2010-03-16T00:35:11 *** dbs has quit IRC 2010-03-16T01:02:03 *** mck9 has left #evergreen 2010-03-16T02:37:01 *** jeff_ has quit IRC 2010-03-16T02:38:29 *** jeff_ has joined #evergreen 2010-03-16T04:08:33 *** natschil has joined #evergreen 2010-03-16T07:54:51 *** mck9 has joined #evergreen 2010-03-16T08:17:39 *** natschil has quit IRC 2010-03-16T08:42:31 *** emrikol has joined #evergreen 2010-03-16T09:05:37 *** bshum has joined #evergreen 2010-03-16T09:10:36 *** dbs has joined #evergreen 2010-03-16T09:11:52 *** sfortin has joined #evergreen 2010-03-16T09:14:27 *** Meliss has joined #evergreen 2010-03-16T09:14:54 *** tsbere has joined #evergreen 2010-03-16T09:23:34 *** sfortin_ has joined #evergreen 2010-03-16T09:26:09 *** sfortin has quit IRC 2010-03-16T09:26:21 *** Dmagick has quit IRC 2010-03-16T09:26:34 *** phasefx has quit IRC 2010-03-16T09:27:08 *** leed has quit IRC 2010-03-16T09:27:30 *** jenny has joined #evergreen 2010-03-16T09:28:57 *** sfortin_ has quit IRC 2010-03-16T09:30:01 *** dmcmorris_esi has quit IRC 2010-03-16T09:30:22 *** Dmagick-mac has quit IRC 2010-03-16T09:30:24 *** sfortin has joined #evergreen 2010-03-16T09:31:24 *** Dmagick has joined #evergreen 2010-03-16T09:31:27 *** leed has joined #evergreen 2010-03-16T09:31:58 *** Dmagick-mac has joined #evergreen 2010-03-16T09:32:06 *** dmcmorris_esi has joined #evergreen 2010-03-16T09:33:31 *** phasefx has joined #evergreen 2010-03-16T09:39:24 *** sfortin has quit IRC 2010-03-16T10:02:17 dbs: re the __make_marc_doc patch, do you want another version that clears out empty subfield nodes? 2010-03-16T10:02:40 sure, if you want to do it; otherwise I'll just add it myself after committing your patch 2010-03-16T10:04:25 dbs: ok, I'll send a replacement patch in a few 2010-03-16T10:04:55 and then datafields? 2010-03-16T10:05:00 yep 2010-03-16T10:05:05 gmcharlt++ 2010-03-16T10:05:10 and a DCO! 2010-03-16T10:05:44 dbs: speaking of DCOs, is a blanket DCO sufficient? 2010-03-16T10:06:09 gmcharlt: so far we haven't gone that route, but it would make sense 2010-03-16T10:06:36 dojo has a similar approach 2010-03-16T10:06:43 many projects do 2010-03-16T10:07:13 a page on the wiki for such DCOs should suffice; people can gpg-sign them if they feel the need 2010-03-16T10:07:22 but i think there might still be a required statement on submissions... 2010-03-16T10:07:31 (in the dojo example) 2010-03-16T10:08:02 ah, no. just a requirement that your full name be associated with the patch so that your CLA can be cross-referenced 2010-03-16T10:09:04 jeff: that's no problem - a Signed-of-by: bit should do it 2010-03-16T10:12:07 typo of the day: 2010-03-16T10:12:13 sue OpenILS::Foo; 2010-03-16T10:12:20 *** r123 has joined #evergreen 2010-03-16T10:17:07 If only we had a foundation associated with a free software conservancy that could offer us actual legal advice! 2010-03-16T10:18:19 So, evergreen is 100% incompatible with running on RHEL4. Even if you custom compile half the libraries in the system as newer versions. 2010-03-16T10:19:34 *** Flubber has joined #evergreen 2010-03-16T10:20:59 *** Flubber has quit IRC 2010-03-16T10:22:33 *** alxp has joined #evergreen 2010-03-16T10:23:51 dbs: is there an issue? 2010-03-16T10:28:25 *** bshum has quit IRC 2010-03-16T10:28:31 gmcharlt: i think dbs was just responding to your "should do it" 2010-03-16T10:29:02 dbs/gmcharlt: doh, sorry, I already committed the patch... 2010-03-16T10:29:51 in the case of associating a name with a patch, I'm not sure I see how complicated it can be 2010-03-16T10:30:01 miker_: no issue, I'll take that into account when I resubmit 2010-03-16T10:30:09 hadn't looked here first :) 2010-03-16T10:30:10 gmcharlt: I meant more about whether the wiki would or could suffice 2010-03-16T10:30:15 CHARGE 2010-03-16T10:31:01 dbs: thanks for clarifying 2010-03-16T10:31:13 sorry, on a call 2010-03-16T10:31:35 *** bshum has joined #evergreen 2010-03-16T10:49:46 dbs: miker_: enhanced killer of emptiness submitted 2010-03-16T10:57:49 on it 2010-03-16T11:00:25 done 2010-03-16T11:00:36 miker_: thanks! 2010-03-16T11:02:55 <_bott_> This is odd... Can anyone verify that for hyphenated names, the barcode is not displayed in circ_summary.xul (i.e. Show item details). I see the patron name, but no barcode in the colored bar for "Current Circulation" 2010-03-16T11:04:03 <_bott_> oops, make that circ_brief.xul 2010-03-16T11:04:19 that doesn't seem right 2010-03-16T11:04:32 _bott_: 1.4? 2010-03-16T11:04:35 miker_: I see a need for a Open-ILS/src/perlmods/OpenILS/Application/Cat/MarcCommon.pm so that auth and bib MARC record handling doesn't diverge; is that namespace OK or do you have a preference for a different location for MARC utility routines 2010-03-16T11:04:53 <_bott_> And 1.6, in my case 2010-03-16T11:05:55 gmcharlt: if it will be useful outside the open-ils.cat app, perhaps OpenILS::Utils::MARC or similar? 2010-03-16T11:06:23 miker_: OpenILS::Utils::MARC it is, then 2010-03-16T11:07:31 <_bott_> It gets stranger... Hypenated last name, and no middle name/initial 2010-03-16T11:07:56 _bott_: NULL or empty? or doesn't matter? 2010-03-16T11:08:40 <_bott_> Empty, in the tests I've just run 2010-03-16T11:08:59 _bott_: you're not geting an error popup, are you? 2010-03-16T11:10:27 <_bott_> Nope, no popup error 2010-03-16T11:12:44 <_bott_> Argh... I'm trying too hard. The hypen doesn't matter, it's just the blank middle name 2010-03-16T11:15:14 _bott_: your version of circ_brief is using patron.util.format_name ? 2010-03-16T11:16:04 * phasefx really needs to let that use a localizable property string 2010-03-16T11:17:00 <_bott_> I've assumed it's line 76, using: var r_au = preq.getResultObject(); 2010-03-16T11:17:00 <_bott_> $('patron_name').value = r_au.family_name() + ', ' + r_au.first_given_name() + ' ' + (r_au.second_given_name() ? r_au.second_given_name() + ' : ' + r_au.card().barcode() : ''); 2010-03-16T11:17:43 in later versions something like that has been wrapped into a utility function 2010-03-16T11:18:25 <_bott_> \me looks into the trunk 2010-03-16T11:18:35 * phasefx can see how that line would go awry 2010-03-16T11:18:38 * _bott_ grumbles about slashes 2010-03-16T11:19:05 <_bott_> yep, makes sense after I realized what was causing it 2010-03-16T11:21:38 as dbs would say, there's nothing wrong with using linefeeds ;) would have made that error more obvious 2010-03-16T11:38:18 yeah, misplaced closing ) 2010-03-16T11:39:03 phasefx: I'll move the paren to before + ':' and round-port, eh? 2010-03-16T11:39:45 <_bott_> I ended up with: $('patron_name').value = r_au.family_name() + ', ' + r_au.first_given_name() + ' ' + (r_au.second_given_name() ? r_au.second_given_name() :'') + ' : ' + r_au.card().barcode(); 2010-03-16T11:40:12 <_bott_> Want me to send the diff? I've got it against 1.6 2010-03-16T11:40:13 _bott_: yep ... just got to that too 2010-03-16T11:40:29 I'm in rel_1_6_0 right now, so no need 2010-03-16T11:40:33 thanks, though! 2010-03-16T11:40:51 <_bott_> no problem. sorry for thinking out loud re: the hyphen 2010-03-16T11:41:20 np ... we all got there :) 2010-03-16T11:51:01 *** jamesrf has joined #evergreen 2010-03-16T11:52:59 *** Flubber has joined #evergreen 2010-03-16T11:53:28 Does the survey feature not work in 1.6.0.3? 2010-03-16T11:59:01 miker_: _bott_: thanks guys 2010-03-16T11:59:28 Flubber: it's a bit borked with multi-question surveys 2010-03-16T11:59:51 a 1.6.0 patch went it in yesterday, I still need to forward-port it 2010-03-16T12:09:41 whats the best evergreen to install if you want everyth 2010-03-16T12:09:51 ing to function properly? 2010-03-16T12:10:09 1.6.0.3+patches 2010-03-16T12:10:39 any software has bugs, so "everything" is a bit of a stretch goal to be honest 2010-03-16T12:11:42 I would say you would want to run the 3.0 series. You will need a time machine. :P 2010-03-16T12:12:16 tsbere: why settle for 3.0, then? go with 4.0 ;) 2010-03-16T12:12:20 heh, apparently our friends in NL are running 6.1.1 2010-03-16T12:12:28 (or better yet, 4.0.1) 2010-03-16T12:12:29 what version of evergreen does the georgia public library use? 2010-03-16T12:12:48 gmcharlt, if you go too far into the future you start running into hardware compatibility issues. ;) 2010-03-16T12:13:17 yeah, I hear the programming on quantum computers can be an uncertain exercise 2010-03-16T12:25:18 Flubber: GPLS/PINES is using 1.4.0.7 2010-03-16T12:26:06 phasefx: if you like, I'll forward-port the survey patch 2010-03-16T12:26:23 Flubber: note that 1.4.0.7 is not without its problems, and the focal point for bug fixing / stability is effectively 1.6.0 right now 2010-03-16T12:27:09 Flubber: if you just want to test, svn://svn.open-ils.org/ILS/branches/rel_1_6_0 would be your best bet, IMO 2010-03-16T12:28:20 *** emrikol has quit IRC 2010-03-16T12:28:31 *** StephenGWills has joined #evergreen 2010-03-16T12:29:14 miker_: go for it. I tried svn merge right away, hit conflicts, and punted for later 2010-03-16T12:29:34 *** tsbere has quit IRC 2010-03-16T12:29:34 any quesses which step of the 1.6.0.3 install i missed if at the end of it, Autogen tells me that I have an empty filename in fieldmapper.pm? 2010-03-16T12:30:01 StephenGWills: you may be using the opensrf_core.xml and opensrf.xml files from opensrf and not from evergreen. 2010-03-16T12:30:03 phasefx: just space/tab stuff 2010-03-16T12:30:44 ok will check 2010-03-16T12:31:01 How do you umark for overlay? i am going to stick with 1.6 2010-03-16T12:31:22 miker_: also, I need to sanity check the new patron editor and see how it handles surveys 2010-03-16T12:31:36 not sure what EG release the new editor gets used 2010-03-16T12:31:40 if any 2010-03-16T12:31:58 s/what/with which/ 2010-03-16T12:32:08 phasefx: trunk only, IIRC, but berick will know for sure 2010-03-16T12:32:32 I think some folks may have edited their constants.js to get at it early, as well 2010-03-16T12:35:51 *** Flubber has quit IRC 2010-03-16T12:50:19 jeff: with the exception of the substitution keys and the passwords, the Open-ILS/example xml's match the /openils/conf ones... i think i got the right ones. 2010-03-16T12:54:30 StephenGWills: sorry for the mis-steer, then! do you have two tags in /openils/conf/opensrf.xml both pointing to /openils/conf/fm_IDL.xml? 2010-03-16T12:56:21 no prob, every day is a school day for me! :) yes to the IDL tags and the file: fm_IDL.xml exists in conf 2010-03-16T13:07:50 StephenGWills: what's the exact error from autogen.sh? can you pastebin the full output? 2010-03-16T13:21:56 will set it up. sorry got called away 2010-03-16T13:26:02 not sure that worked... never done it before 2010-03-16T13:28:24 StephenGWills: the pastebot isn't working, but we can track it down 2010-03-16T13:28:32 oh 2010-03-16T13:29:04 http://paste.lisp.org/display/96478 2010-03-16T13:29:21 it really isn't that huge an output. stops right away 2010-03-16T13:31:50 is opensrf running? 2010-03-16T13:31:58 opensrf.settings in particular? 2010-03-16T13:35:17 looks like it but there are boat load of things running, like a couple of pages worth. will shut them down and restart osrf_ctl and try again 2010-03-16T13:40:54 checking out pre-cats in rel_1_6_0 is borked :-/ 2010-03-16T13:41:17 first we get a "no_item" event 2010-03-16T13:41:24 override, and we get the pre-cat prompt 2010-03-16T13:42:00 go farther, we get an error: Can't call method \"duration_rule\" on an undefined value at /openils/lib/perl5/OpenILS/Application/Circ/Circulate.pm line 1374 2010-03-16T13:45:45 works correctly if we use scripts instead of in-db 2010-03-16T13:46:15 yay scripts 2010-03-16T13:46:58 I'll ticket it if miker_ or senator doesn't swoop in with a fix, otherwise moving on. tangents, chaos powers, har :) 2010-03-16T13:47:42 *** sfortin has joined #evergreen 2010-03-16T13:54:19 <--- dummy. works fine after a reboot. 2010-03-16T13:55:53 * senator defers. not sure what else i'd break, not knowing in-db circ that well. 2010-03-16T13:57:18 action.item_user_circ_test() probably needs to return some row out of conf.circ_matrix_matchpoint by default, instead of null, but... yeah. 2010-03-16T13:57:36 unsure of the side effects 2010-03-16T13:58:11 * dbs wishes again for a gift from the testing gods of several data sets, a test harness, and a ton of unit tests 2010-03-16T13:58:51 dbs: us mere mortals will have to do, in time :) 2010-03-16T14:02:03 gmcharlt: still, a human sacrafice or two couldn't hurt. ;) 2010-03-16T14:02:15 pick me! pick me! 2010-03-16T14:02:17 oh, wait 2010-03-16T14:02:34 * gmcharlt thinks he would prefer to have cake instead 2010-03-16T14:11:48 Though I am hesitant to bring it up again, I guess there is no time like the present. I think we need to come to a consensus of some kind on tabs vs spaces :) 2010-03-16T14:12:10 I was looking through recent commits, and found a bunch like this, converting spaces to tabs: http://svn.open-ils.org/trac/ILS/changeset/15725/trunk/Open-ILS/src 2010-03-16T14:12:42 dbwells: we have consensus for the most part 2010-03-16T14:12:51 Yet phasefx's vimrc in trunk http://svn.open-ils.org/trac/ILS/browser/trunk/Open-ILS/examples/vimrc has expandtab set 2010-03-16T14:13:06 in C code, it's basically scott. and tabs are traditional there 2010-03-16T14:13:15 this is contrary, isn't it? 2010-03-16T14:13:24 elsewhere, with limited exceptions, spaces is our convention 2010-03-16T14:13:49 yes, the commit you linked is C code 2010-03-16T14:14:23 atz: thanks, that is very good to know. I work 90% in Perl, and was starting to worry about converting all my spaces back to tabs. 2010-03-16T14:14:47 may i ask how many spaces to the tab? 2010-03-16T14:14:53 4 2010-03-16T14:14:59 k 2010-03-16T14:15:35 i'm an old C guys and tend to tab my perl and php 2010-03-16T14:15:42 will be good :) 2010-03-16T14:16:10 It would probably be wise to add a noet exception to the vimrc for .c files 2010-03-16T14:16:24 similar to the makefiles and txt files. 2010-03-16T14:16:30 good idea 2010-03-16T14:18:10 i once wasted an afternoon on a non-tab space in a makefile. grrr.... 2010-03-16T14:18:20 mck9 has his own conventions 2010-03-16T14:19:13 yeah, i don't feel the need to force my perl/css/html/js-centric formatting on our very small population of C coders 2010-03-16T14:19:35 mostly because i don't want to read that stuff anyway! 2010-03-16T14:19:38 hah 2010-03-16T14:20:03 atz: Is there also a consensus for xml files and sql files? 2010-03-16T14:20:37 dbwells: miker_ uses lots of tabs in SQL files, but I hatesess the tabs because a copy & paste into psql results in tab-completion (argh!) 2010-03-16T14:20:41 * gmcharlt prefers spaces instead of tabs for everything except C code 2010-03-16T14:21:27 dbwells: in general, i think so. probably a lot of aberrant formatting in the codebase though. 2010-03-16T14:21:30 * dbs prefers spaces instead of tabs for everything. he even has spaces on his file folders. 2010-03-16T14:21:48 dbs: :D 2010-03-16T14:22:10 If there is a need to change the C from tabs to spaces that can be done easily with expand -t4. In my experience unexpand doesn't work so well, so I have been reluctant to switch to spaces because it would be hard to switch back. 2010-03-16T14:22:48 mck9: there's no need :) 2010-03-16T14:24:53 Good. It's bad enough to be putting my curly braces in the wrong places... 2010-03-16T14:26:56 i'll start working on the APL port so that this space issue will go away :) 2010-03-16T14:27:23 StephenGWills: only if you give us all free keyboards 2010-03-16T14:28:14 does APL have its own character range in Unicode? 2010-03-16T14:28:19 who am I kidding? I can't even get postgreSQL to run as something the staff client can recognize. /me is a human sacrifice candidate. 2010-03-16T14:28:55 the staff client's technology stack does worship XUL 2010-03-16T14:30:08 it sure don't worship me! ;) 2010-03-16T14:30:20 (incidentally, there's a new Ghostbuster's movie in the works:) 2010-03-16T14:34:56 So, while we are all thinking about it, do we want sql files in the real-tabs exception list? I have also seen the auto-complete paste behavior dbs mentioned, but it hasn't actually hurt anything for me yet. A quick scan through some of the representative files does seem to favor tabs. 2010-03-16T14:35:24 Especially the older files which haven't been distorted by we mortals. 2010-03-16T14:35:25 dbwells: the SQL is currently 95% tabs or thereabout 2010-03-16T14:36:00 I am used to :%s/^T/ /g by now 2010-03-16T14:36:09 (before copying/pasting) 2010-03-16T14:36:30 Also... I tried using perltidy, and I can't stand it 2010-03-16T14:37:45 dbs: great (re: sql %). I'll make a few updates to my bare-bones code formatting page so the de facto can become a bit more de jure. 2010-03-16T14:38:11 dbwells++ 2010-03-16T14:39:23 dbs: Also, perltidy does seem to make that first impression. I think one can get used to it and make it work for them, but it might not be worth the effort for all but the most mis-formatted files. 2010-03-16T14:39:51 dbwells: i use it for sections of code at a time 2010-03-16T14:39:58 dbs: instead of copying and pasting db upgrade scripts, i just use \i in psql, fwiw 2010-03-16T14:40:47 senator: easy to do on a local system 2010-03-16T14:41:00 dbwells: in vim, "V" (visual mode, highlight X number of rows), "!perltidy" 2010-03-16T14:41:23 senator: harder to do when you have to schlep said scripts around 2010-03-16T14:41:28 atz: interesting, I have not tried that, thanks for the tip! 2010-03-16T14:41:55 np 2010-03-16T14:42:14 dbs: keep an svn checkout on the db server just for db upgrades? 2010-03-16T14:42:46 senator: if only we could afford to do upgrades at actual release intervals 2010-03-16T14:47:36 the sql you're copying and pasting now is from commits to EG right? wherever you run that sql today, i just mean you might svn checkout the branch they come from on whatever machine you run psql on (wouldn't actually have to be the server) to avoid the need for copy and paste, right? 2010-03-16T14:48:07 in any case, the tabs don't need to stay there, just sayin' 2010-03-16T14:50:13 senator: umm... but normally we only need to apply one part of the upgrade (say, a single CREATE OR REPLACE FUNCTION stmt), rather than the entire schema file, or upgrade SQL script 2010-03-16T14:50:26 ah that's what i'm missing 2010-03-16T14:50:44 *** pmp_afk has quit IRC 2010-03-16T14:51:12 I could still check out the file, copy it, edit it down to what we need... but that's more work than :%s/^t/ /g :) 2010-03-16T15:00:19 *** r123 has left #evergreen 2010-03-16T15:04:01 *** pmplett has joined #evergreen 2010-03-16T15:14:23 *** natschil has joined #evergreen 2010-03-16T15:19:57 *** natschil has quit IRC 2010-03-16T15:31:33 *** r123 has joined #evergreen 2010-03-16T15:38:34 *** alxp has quit IRC 2010-03-16T15:51:13 *** leed- has quit IRC 2010-03-16T15:56:54 *** leed- has joined #evergreen 2010-03-16T16:00:17 *** Meliss has quit IRC 2010-03-16T16:06:56 *** sfortin has quit IRC 2010-03-16T16:06:59 *** bshum has quit IRC 2010-03-16T16:11:09 *** jenny has left #evergreen 2010-03-16T16:18:05 phasefx: longing for a nice text file to hold all of the circ data for fast batch renewals? heh 2010-03-16T16:18:57 I don't follow, but I've been extraordinarily slow these past few days :) 2010-03-16T16:19:58 maybe what we really need is a batch renewal method 2010-03-16T16:20:18 that could take care of the repeated override prompt problem too 2010-03-16T16:20:30 in this case 2010-03-16T16:28:18 that would be awesome! I'm off to take notes at a meeting... 2010-03-16T16:32:49 *** dbs has quit IRC 2010-03-16T16:36:51 *** jamesrf has quit IRC 2010-03-16T16:47:50 *** dbs has joined #evergreen 2010-03-16T17:00:36 *** pmplett is now known as pmp_afk 2010-03-16T18:20:39 *** pmp_afk is now known as pmplett 2010-03-16T18:34:39 *** dbs has quit IRC 2010-03-16T19:09:20 *** jamesrf has joined #evergreen 2010-03-16T19:20:38 *** lisppaste3 has joined #evergreen 2010-03-16T19:35:46 *** mck9 has quit IRC 2010-03-16T20:41:18 don't you think the message "A *TERRIBLE* problem has occurred..." is a little too extreme? 2010-03-16T20:43:06 may not be too extreme, depending on the circumstances 2010-03-16T20:43:19 although I'll grant that it's not a terribly specific error message 2010-03-16T20:44:09 it's a credit 2010-03-16T20:44:09 card transaction was processed successfuly, but the patron's 2010-03-16T20:44:09 payment could not be recorded within Evergreen. Please seek 2010-03-16T20:44:09 assistance. 2010-03-16T20:44:53 i would say "serious problem" or "*serious* problem" 2010-03-16T20:45:05 probably not even *SERIOUS* 2010-03-16T20:53:48 "a problem has occurred" 2010-03-16T20:54:12 or even "an unexpected problem has occurred" -- but then when would you ever have "an expected problem has occurred"? :) 2010-03-16T20:54:39 "an expected problem has occurred: you were cataloging in MARC21, right?" 2010-03-16T20:54:47 "This can never happen" 2010-03-16T20:55:11 phasefx: getting some DB errors after upgrade 2010-03-16T20:55:29 2010-03-17 00:39:30 UTC ERROR: "search_result" is a composite type 2010-03-16T20:55:52 ERROR: "marc21_physical_characteristics" is a composite type 2010-03-16T20:56:51 what version of postgres are you using? 2010-03-16T20:57:13 8.3.7 2010-03-16T20:58:44 gots me. I asked because I remember mck9 implementing a stored procedure using composite types in a certain way and it wasn't supported in pg 8.2, so he re-did it 2010-03-16T20:59:06 yeah, that's what i was thinking of 2010-03-16T21:02:09 i guess i'll ask him tomorrow 2010-03-16T21:02:19 *** dbs has joined #evergreen 2010-03-16T21:14:20 *** jamesrf has quit IRC 2010-03-16T22:07:29 *** pmplett has quit IRC 2010-03-16T22:07:40 *** jamesrf has joined #evergreen 2010-03-16T22:28:33 atz: that's normal, and they're mine, not mck9's ;) 2010-03-16T22:29:54 it comes from cstore trying to interogate what it thinks is a table, but is not, in fact, a table but just a shell type describing (yes) the output structure of a stored proc 2010-03-16T22:32:57 just noise 2010-03-16T22:35:12 * dbs attempts to figure out if it truly is possible to make keywords-in-title more relevant than keywords elsewhere for a keyword search simply by adjusting search.relevance_adjustment 2010-03-16T22:37:33 getting flamed by the comp sci kids for "programming languages" returning a windows regex book as the most relevant hit (because it has about 6 subjects with "programs / languages" in them) 2010-03-16T22:40:24 dbs: if you me a title search, yes. if you mean a keyword (search class) search, no. but I've been musing the use of weights (tsearch2 capability) 2010-03-16T22:40:37 that would require ... lots of new work, though 2010-03-16T22:40:45 I mean a keyword search, of course :) 2010-03-16T22:41:14 Good. Then I should have trusted my first reading of things. 2010-03-16T22:42:16 Also... first_word truly is the first word, not the left-anchored search phrase (if I understand that LIMIT 1 correctly) 2010-03-16T22:43:17 it means the first word of the query, as typed, must match the first word in the value column on the field_entry table, yes 2010-03-16T22:49:38 Right-o. 2010-03-16T22:50:25 So the real answer to Melissa's question from about a month or so back was "Do a title search if you want better relevance ranking for search terms in a title" 2010-03-16T22:50:41 Which the Solr kids would mock. 2010-03-16T22:51:14 But so be it. Better to have that out on the table. 2010-03-16T23:28:42 *** jeff__ has joined #evergreen 2010-03-16T23:31:35 *** leed- has quit IRC 2010-03-16T23:31:35 *** jeff_ has quit IRC 2010-03-16T23:31:35 *** jeff has quit IRC 2010-03-16T23:31:55 *** leed- has joined #evergreen 2010-03-16T23:38:46 *** jeff has joined #evergreen 2010-03-16T23:39:01 * jeff plays with open-ils.circ.money.payment