona | as a side effect, this has exposed some bad subfields in my records. |
| # | 17:36:01 | olgagirl has joined #evergreen |
| # | 17:36:06 | moodaepo has joined #evergreen |
| # | 17:36:30 | eby has quit IRC |
| # | 17:36:30 | pmplett has quit IRC |
| # | 17:36:31 | mck9 has quit IRC |
| # | 17:36:31 | youdonotexist has quit IRC |
| # | 17:36:31 | phasefx has quit IRC |
| # | 17:36:33 | rsinger has quit IRC |
| # | 17:36:33 | jeffdavis has quit IRC |
| # | 17:36:33 | Dmagick-home has quit IRC |
| # | 17:36:34 | gdunbar has quit IRC |
| # | 17:36:34 | kbeswick_ has quit IRC |
| # | 17:36:34 | gmcharlt has quit IRC |
| # | 17:36:36 | sylvar has quit IRC |
| # | 17:36:36 | bradl has quit IRC |
| # | 17:36:37 | _bott_ has quit IRC |
| # | 17:38:02 | Dyrcona | the reason that i say it is a bug is because it has caused my bib load to fail when there was a 245 with an erroneous subfield $0. $0 is not supposed to be used in a 245, and in this case, the $0 should have $a. |
| # | 17:38:59 | Dyrcona | i'm still looking through the MARC standard, but from what I can see $0 is only valid in 1xx and some 7xxx. If Evergreen is using $0 as an extension to MARC, it is wrong imho. |
| # | 17:39:19 | mck9 has joined #evergreen |
| # | 17:39:19 | eby has joined #evergreen |
| # | 17:39:19 | youdonotexist has joined #evergreen |
| # | 17:39:19 | gdunbar has joined #evergreen |
| # | 17:39:19 | pmplett has joined #evergreen |
| # | 17:39:19 | jeffdavis has joined #evergreen |
| # | 17:39:19 | phasefx has joined #evergreen |
| # | 17:39:19 | rsinger has joined #evergreen |
| # | 17:39:19 | Dmagick-home has joined #evergreen |
| # | 17:39:19 | kbeswick_ has joined #evergreen |
| # | 17:39:19 | _bott_ has joined #evergreen |
| # | 17:39:19 | gmcharlt has joined #evergreen |
| # | 17:39:19 | sylvar has joined #evergreen |
| # | 17:39:19 | bradl has joined #evergreen |
| # | 17:39:49 | gmcharlt has left #evergreen |
| # | 17:39:57 | gmcharlt has joined #evergreen |
| # | 17:40:50 | berick has quit IRC |
| # | 17:41:57 | gmcharlt | Dyrcona: 1xx in the bib records, you mean? |
| # | 17:42:07 | Dyrcona | yes. |
| # | 17:42:18 | jenny has joined #evergreen |
| # | 17:42:41 | berick_ has joined #evergreen |
| # | 17:42:48 | berick_ is now known as berick |
| # | 17:42:48 | gmcharlt | Dyrcona: lots of ohter bib fields can have $0 and can be linked to authority records |
| # | 17:42:48 | gmcharlt | the 6xx fields in particular |
| # | 17:42:55 | Dyrcona | the function i mentioned above is looking for any datafield in the bib records with a $0 and trying to convert that to an integer. |
| # | 17:43:15 | gmcharlt | and many of the 7xx |
| # | 17:43:39 | Dyrcona | i think the function should only check the relevant fields, and not just any field. |
| # | 17:44:01 | Dyrcona | i am going to fix my records before loading. |
| # | 17:44:01 | gmcharlt | and 8xx ... |
| # | 17:44:30 | Dyrcona | i'm doing a survey on the LOC web site. I use it so much I thought I'd do their survey. |
| # | 17:44:40 | gmcharlt | not necessarily disagreeing about it checking only the fields that (per standard) can be authority controlled |
| # | 17:44:47 | Dyrcona | so, that interrupted my standards surfing. |
| # | 17:45:35 | Dyrcona | well, its blowing up on bad bib records, and this is a problem when 1 bad record prevents 9,999 others from loading. |
| # | 17:46:09 | gmcharlt | blowing specifically on the int conversion? |
| # | 17:47:15 | gmcharlt | it would be a bad assumption to assume that $0 can only contain an int, as bib records in the wild that have $0 pointing to national authority records ... aren't restricted to just numeric authority record IDs |
| # | 17:49:35 | Dyrcona | i'll past the exception message, it is blowing on the int conversion. |
| # | 17:50:11 | Dyrcona | 2010-08-12 16:19:36 EDT bib: [exception] org.postgresql.util.PSQLException: ERROR: invalid input syntax for integer: "4559C0" |
| # | 17:50:11 | Dyrcona | Where: SQL function "map_authority_linking" statement 2 |
| # | 17:50:12 | Dyrcona | SQL statement "SELECT biblio.map_authority_linking( $1 , $2 )" |
| # | 17:50:12 | Dyrcona | PL/pgSQL function "indexing_ingest_or_delete" line 20 at PERFORM |
| # | 17:51:12 | Dyrcona | dbs mentioned it looking for authority.record_entry.id in $0 the other day, but it should probably be looking for ARN instead, no? |
| # | 17:54:34 | Dyrcona | if you think it is worth pursuing, i'll be happy to make a patch that searches arn and/or at least doesn't blow up on int conversion. |
| # | 18:01:17 | jenny has left #evergreen |
| # | 18:14:05 | gmcharlt | Dyrcona: are.id is better, but a patch that makes it not blow up on it conversion would be god |
| # | 18:14:59 | Dyrcona | ok. sometime by next tuesday, i should be able to do it. i'm in consulting with shae again tomorrow. |
| # | 18:27:18 | miker_ | Dyrcona: I'll address the int conversion tonight or tomorrow, but we won't be changing from are.id. please see the -dev list for in-depth discussion on the implementation, and its standards compliance (also, only checking {some subset of all bib fields} will probably be coming with funded development) |
| # | 18:28:30 | Dyrcona | ok. i'm changing my bib loader/cleaner to replace $0 with $a where $0 is not valid for a field. it should avoid this int conversion problem with our records, i hope. |
| # | 18:28:58 | rsinger has quit IRC |
| # | 18:29:02 | Dyrcona | i haven't yet seen an exception thrown with a record that had $0 where it should be. |
| # | 18:29:31 | Dyrcona | rather, i've only seen the problem with bad records. |
| # | 18:30:46 | rsinger has joined #evergreen |
| # | 18:32:02 | youdonotexist has quit IRC |
| # | 18:48:56 | miker_ | Dyrcona: right. I just need to make sure I don't try to cast, say, an empty string to an int. Once that check is in, you won't get failures even on bad records |
| # | 19:02:59 | Dyrcona | ok. that's it for me. good night/day everyone. |
| # | 19:03:02 | Dyrcona has quit IRC |
| # | 19:29:59 | jamesrf has quit IRC |
| # | 19:59:43 | tsbere has left #evergreen |
| # | 20:01:43 | jamesrf has joined #evergreen |
| # | 20:14:07 | pmplett has quit IRC |
| # | 20:19:52 | miker_ | @later tell Dyrcona I think you'll want to turn on the 'ingest.disable_authority_linking' flag during initial import, turn it back on, and then consider using Open-ILS/src/support-scripts/authority_control_fields.pl to relink them properly |
| # | 20:19:52 | pinesol | miker_: The operation succeeded. |
| # | 20:21:33 | miker_ | @later tell Dyrcona also, will be committing a protection against bad $0 data |
| # | 20:21:33 | pinesol | miker_: The operation succeeded. |
| # | 21:00:21 | miker_ | @later tell Dyrcona protection committed |
| # | 21:00:21 | pinesol | miker_: The operation succeeded. |
| # | 21:34:44 | rsinger_ has joined #evergreen |
| # | 21:38:16 | rsinger has quit IRC |
| # | 21:38:16 | rsinger_ is now known as rsinger |
| # | 21:47:13 | rsinger has quit IRC |
| # | 22:13:54 | dbs has joined #evergreen |
| # | 22:14:42 | dbs | miker_++ # auth-link fix |
| # | 22:18:57 | dbs has quit IRC |
| # | 22:24:22 | dbs has joined #evergreen |
| # | 22:46:01 | jamesrf has quit IRC |
| # | 23:20:15 | jamesrf has joined #evergreen |
| # | 23:52:50 | dbs | agJohn: there are definitely some post-1.6.1.1 patches that you will want to apply. certainly the 1.6.1.1-1.6.1.2 upgrade SQL, plus other bits |
April 10-13