| # | Time | Nick | Message |
|---|
| # | 00:04:45 | tsbere has quit IRC |
| # | 00:04:59 | tsbere has joined #evergreen |
| # | 00:10:23 | bshum has quit IRC |
| # | 02:30:11 | tpham has quit IRC |
| # | 03:17:42 | dave-esi-away has quit IRC |
| # | 03:18:00 | Callender has quit IRC |
| # | 05:51:15 | rickd_ has quit IRC |
| # | 06:42:56 | rickd_ has joined #evergreen |
| # | 06:45:33 | rickd_ has quit IRC |
| # | 06:49:00 | alxp has joined #evergreen |
| # | 07:33:41 | rickd_ has joined #evergreen |
| # | 07:52:34 | Dyrcona has joined #evergreen |
| # | 07:55:50 | ebyr has joined #evergreen |
| # | 08:44:07 | kmlussier has joined #evergreen |
| # | 08:44:10 | rsoulliere has joined #evergreen |
| # | 08:46:51 | Melissa1 has joined #evergreen |
| # | 08:48:57 | r123 has joined #evergreen |
| # | 08:51:23 | bshum has joined #evergreen |
| # | 09:12:22 | Melissa1 has left #evergreen |
| # | 09:15:52 | Melissa2 has joined #evergreen |
| # | 09:20:47 | Dyrcona has quit IRC |
| # | 09:21:06 | Dyrcona has joined #evergreen |
| # | 09:24:48 | Melissa1 has joined #evergreen |
| # | 09:24:54 | atz | a hole in the specs? but that's unpossible! |
| # | 09:26:38 | Melissa2 has quit IRC |
| # | 09:26:41 | dbs has joined #evergreen |
| # | 09:31:18 | kmlussier_ has joined #evergreen |
| # | 09:33:02 | kmlussier has quit IRC |
| # | 09:33:16 | kmlussier_ is now known as kmlussier |
| # | 09:38:58 | dbs | @later tell agJohn Just backported the 1.6.1-2.0 upgrade fixes for custom indexes from trunk into the rel_2_0 branch - please give it a shot and report back! bshum that goes for you too :) |
| # | 09:38:58 | pinesol | dbs: The operation succeeded. |
| # | 09:43:33 | jenny has joined #evergreen |
| # | 09:52:30 | bshum | dbs: Awesome! Going to give it a whirl right after I get a reloaded DB on our test server. |
| # | 09:52:50 | dbwells | senator: From what I have read, the 1st week of a month is the first one with a Thursday. Based on that definition, it is correct to skip 12/19, not 12/26. If you want it to skip the 4th Monday, I think you should specify 1204mo in this case. |
| # | 09:54:45 | dbwells | As for counting weeks backwards, I also am not aware of what qualifies as the 'last' week. |
| # | 09:55:38 | dbs | time-- # see also timezones-- |
| # | 09:55:41 | Guest56504 has quit IRC |
| # | 09:58:32 | dbs | @later jamesrf Maybe need to do something like store everything in the database in UTC based on the circ_lib's timezone (assuming an OU setting for that) and pass the UTC around to various clients to format to their localized form? |
| # | 09:58:32 | pinesol | dbs: Error: The "Later" plugin is loaded, but there is no command named "jamesrf" in it. Try "list Later" to see the commands in the "Later" plugin. |
| # | 09:58:38 | dbs | @later tell jamesrf Maybe need to do something like store everything in the database in UTC based on the circ_lib's timezone (assuming an OU setting for that) and pass the UTC around to various clients to format to their localized form? |
| # | 09:58:38 | pinesol | dbs: The operation succeeded. |
| # | 09:59:04 | senator | dbwells: while i didn't find anything to read about mfhd that says the counted weeks must have a thursday (although the DateTime module for perl does), being about to specify 1204mo does provide a workaround, so pending a quick test that our mfhd date code does the right thing on thursday, i'll revert my commit to trunk |
| # | 09:59:18 | senator | s/being about/being able/ |
| # | 10:00:12 | dbs | Perhaps add an explicit test to MFHD::Date for that case to document our interpretation, too? |
| # | 10:03:38 | senator | hm k yeah the code the way it was does work based on thursday counting forward, and based on whatever the last day of the month being tested is when going backward (99,98,97) |
| # | 10:03:46 | ebyr has quit IRC |
| # | 10:04:16 | bshum | dbs: Ran the new 1.6.1-2.0 upgrade with our 1.6.1 that has a custom index in it and it ran through without any errors breaking it. |
| # | 10:04:17 | senator | dbs: tests, humbug! no i'm kidding. sure i'll add a test |
| # | 10:04:19 | dbs amused by "0599tu Last Tuesday in May" |
| # | 10:04:30 | dbs | bshum++ # thanks |
| # | 10:04:45 | bshum | dbs: So it's just adding 100 to the custom ID right? |
| # | 10:04:50 | bshum | Like if ours was 17, it's now 117 |
| # | 10:05:04 | dbs | bshum: yep - and adjusting all of the references to that ID |
| # | 10:05:14 | bshum | Fancy! |
| # | 10:05:16 | bshum | dbs++ |
| # | 10:05:23 | senator | dbwells: so if we document that we work by counting thursdays for going forward, that's fine by me. but would you agree we should also county by thursdays when going backward? |
| # | 10:05:37 | dbs | bshum: and then setting the sequence value to MAX(id) + 1 or 100, whichever is greater, so that new custom indexes go in the right place |
| # | 10:06:40 | dbwells | senator: in your example it skipped 11/14 when counting backwards, so isn't that already using the Thursday idea? |
| # | 10:06:49 | bshum | dbs: Ahh, neat! |
| # | 10:07:17 | dbs reads 1204mo as "4th monday in december" - not sure how thursday comes into play |
| # | 10:07:31 | senator | it's acutally using wednesday, and only because 11/30 happens to be a wednesday that year |
| # | 10:07:44 | senator | if you change the start holding to start on a wednesday |
| # | 10:07:48 | senator | then you get: |
| # | 10:08:45 | senator | hmmm |
| # | 10:08:50 | senator | yeah: |
| # | 10:09:08 | senator | 2,9,16,30 |
| # | 10:09:19 | senator | with 1198 as the chron code |
| # | 10:10:06 | senator | if we were assuming the count-by-thursdays rule, we'd expect 2,9,23,30, no? |
| # | 10:10:27 | ebyr has joined #evergreen |
| # | 10:10:35 | senator | dbs: the thursday talk comes from what dbwells said at 9:52 about what should be done in the case with no weekday specified |
| # | 10:11:20 | dbs | senator: ah - thus the hole in the spec; I was just working through your commit log example |
| # | 10:12:43 | senator | right yeah sorry about all the extra $'s, but anyway the chron codes there are 1198 and 1204, no specific day specified, so day of the week is only known by the date of the example holding (863) |
| # | 10:12:49 | dbs | are there examples in http://www.loc.gov/marc/holdings/hd853855.html including a $y for weekly patterns that don't specify the weekday? Perhaps that's an implicit requirement. |
| # | 10:12:57 | dbwells | senator: from what I can tell, the backwards week counting seems to be broken by any definition. Just to get things straight, you are proposing that 1199 will be the last week in November with a Thursday in it? |
| # | 10:14:01 | senator | dbwells: if 1201 is to mean the first week in december with a thursday in it, doesn't it make sense that 1199 would mean the last week in nov with a thursday in it? that only seems consistent. |
| # | 10:15:22 | kmlussier_ has joined #evergreen |
| # | 10:15:22 | kmlussier has quit IRC |
| # | 10:15:23 | kmlussier_ is now known as kmlussier |
| # | 10:15:29 | dbs | o coded $y weekly patterns, that is. |
| # | 10:15:47 | dbs | dbwells: where did you find the reference to "if no weekday is provided for a weekly pattern, assume thursday"? |
| # | 10:16:34 | dbs | "Questions on coding patterns or establishing new ones should be addressed to the Network Development and MARC Standards Office, Library of Congress, Washington, D.C. 20540-4402 (Email: ndmso@loc.gov)." |
| # | 10:18:25 | dbwells | dbs: like senator, I found that definition in the Perl DateTime module, which claims it is "based on the ICU definition of week of month, and correlates to the ISO8601 week of year definition" |
| # | 10:18:40 | dbs | dbwells: oh, so that has no correspondence to the MFHD specs |
| # | 10:19:19 | dbwells | dbs: right, I can't find it clearly defined anywhere in the MFHD documents. |
| # | 10:19:22 | jeff | unless the MFHD specs reference ISO8601 or the ICU definition? |
| # | 10:19:42 | jeff breaks own fingers in an attempt to avoid MFHD |
| # | 10:21:38 | senator | gmcharlt volunteered to forward the question onto someone like that last night. i'm hopefully assuming that he has some faster means than snail mail |
| # | 10:22:08 | dbwells | I am not sure if basing a pattern off weeks of the month without specific days is ever terribly sensible. Not that a publisher wouldn't try! |
| # | 10:23:21 | senator | right. it is part of the spec though, and i'm pretty sure it appears in actual records, so i just want our support for it to be at least consistent if no firm definition of "correct" can be had |
| # | 10:23:46 | dbs | dbwells: I was going to ask if it would be possible to raise a warning to say "No weekday is supplied, we're going to assume XXX until told otherwise." |
| # | 10:23:56 | senator | the caption/pattern wizard that allows relatively easy building of 85x tags ... |
| # | 10:24:01 | senator | doesn't yet handle $y |
| # | 10:24:26 | senator | i was thinking of extending it to handle $y, and that seems like a good place to me to let the user know we assume thursday, or whatever we do |
| # | 10:25:23 | dbs looks briefly at z39.71, gouges eyes out |
| # | 10:26:33 | Administrator_ has joined #evergreen |
| # | 10:26:50 | Administrator_ is now known as sfortin |
| # | 10:27:22 | senator | in other words, just to recap since understandings have evolved since this conversation started, i'm proposing reverting my commit to trunk from last night, and instead introducing a new patch that leaves forward counting (1201) alone, but changes backward counting (1198) to be thursday-based |
| # | 10:28:13 | senator | UI work in the caption/pattern wizard is a separate thing and won't go into rel_2_0 until post-release anyway, so it's only the 1201/1198 stuff that i'm talking about for the immediate term. |
| # | 10:29:05 | dbwells | senator: I actually think your change was certainly a more natural interpretation. But since we can already handle it by specifying the day-of-week, I think the Thursday-based idea is the best we have to go on. |
| # | 10:31:12 | dbwells | senator: so, in our current example case, if you chose to omit 1201, then Nov. 28 would be the actual issue omitted? (as I said, not exactly natural!) |
| # | 10:31:54 | atz | need to develop a religion w/ a thrusday sabbath... |
| # | 10:32:00 | atz | *thursday |
| # | 10:32:27 | dbwells | and if you mean Dec. 5, you would need to omit 1201mo. |
| # | 10:33:37 | senator | dbwells: for our current example (now for wednesday issues, my commit reverted) 1201 actually winds up omitting nothing |
| # | 10:33:41 | senator | because of bounds checking |
| # | 10:34:07 | senator | there's some code in there that makes sure 1201 will only apply to a date that's actually in december |
| # | 10:34:46 | senator | certainly not super intuitive, but again consistent enough to be labeled ? |
| # | 10:34:59 | dbwells | right, but shouldn't it omit 11/28, strictly speaking? Or is that crazy? |
| # | 10:35:28 | senator | maybe we should actually wait for word from marc world headquarters |
| # | 10:35:32 | dbwells | :) |
| # | 10:36:01 | dbwells hopes to not encounter such patterns in real life |
| # | 10:36:39 | rjackson-isl has joined #evergreen |
| # | 10:37:24 | ChanServ changes topic to "Welcome to the #Evergreen library system channel! | We are publicly logged. | Large pastes at http://paste.lisp.org/new/evergreen" |
| # | 10:38:08 | rjackson-isl | Good morning Evergreen community |
| # | 10:38:38 | rjackson-isl | I have an issue with DOB column in the actor.usr table not displaying correctly in 1.6.0.0 |
| # | 10:38:56 | phasebb has joined #evergreen |
| # | 10:39:04 | rjackson-isl | was instructed to set up a trigger and function to change hour to 12 to allow correct display in staff client |
| # | 10:39:36 | rjackson-isl | this doesn't work for pre-existing actor.usr entries that did not have a dob and are subsequently updated to have one |
| # | 10:40:18 | rjackson-isl | rule checks if NEW.dob != OLD.dob |
| # | 10:40:23 | rjackson-isl | any thoughts? |
| # | 10:45:12 | moodaepo | rjackson-isl: Not that I have an answer but where did you get the instruction to change the hour? |
| # | 10:46:59 | dbs | dbwells: I suspect 1201 omitting 11/28 would be an unpleasant surprise |
| # | 10:48:00 | dbs | rjackson-isl: maybe if you paste the trigger and function to paste.lisp.org so we don't have to guess? |
| # | 10:48:26 | dbwells | dbs: unfortunately, the alternative is that no issues are omitted, which is also at least a surprise. |
| # | 10:50:50 | dbs | dbwells: sure is. at the very least, if we can't surface this in the UI, I think we need to generate an explicit log message that says "weekly pattern X doesn't specify a day and you might be surprised by the results" |
| # | 10:52:37 | dbwells | senator: dbs: while a part of me still wants to support this esoteric idea, and rely on specificity to clear up the confusion, I am becoming more convinced that senator's changes will probably be the best bet (and I'll put my money there as far what the standards guys have to say as well). |
| # | 10:53:39 | rjackson-isl | dbs: her is the trigger http://pastie.org/1452732 |
| # | 10:53:48 | senator | heh, ok, well i reverted my commit for now in the spirit of leaving things the way they were until we get something more solid to go on |
| # | 10:53:57 | senator | could always recommit if that's the way we want to go |
| # | 10:54:06 | senator will ping gmcharlt about his contact later |
| # | 10:54:48 | rsoulliere | Hi, Could or should the DIG use this channel for a DIG meeting if we book ahead -- use the official calendar? |
| # | 10:55:17 | mrpeters-isl thinks maybe an #evergreen-doc channel could be useful? |
| # | 10:55:32 | Dyrcona | rjacksion-isl: null fails all = or != comparisons. |
| # | 10:55:37 | mrpeters-isl | or even #evergreen-mtg |
| # | 10:55:43 | dbs | rsoulliere: I say just go ahead and use the #evergreen channel |
| # | 10:55:55 | dbs | that way it gets logged automatically |
| # | 10:56:09 | Dyrcona | rjackson-isl: the trigger probably also needs a check for "NEW.dob is not null". |
| # | 10:56:34 | dbs | and we don't need to set up separate searches, etc just for meeting logs |
| # | 10:56:43 | Dyrcona | rjackson-isl: null != null return false. |
| # | 10:56:44 | dbs | Dyrcona++ |
| # | 10:57:20 | rjackson-isl | Dyrcona: thanks! |
| # | 10:57:34 | Dyrcona | rjackson-isl: you're welcome. |
| # | 10:58:31 | dbwells | senator: After all, if we are considering a weekly, and the world revolves around that publication, then weeks don't start until it is published |
| # | 10:59:01 | senator | yeah, that would make sense |
| # | 10:59:54 | dbwells | senator: sorry, just restating the obvious, really :) |
| # | 11:00:09 | dbs | rjackson-isl: add this line first: IF OLD.dob IS NULL THEN NEW.dob := NEW.dob + interval '12 hour'; END IF; |
| # | 11:00:20 | dbs | (right after the BEGIN) |
| # | 11:00:39 | bshum | dbs: berick: I got a preliminary list of program topics from afterl, you guys were wondering about that right? |
| # | 11:00:50 | dbs | bshum: would be great if that could be shared |
| # | 11:01:03 | berick | bshum: indeed |
| # | 11:01:23 | rjackson-isl | dbs++ |
| # | 11:02:45 | dbs | (if NEW.dob IS NULL, then NULL + '12 hours' still equals NULL, so a check on NEW.dob NULLness isn't explicitly needed) |
| # | 11:03:17 | senator | dbwells: no prob |
| # | 11:03:52 | rjackson-isl | dbs: That makes sense :-) now off to test and implement |
| # | 11:04:04 | phasefx | rjackson-isl: may also want to use date_trunc. something like NEW.dob = date_trunc('day',NEW.dob) + interval '12 hour'; |
| # | 11:04:48 | phasefx | which resets the hour to 0 before you add 12 |
| # | 11:04:49 | rjackson-isl | phasefx: Sorry for the question but what will this get me? |
| # | 11:05:01 | phasefx | maybe nothing depending on the data |
| # | 11:05:04 | dbs | phasefx: would be good for correctness, but I'm assuming the staff client passes in +00:00:00 |
| # | 11:05:18 | dbs | (or something like that) |
| # | 11:05:29 | rjackson-isl | phasefx: thanks - that will help as some dobs have hour other than 0 (about 10%) |
| # | 11:05:30 | phasefx | right, could be overkill, but just occurred to me |
| # | 11:08:47 | rsoulliere | Any guidelines on dates or times for the DIG meetings if we use the channel...? other than not conflict with other meetings taking place on this channel of course. |
| # | 11:10:06 | phasefx | rsoulliere: I say aim for the convenience of the DIGgers. If we ever get too crowded here, we can address that when it happens |
| # | 11:10:41 | rsoulliere | OK, thanks. |
| # | 11:10:56 | dbs | rsoulliere: right - if the DIG meeting doesn't conflict with another meeting on http://evergreen-ils.org/calendar then it should be okay to use this channel |
| # | 11:11:07 | ebyr is now known as eby |
| # | 11:11:08 | eby has joined #evergreen |
| # | 11:11:10 | dbs | (and the DIG meeting will go on http://evergreen-ils.org/calendar right?) :) |
| # | 11:11:17 | rsoulliere | Yes, |
| # | 11:11:49 | rsoulliere | dbs: We might even have a set schedule perhaps... |
| # | 11:12:02 | dbs | btw: DIG++ |
| # | 11:12:32 | rsoulliere | We are going to try to follow the developer meeting model -- with action items and everything. :-) |
| # | 11:17:57 | dbs reads the translation process for Bugzilla: http://www.bugzilla.org/docs/localizer.html "Once you've got a copy of the templates, start translating the English in them. It should be obvious what to translate and what not to translate." |
| # | 11:18:02 | finnapz has left #evergreen |
| # | 11:19:36 | dbs | (still trying to find a good solution for Template::Toolkit i18n... the Bugzilla model seems translator unfriendly) |
| # | 11:20:18 | eby has quit IRC |
| # | 11:21:34 | finnapz has joined #evergreen |
| # | 11:22:21 | phasefx | maybe a harmless tag in a template to designate an identifier for the text that follows? |
| # | 11:23:01 | dbs | http://mail.template-toolkit.org/pipermail/templates/2008-July/010293.html seems to be the right approach - lets us still use PO files |
| # | 11:24:18 | dbs | It's going to run into problems with stuff like "<div>You are <strong>not</strong> allowed to do <span style="foo">X</span>.</div>" though, I think. |
| # | 11:24:39 | jenny has quit IRC |
| # | 11:25:08 | dbs | eeevil is busy driving from Atlanta to San Diego |
| # | 11:25:51 | dbs | we should have ponied up for a driver + 3g mobile hotspot for him so he could code in the back seat |
| # | 11:26:03 | berick | s/from/to/ s/to/from/ |
| # | 11:26:44 | jeff | dbs: i came across this old article, but the listings all appear broken -- http://www.drdobbs.com/web-development/184415992 |
| # | 11:27:55 | dbs | berick: geez, why would he want to go back to Atlanta? :) |
| # | 11:28:58 | berick | furreal |
| # | 11:29:07 | dbs | jeff: thanks! |
| # | 11:30:46 | jeff | alas, a quick glance indicates that the techniques didn't make it into their book. |
| # | 11:31:09 | dbs | the i18n as a roll-your-own afterthought to TT2 makes me long for something more like genshi (http://genshi.edgewall.org/) |
| # | 11:32:02 | jeff | well, berick HAS been doing all that work on the python opensrf bits... ;-) |
| # | 11:35:50 | phasefx | wow, I like that (genshi examples, tag/id when you need to, handles nested markup cleanly) |
| # | 11:36:41 | phasefx | <p i18n:msg="name"><i>Please</i> visit <b>the site <a href="${site.url}">${site.name}</a></b>for help.</p> => [1:Please] visit [2:the site [3:%(name)s]] for help. |
| # | 11:37:50 | phasefx | mmm.. i18n:choose, i18n:singular, i18n:plural |
| # | 11:41:34 | dbs peeks at jinja w/ http://jinja.pocoo.org/extensions/#i18n-extension for an option that doesn't require strict XML templates |
| # | 11:43:36 | mrpeters-isl | have a server reporting a Internal Server Error when attempting to use vandelay...likely a permissions issue, but not sure where to look for the vandelay binaries? perhaps a different name on the binaries than "vandelay"? |
| # | 11:44:52 | dbs | vandelay binaries? |
| # | 11:45:03 | phasefx | mrpeters-isl: apache error logs will likely show you the files (and more info on the error) |
| # | 11:45:17 | mrpeters-isl | well, maybe not...but i'm betting a permissions issue on something relating to vandelay |
| # | 11:45:23 | dbs | mrpeters-isl: like /openils/var/web/templates/default/vandelay/ ? |
| # | 11:46:24 | mrpeters-isl | phasefx: right |
| # | 11:46:52 | jenny has joined #evergreen |
| # | 11:49:05 | mrpeters-isl | what makes it really fun is there appears to be no apache logs for this one server...how lovely |
| # | 11:51:19 | mrpeters-isl | syslog-ng.conf IP fail :( |
| # | 11:51:19 | brian_f has joined #evergreen |
| # | 11:53:29 | mrpeters-isl | thoughts on these? |
| # | 11:53:30 | mrpeters-isl | ap_error.log:2011-01-12 11:51:07 192.168.80.21 osrf_json_gw: [error] [client 192.168.80.230] File does not exist: /openils/var/web/xul/rel_1_6_0_0/server/main/OpenILS, referer: http://192.168.80.21/xul/rel_1_6_0_0/server/ |
| # | 11:53:30 | mrpeters-isl | ap_error.log:2011-01-12 11:51:07 192.168.80.21 osrf_http_translator: [error] [client 192.168.80.230] File does not exist: /openils/var/web/xul/rel_1_6_0_0/server/main/util, referer: http://192.168.80.21/xul/rel_1_6_0_0/server/ |
| # | 11:53:30 | mrpeters-isl | ap_error.log:2011-01-12 11:51:07 192.168.80.21 osrf_json_gw: [error] [client 192.168.80.230] File does not exist: /openils/var/web/xul/rel_1_6_0_0/server/main/util, referer: http://192.168.80.21/xul/rel_1_6_0_0/server/ |
| # | 11:53:30 | mrpeters-isl | ap_error.log:2011-01-12 11:51:07 192.168.80.21 osrf_json_gw: [error] [client 192.168.80.230] File does not exist: /openils/var/web/xul/rel_1_6_0_0/server/main/util, referer: http://192.168.80.21/xul/rel_1_6_0_0/server/ |
| # | 11:56:04 | mrpeters-isl | other than vandelay.css I don't see anything related |
| # | 11:56:10 | phasefx | maybe normal 404's coming from JSAN initialization? |
| # | 11:59:36 | b_bonner has joined #evergreen |
| # | 12:11:30 | youdonotexist has joined #evergreen |
| # | 12:18:37 | dbs muses that when IE 10 is produced, EG will suddenly start applying nokeepalive ssl-unclean-shutdown downgrade-1.0 force-response-1.0 applied to IE 10 clients |
| # | 12:22:23 | gmcharlt | senator: dbs: my contact is somebody who's been active in serials and MARBI for years; I'll let you know if I hear from her |
| # | 12:26:03 | jamesrf has joined #evergreen |
| # | 12:26:34 | alxp has quit IRC |
| # | 12:27:15 | senator | gmcharlt: neat, thanks |
| # | 12:28:22 | atz | dbs: assuming they don't call it IE X |
| # | 12:28:57 | senator giggles at that |
| # | 12:29:02 | dbs | atz: good point! hopefully they'll call it 9+1 |
| # | 12:29:36 | dbs | gmcharlt: I'm suspicious; are you sure your contact is not insane? |
| # | 12:30:39 | gmcharlt | dbs: heh. on the other hand, would you seriously argue that sanity is a required attribute for anybody who has anything to do with library automation? |
| # | 12:30:58 | dbs | gmcharlt: touché |
| # | 12:31:02 | gmcharlt | (and why are the walls in this IRC channel so plush?) |
| # | 12:46:00 | moodaepo_ has joined #evergreen |
| # | 12:47:41 | rjackson-isl has quit IRC |
| # | 12:47:54 | moodaepo_ has quit IRC |
| # | 12:58:02 | dbs | hrm. reingest-1.6-2.0.sql is unhappy with XPath on our test server - http://paste.lisp.org/display/118573 |
| # | 13:00:44 | gmcharlt | dbs: check to make sure that the namespace prefixes in the xpath match the config.xml_transform values |
| # | 13:01:08 | dbs | gmcharlt: right-o, will do. thanks |
| # | 13:01:09 | gmcharlt | unlike 1.6, 2.0 requires that the ns prefixes actually be consistent |
| # | 13:03:13 | dbs | ahh. name = marcxml, prefix = marc is probably the problem |
| # | 13:05:29 | dbs is tempted to throw in a regex for s/\/marcxml:/\/marc:/g into the custom index fixer-upper for config.metabib_field.xpath |
| # | 13:05:36 | dbs | many thanks, gmcharlt |
| # | 13:12:45 | dbs | annotated http://paste.lisp.org/display/118573 with my solution |
| # | 13:24:59 | artunit has quit IRC |
| # | 13:49:47 | jenny1 has joined #evergreen |
| # | 13:50:01 | jenny1 has left #evergreen |
| # | 13:51:50 | jenny has quit IRC |
| # | 13:59:33 | rsoulliere | Hi, We installed a new 3M SmartChute but are getting a SIP related error when trying to checkin an item. |
| # | 13:59:46 | rsoulliere | Item config script [ARRAY(0x2650f60) : circ/circ_item_config.js] failed to run: Error: ReferenceError: js2JSON is not defined at line 0: (null) |
| # | 14:00:14 | rsoulliere | I've seen this error before in the logs of this list. Does this ring any bells? |
| # | 14:01:37 | phasefx | coming from a logging statement maybe? |
| # | 14:02:57 | rsoulliere | the checkin is also not completed.. |
| # | 14:03:39 | jeff | oh hey, i think i just solved our auto-lost and forgive on checkin BUT DON'T BREAK mmpbbt and DON'T OWE A REFUND |
| # | 14:06:46 | finnapz1 has joined #evergreen |
| # | 14:07:10 | finnapz1 has left #evergreen |
| # | 14:07:25 | finnapz has quit IRC |
| # | 14:08:06 | finnapz has joined #evergreen |
| # | 14:31:43 | Melissa1 has quit IRC |
| # | 14:37:50 | Melissa1 has joined #evergreen |
| # | 14:42:02 | AaronZ has joined #evergreen |
| # | 14:43:26 | jamesrf has quit IRC |
| # | 14:44:46 | mrpeters-isl | rsoulliere: http://www.mail-archive.com/open-ils-general@list.georgialibraries.org/msg00380.html perhaps? |
| # | 14:45:33 | mrpeters-isl | seems to be identical, except for the array portion |
| # | 14:48:14 | AaronZ | Reports question: Is there a way to create a report listing all Title Records for which all Item and Volume Records have been deleted? I can get all Title Records which do not have any Item or Volume Records attatched, but it seems that deleted items are still attatched to the Title Record |
| # | 14:49:47 | mrpeters-isl | AaronZ: by title record do you mean a reporter.materialized_simple_record row? |
| # | 14:50:14 | mrpeters-isl | aka Simple Record Extract > Title (Proper) IIRC... |
| # | 14:51:22 | mrpeters-isl | or by "title" are you referring to a biblio.record_entry row |
| # | 14:51:44 | mrpeters-isl | aka bib record |
| # | 14:52:30 | AaronZ | mrpeters-isl:I am referring to the bib record that the volume and copy records are attatched to |
| # | 14:52:48 | mrpeters-isl | ok. so you want to find all biblio.record_entry rows that don't have an associated call number |
| # | 14:53:05 | AaronZ | Yes |
| # | 14:53:25 | mrpeters-isl | well, if you have access to the SQL database you could do this quicker |
| # | 14:54:17 | mrpeters-isl | you could do something like SELECT * FROM asset.call_number WHERE record NOT IN (SELECT id FROM biblio.record_entry); and dump it to csv |
| # | 14:54:34 | mrpeters-isl | rather - SELECT * FROM asset.call_number WHERE record NOT IN (SELECT id FROM biblio.record_entry WHERE deleted='t'); |
| # | 14:54:44 | mrpeters-isl | that'd get you all of the call number's linked to deleted bibs |
| # | 14:54:54 | AaronZ | I have access to the SQL database, but for repeatibility, I would prefer to setup a report that can be scheduled to run monthly |
| # | 14:55:09 | mrpeters-isl | you could then take it a bit further and get asset.copy rows |
| # | 14:55:32 | mrpeters-isl | ok. let me open up the reports module and think about this a second. i've just gotten better at using SQL to get the info i need than reporter :) |
| # | 14:55:42 | mrpeters-isl | the theory, i think is the same though |
| # | 14:56:33 | AaronZ | This is to do a batch delete for OCLC. We want to have them remove the titles that we no longer have holdings for from their listing for our system |
| # | 14:57:00 | mrpeters-isl | sure |
| # | 14:57:29 | mrpeters-isl | so, we want to use the Item source, id say |
| # | 14:57:54 | mrpeters-isl | maybe use barcode, call number label, owning library, bib id as displayed fields |
| # | 14:58:03 | mrpeters-isl | maybe even copy id |
| # | 14:58:36 | mrpeters-isl | and then a filter on Item > Bib Record > Call Number/Volume > Is Deleted? |
| # | 14:59:18 | mrpeters-isl | and probably an Is Deleted filter on Item = false |
| # | 14:59:22 | eby has joined #evergreen |
| # | 15:03:03 | mrpeters-isl | i just ran that on my system and got seemingly good results |
| # | 15:03:19 | mrpeters-isl wonders if there is a way to export this template via SQL to send to you |
| # | 15:06:14 | mrpeters-isl | AaronZ: any luck? |
| # | 15:06:22 | AaronZ | Is that returning the item records (ie: a list of items that have been deleted), or the bib records (ie: a list of bib records which no longer have any items attatched to them)? |
| # | 15:07:05 | AaronZ | I got: ERROR: table name "19a58c77dfa7418c489d7591bce37fc6" specified more than once at /openils/bin/clark-kent.pl line 219. |
| # | 15:07:14 | mrpeters-isl | AaronZ: it lists Copy ID (asset.copy.id), barcode (asset.copy.barcode), call number label (asset.call_number.label), and Owning Lib (asset.copy.owning_lib) |
| # | 15:07:50 | mrpeters-isl | IIRC, tables get aliased with an MD5 string (i think?) so perhaps you have the same filter in there twice or something? |
| # | 15:08:14 | AaronZ | Looks like I specified the call number twice |
| # | 15:08:19 | mrpeters-isl | there you go! |
| # | 15:08:29 | mrpeters-isl | also, if you wanted to have the bib id's, just add that as a displayed field |
| # | 15:08:44 | mrpeters-isl | Item > Call Number > Bib Record > Record ID |
| # | 15:08:59 | mrpeters-isl | but barcode and copy id are probably most useful for OCLC to remove your holdings |
| # | 15:10:02 | AaronZ | I dont really need the copy info (call number or barcode), I just need the bib info (record ID, |
| # | 15:10:12 | mrpeters-isl | ok so just add/remove what you do and don't need |
| # | 15:10:14 | AaronZ | and TCN |
| # | 15:10:16 | mrpeters-isl | and you should be on your way |
| # | 15:11:12 | mrpeters-isl | TCN should be in Item > Call Number > Bib Record > Simple Record Extracts > TCN Value |
| # | 15:12:36 | AaronZ | ok |
| # | 15:12:40 | AaronZ | I will try that |
| # | 15:13:16 | mrpeters-isl | good luck |
| # | 15:21:44 | AaronZ | mrpeters-isl: It appears to be returning all copy level records that have had a copy deleted. Is there a way to modify this to only return title level (aka bib) records where ALL of the copies have been deleted? |
| # | 15:23:48 | mrpeters-isl | hmm |
| # | 15:24:22 | mrpeters-isl | you probably don't need a copy level filter |
| # | 15:24:36 | mrpeters-isl | just a Bib Record - Is Deleted = True |
| # | 15:25:10 | bshum | Assuming that their system is configured to auto-delete bibs with no active volumes/items, right? But that's default behavior last I checked. |
| # | 15:25:26 | mrpeters-isl | not sure on that bshum |
| # | 15:25:33 | AaronZ | As I understand it, we only have bib level holdings listed in OCLC/Worldcat (ie: that the Pioneer Library System owns a copy of book x, not that Library y in the Pioneer Library System owns book x), so we ONLY want to remove the bibs where none of our libraries own a copy anymore |
| # | 15:26:04 | bshum | mrpeters-isl: Well, I was just saying that because our system didn't automatically delete bibs at first, so our targets against bibs with deleted = true yielded no results for awhile. |
| # | 15:26:14 | bshum | One hopes they don't suffer our fate. |
| # | 15:26:21 | mrpeters-isl | well, but if a bib is deleted then in reality the items may as well be too...because they're not going to show up except by a direct barcode search in the staff client...IIRC |
| # | 15:26:29 | dbs | Sounds like a COUNT() on copy where copy IS NOT deleted, if you can manage that |
| # | 15:26:52 | dbs | (and the result of that count needs to be 0) |
| # | 15:27:08 | mrpeters-isl | good call dbs\ |
| # | 15:27:13 | AaronZ | dbs: hmmmm that might work with one of my trial reports |
| # | 15:27:29 | dbs | mrpeters-isl: we have lots of bibs with no copies that show up in search - ye olde 856 $9 |
| # | 15:27:40 | mrpeters-isl | well, they have trancendence set, righT? |
| # | 15:27:46 | dbs | mrpeters-isl: no |
| # | 15:27:51 | mrpeters-isl | hmm |
| # | 15:28:11 | dbs | $9 = shortname of library, avoids transcendence and provides scoping to that library scope |
| # | 15:28:27 | mrpeters-isl | interesting. i didn't know of that feature! |
| # | 15:28:51 | mrpeters-isl | i just always set trancendence for bibs that are never going to have holdings, but that we want to appear for everyone |
| # | 15:30:10 | dbs | http://docs.evergreen-ils.org/1.6/draft/html/cataloguing.html doesn't document it, but should |
| # | 15:31:09 | dbs | although it's mentioned in the release notes for 1.6.0.0 at http://docs.evergreen-ils.org/1.6/draft/html/releasenotes.html |
| # | 15:32:03 | mrpeters-isl | well cool |
| # | 15:34:18 | gmcharlt prepares to drop CurrencyTextBox from AutoFieldWidget in favor of NumberTextBox |
| # | 15:34:40 | gmcharlt | CurrencyTextBox in Dojo 1.3 is just too broken if you need to enter negative amounts |
| # | 15:34:49 | gmcharlt | thoughts/objections? |
| # | 15:35:43 | dbs | I say do it, and open a bug to switch it back once we can move to Dojo 1.5 or 1.6 or whatever |
| # | 15:36:52 | dbs | thoughts / objection to my marking email_notify "false" in opensrf.xml.example for 2.0 for predues and holds? And to mark that whole section DEPRECATED in favour of action-trigger-based notification? |
| # | 15:40:17 | bshum | dbs: By that whole section, you mean the email portion right? |
| # | 15:40:46 | dbs | bshum: notifications/overdue and notifications/predue |
| # | 15:47:22 | bshum | dbs: If overdues / predues are all action-trigger-based in 2.0, then that's pretty cool. I'm not the most familiar with that, I think we still use the print portions of that file in our 1.6.1 |
| # | 15:51:27 | dbs | bshum: email notifications can be all action-trigger based, but I can imagine the print use case living on in 2.0 |
| # | 15:52:21 | gmcharlt | dbs: +1 |
| # | 15:52:50 | bshum | dbs: Oh I see how it's working now. (Had to examine the file) |
| # | 15:52:55 | bshum | +1, let it burn |
| # | 15:53:41 | dbs | went through the code-side of the 1.6.1->2.0 upgrade for our test instance with a copy of our production data today, so some of this is still nice and fresh :) |
| # | 15:57:16 | bshum | On a side note, I created a page http://evergreen-ils.org/dokuwiki/doku.php?id=eg11:proposals |
| # | 15:57:24 | bshum | It's temporary till they add it to the conference site |
| # | 15:57:30 | bshum | But has the pertinent info on there. |
| # | 15:59:57 | dbs | bshum++ |
| # | 16:00:36 | bshum | So, prepare and submit stuff! |
| # | 16:02:20 | dbs | aha! Net::SSH2 is missing on our debian lenny system because it's not packaged on lenny |
| # | 16:03:43 | rsoulliere | mrpeters++ thanks for the lead re: my SIP error |
| # | 16:04:38 | brian_f has quit IRC |
| # | 16:05:19 | rsoulliere | fixed by adding <path>/openils/lib/javascript/</path> to oils_sip.xml |
| # | 16:06:07 | rsoulliere | which is where the the JSON_v1.js script is located. |
| # | 16:07:14 | rsoulliere | Should I follow up with any bug reporting /patch submitting in launchpad? |
| # | 16:08:02 | rsoulliere | One other funny thing is that the 3M gui seems to be reporting a type of error at the SmartChute terminal |
| # | 16:08:23 | dbs | rsoulliere: please do |
| # | 16:09:24 | rsoulliere | However, Evergreen is checking in the item and changing the status. |
| # | 16:10:25 | rsoulliere | I will also follow up with the 3M guy regarding this behaviour. |
| # | 16:10:42 | rsoulliere | thanks all. regards. |
| # | 16:10:47 | mrpeters-isl | yay! |
| # | 16:10:50 | rsoulliere has quit IRC |
| # | 16:11:07 | b_bonner has quit IRC |
| # | 16:12:39 | b_bonner has joined #evergreen |
| # | 16:27:40 | jenny1 has joined #evergreen |
| # | 16:28:45 | bshum has quit IRC |
| # | 16:42:14 | bshum has joined #evergreen |
| # | 16:51:40 | kmlussier has quit IRC |
| # | 17:03:40 | Melissa1 has quit IRC |
| # | 17:07:28 | mjg_ has joined #evergreen |
| # | 17:09:50 | mjg_ has left #evergreen |
| # | 17:20:59 | Dyrcona has quit IRC |
| # | 17:22:33 | AaronZ has quit IRC |
| # | 17:40:18 | sfortin has quit IRC |
| # | 17:42:47 | b_bonner has quit IRC |
| # | 17:53:26 | b_bonner has joined #evergreen |
| # | 17:56:41 | b_bonner has quit IRC |
| # | 18:03:07 | atheos_ has quit IRC |
| # | 18:04:51 | eby has quit IRC |
| # | 18:05:29 | phasebb has quit IRC |
| # | 18:06:14 | eby has joined #evergreen |
| # | 18:06:53 | jenny1 has left #evergreen |
| # | 18:07:35 | atheos_ has joined #evergreen |
| # | 18:35:00 | b_bonner has joined #evergreen |
| # | 19:29:44 | b_bonner has quit IRC |
| # | 19:31:46 | youdonotexist has quit IRC |
| # | 20:03:34 | corretico has joined #evergreen |
| # | 20:40:30 | atheos_ has quit IRC |
| # | 21:13:48 | youdonotexist has joined #evergreen |
| # | 22:04:00 | bshum has quit IRC |
| # | 22:59:43 | artunit has joined #evergreen |
| # | 23:36:37 | dbs has quit IRC |
April 10-13