Open Source Integrated Library System

Evergreen on IRC

#openils-evergreen Logs for Tuesday, March 31st, 2009

< Monday, March 30th, 2009Raw Log FileWednesday, April 1st, 2009 >
#TimeNickMessage
#00:02:06jeffGOT IT!
#00:02:13jeffquirks mode.
#00:02:19jeffit was quirks mode.
#00:02:28dbsjeff: i was just running it through tidy and saw that
#00:02:41dbsline 113 column 49 - Warning: discarding unexpected <table>
#00:02:44jeffthe config changes in 1.4 that pass it through XMLENT also set a content-type that forces quirks mode.
#00:02:50jeffwoo!
#00:03:00jefftidy++
#00:03:01dbscan't have tr/table
#00:03:04dbsjeff++
#00:03:16jeffdbs++
#00:03:18jeffphasefx++
#00:03:19jeffand...
#00:03:39jeffscroll, scroll, scroll...
#00:03:49jeffdchristens++
#00:03:50jeffdchristens++
#00:04:06jeffugh. what fun!
#00:04:50phasefxhuzzah
#00:07:50dbsyeah, just throwing a <td></td> around the offending table will fix it I think, just testing now
#00:10:13dbsyep
#00:10:26jeffworks here!
#00:12:23dbsI'll delete the errant Cookie.js while I'm at it
#00:15:53jeffand we'll have a working closed dates editor again.
#00:17:06brendan_ga has quit IRC
#00:18:19jeffso the other amusing bit here is that standards mode was fine with the error, and quirks mode choked on the error. :)
#00:18:52jeffunless i'm flipping a bit somewhere
#00:21:15jeffchecked the passed-through-xmlent output of the other xhtml files in xul/server/admin, no similar table errors.
#00:23:40dbsjeff++
#00:24:20jeff(i'd have been surprised to find any, since it was so deep in that specific interface)
#00:25:10jeffand re standards/quirks... quirks mode just handled it worse. i suppose that's not surprising.
#00:25:20dbsbtw, we force quirks mode because of IE6, IIRC
#00:25:27dbsdie, IE6, die.
#00:26:44jeffwell, this specific bit only affects things in /xul/*.xhtml
#00:29:31jeffi'm pretty sure that XMLEntContentType was what was forcing quirks mode on this file.
#00:32:25jeffbut i'm losing focus:
#00:32:31jeffdie, IE6, die.
#00:34:36brendan_ga has joined #OpenILS-Evergreen
#00:36:11brendan_gais there a 1.4.0.4 client version
#00:36:56jeffbrendan_ga: not packaged for download yet.
#00:37:00jeffphasefx: get cracking!
#00:37:27brendan_gaok -- any idea how long?
#00:37:46phase_bbHa, .2 isn't good enough for you guys? :)
#00:38:03brendan_gacould I use .2 on my .4 server?
#00:39:08jeffdbs: for the record, the &quot; bits in the onclick= attributes seem to be just fine, complete red herring. :P
#00:39:46dbsjeff: yeah, I thought that was the case but it's less confusing to have it show up that way; I was in the file anyway, so
#00:39:49dbs...
#00:40:22jeff nods
#00:40:24dbsbrendan_ga: sure - just create a symlink for rel_1_4_0_2 in /openils/var/web/xul
#00:40:39dbsthere were only a few minor client-side changes
#00:40:44brendan_gathanks that is what i was wondering....
#00:41:12brendan_gathanks, dbs, jeff, phase_bb off to hack in my server
#00:41:38dbs crashes (before his laptop does, for a change!)
#00:41:45dbs has quit IRC
#04:13:13till_ has quit IRC
#04:17:51till_ has joined #OpenILS-Evergreen
#05:22:51BenO has joined #openils-evergreen
#05:41:36till_ has quit IRC
#07:23:09till__ has joined #OpenILS-Evergreen
#07:43:20till__ has quit IRC
#08:27:39atheos has quit IRC
#08:54:10BING2 has joined #openils-evergreen
#08:59:57BING2 has left #openils-evergreen
#09:09:30atheos has joined #OpenILS-Evergreen
#09:34:26till_ has joined #OpenILS-Evergreen
#10:25:44rsinger has quit IRC
#10:30:53rsinger has joined #OpenILS-Evergreen
#10:32:03meliss has joined #OpenILS-Evergreen
#10:37:49phasefxha, so I tried building a 1.4.0.4 client and the new_tab bug from trunk had slipped in afterall
#10:39:57jeffheh
#10:40:03jeffoops. :P
#10:42:08phasefxbut strangely enough.. just doing a dirdiff on .2 versus .4, I don't see how
#10:47:30jeffwhat's the symptom?
#10:49:33phasefxwindow.xulG.new_tab is not a function for things in Local Admin that should spawn a new tab
#11:26:28_bott_phasefx: sockeye should be ready for you now
#11:31:13phasefx_bott_++
#11:31:13_bott_ has quit IRC
#11:32:12_bott_ has joined #OpenILS-Evergreen
#11:46:27rickd_ has quit IRC
#11:46:53rickd has joined #OpenILS-Evergreen
#11:53:59dchristens has joined #OpenILS-Evergreen
#11:57:29eby has joined #OpenILS-Evergreen
#12:06:48jamesrf has joined #openils-evergreen
#12:29:05dchristenswow. reading the IRC logs from last night and this morning... jeff++ dbs++ phasefx++ ... you guys rock!
#12:29:29dchristensoop - miker_++ too :-)
#12:30:51jeffdchristens++ # thanks for reporting it!
#12:30:53dbs has joined #openils-evergreen
#12:41:47brendan_ga has quit IRC
#13:01:00brendan_ga has joined #OpenILS-Evergreen
#13:14:37phasefxjeff: okay, for whatever reason, the bug is only happening with 1.4 clients against trunk, not against 1.4 servers (half-phew)
#13:14:52phasefxwhich blows my mind, but oh well
#13:15:38jeffphasefx: there was a fancyprompt bug that only threw when running against a server with versionless support, i think...
#13:16:22phasefxeww
#13:16:24jeffphasefx: might ask _bott_ about it, we had it on the 1.4 dev/testing box. any fancyprompt failed, until (i think) whatever supported the versionless client builds was removed.
#13:16:26BenO has quit IRC
#13:16:36jeffi experienced, he fixed.
#13:17:21dbs goes to check backlog
#13:17:36phasefxyou can still have a client without a version, such that it's looking for /xul/server/ and not /xul/BUILD_ID/server/, but what's not supported anymore is running the client directly against application.ini in xul/staff_client/.. has to be out of xul/staff_client/build/
#13:20:44dbsthink I'm going to backport the slimpac apostrophe fix now, before I forget about it, unless there are any objections to my braindead approach
#13:21:40phasefxno objections from me, but I tend to go for braintormented approaches
#13:21:44dbshttp://svn.open-ils.org/trac/ILS/changeset/12717
#13:22:08jeffphasefx: yeah, i don't know the specifics, because i haven't had do to it myself. it could be something other than the versionless stuff. maybe some other wildcard? dunno. i do know i was running a stock 1.4.0.2 staff client for windows.
#13:22:24dbs-= THIS MESSAGE NOT LOGGED =-
#13:22:35jeff-= THIS MESSAGE NOT LOGGED =-
#13:22:35phasefxdbs: jeff: you guys can give this a try? http://evergreen-ils.org/~phasefx/evergreen-setup-rel_1_4_0_4.exe
#13:23:06phasefxI had to learn how to compile the client without actually installing OpenSRF and Evergreen
#13:23:32dbsphasefx++
#13:23:40dbswill test it out for you, sir
#13:24:31phasefxdbs: you pointed me in the right direction ./configure --disable-core --disable-web --disable-reporter --with-opensrf-libs=/some_installation_directory/src/lib/opensrf/.libs/ --with-opensrf-headers=/some_installation_directory/include/
#13:25:10phasefxand then going into the client directory, make STAFF_CLIENT_BUILD_ID=foo OPENSRF_JSLIBS=/some_installation_directory/src/javascript/
#13:25:15dbsphasefx: cool! although I still think --disable should disable the checks for those libs and headers :)
#13:25:16phasefxand then going into the client directory, make STAFF_CLIENT_BUILD_ID=foo OPENSRF_JSLIBS=/some_installation_directory/src/javascript/ install
#13:26:07jeffthrow that on a scratchpad!
#13:26:20phasefxnow I'm wondering about the demo server.. it appears to have 1.4 on it, but all the symlinks for the client are for 1.2 :) and the download page mentions 1.2
#13:26:53jeffdemo was 1.2.0.4 i thought.
#13:27:04dbsdemo is 1.2.4.0
#13:27:09dbsdev is 1.4
#13:27:10jeffer, yeah.
#13:27:16dbsacq is trunk
#13:27:37jeffnot the first time i've merged 1.2.4.0 and 1.4.0.4 like that. :P
#13:27:45phasefxoh duh, I was using a bookmark for the ils_version method, and I could have swore it was pointing to demo
#13:28:27phasefxso demo has some rel_1_2 branch pulled from I don't know when
#13:28:53jeffmaybe from when rentals were added for me to test?
#13:32:12phasefxtime stamps in http://demo.gapines.org/opac/common/js/ are from November
#13:36:57atheoswow, only documentation for installing Evergreen has come a long way. Just did an install of 1.4.0.4 and it was a breeze.
#13:38:46dbs makes a note: "March 31, 2009: first positive comment about installing Evergreen"
#13:39:31phasefxhaha
#13:39:42phasefx"it could be worse"
#13:39:45atheoseek, and my sentence didn't make sense. my thought was "only started a couple hours ago, and documentation....."
#13:41:19atheosnfs export, and soon I'll be editing evergreen from within TextMate. I'm stoked
#13:41:59phasefxatheos: _bott_: I forget, are either of you guys actually supporting Mac staff clients, or just playing with them? You may want to poke UPEI and see if there is any room for collaboration there.
#13:42:54atheosI've only installed the staff client on my own Macs, but I don't know how much I could offer anyone else in that area.
#13:43:35phasefxthey posted a patch earlier, but I don't even have a mac (yet) to test it against
#13:43:42dbsthere is also Mac interest at one of our college libraries
#13:43:52jeffwe have at least one mac staff client
#13:44:08atheosquite frankly, I've had zero issues getting the staff client to build.
#13:44:35phasefxin the early days, when I did have a mac (I break things), the mac xulrunner seemed kind of flaky
#13:45:32atheosI interpret (I break things) as, I throw laptops too
#13:45:58phasefx did put a dent in a titanium powerbook, but swears it was just by normal handling
#13:46:01atheosmy powerbook screen just died. I can't get 2 full years from a powerbook without failure
#13:46:10jeffphasefx: does 1.4 work with xulrunner 1.9 yet?
#13:46:27dbsI don't think it's the building of the staff client, just some of the keybindings
#13:46:40phasefxjeff: it probably works, _if_ you tweak application.ini to up the allowed xulrunner version
#13:46:52dbsat least, as reported by our folks on the ground at the college - problems in the marc editor interface
#13:47:07phasefxI think there may be some xulrunner 1.9 tweaks that are only in trunk at the moment
#13:47:18atheosah, I don't use those features myself. I did read the email that supplemented that patch though. something about conflicting keybindings with osx
#13:47:44phasefxmy plan is to support only xr1.9 after eg1.4
#13:47:49eby has quit IRC
#13:48:17dbsyeah, would like to point our people at that patch and encourage them to generate more such patches if necessary, but the technical skill isn't there
#13:52:11dbsphasefx: so far, so good on 1.4.0.4 client, although I stalled at import z39.50 to track down a "record saved but no ingest" problem at http://dwarf.cs.uoguelph.ca/ingest_fail.txt
#13:53:41phasefxczech work?
#13:54:33phasefxoy, ingest :)
#13:55:48jeffphasefx: ah, yes. i was looking at trunk and not 1.4, and seeing the minversion at 1.9 already. :)
#13:56:50phasefxtrunk client still needs some polish for xulrunner1.9 too; we lost some CSS styling since 1.9 won't let remote pages load chrome CSS files
#13:57:54jeffand in a similar vein, is postgresql 8.3 support "just" a matter of a lot of explicit ::text casting?
#13:58:10dbsphasefx: czech is complaining about the version for some reason - are you checking (czeching?) for server-side locale now? if so, cool
#13:58:22dbsjeff: I'm running 1.4 against 8.3
#13:58:23jeffi seem to remember there was some fulltext stuff too.
#13:58:32jeffdbs: ah, i think you were who i was thinking of.
#13:58:56dbsit's all been backported to 1.4.0.2, at least as much as I'm aware of. that said, still not "supported"
#13:59:16jeffunderstood.
#13:59:41phasefxdbs: I'm not doing such a thing explicitly, but it may be the server noticing the locale and not falling back to en-US?
#14:00:11dbsphasefx: mebbe. very interesting
#14:00:21dbs goes to symlink hy-AM to cs-CZ
#14:02:07phasefxdbs: for example, I'm getting a 500 internal error with the version check against my trunk instance for czech
#14:02:26phasefxen-CA is working, though
#14:03:19dbsthe symlink works perfectly
#14:04:04phasefxso what did you symlink exactly?
#14:04:26phasefxis it xmlent looking for something that doesn't exist based on locale?
#14:04:28dbscd /openils/var/web/opac/locale; ln -sf hy-AM cs-CZ
#14:04:45dbs(switch hy-AM to en-US if desired)
#14:05:02dbshmm, problem when logging in though, the cs-CZ build isn't complete on the client side
#14:05:02phasefxcoool
#14:06:45dbsahh, no
#14:06:56phasefxhey! I got success against that server Vaclav Jansa advertised
#14:07:07dbsalso need to cd /openils/var/web/xul/rel_1_4_0_4/server/locale; ln -sf hy-AM cs-CZ
#14:07:28natschil has joined #openils-evergreen
#14:07:29phasefxsome things are incomplete.. New Window and New Tab aren't translated
#14:07:41dbsso the client side is fine, and the only reason I need to create those symlinks is because I didn't build our server with cs-CZ
#14:07:46phasefxCheck Out Items, Check In Items, Register Patron
#14:07:57phasefxSearch the Catalog
#14:07:59dbsphasefx: oh, I know. Some of those strings were added post 1.4
#14:08:10dbsand some of them are still hardcoded.
#14:08:22phasefxI'm shocked they didn't translate "stub" :)
#14:08:47phasefxgood enough to put up for download you think?
#14:08:55cpeach has joined #openils-evergreen
#14:10:47dbs believes so
#14:14:31natschilhello, I just realized that when adding the evergreen search to firefox, it uses the html only search... is there a specific reason for it doing so?
#14:16:21cpeachHi all, im wondering if anyone might know if there is any webservice related documentation or a simple list of service definitions available for Evergreen. Any direction would be muchly appreciated!
#14:17:39natschilcpeach: have you seen the dokuwiki?
#14:17:55natschilcpeach www.open-ils.org/doku/doku.php? or something like that
#14:18:17jeffhttp://evergreen-ils.org/dokuwiki/doku.php
#14:18:20dbsor are you looking for something like http://dev.gapines.org/opac/extras/docgen.xsl
#14:18:25_bott_phasefx: Re: Mac I'm not really supporting them, I only have myself to worry about. John @ BDL may have some, but they may only be OPAC
#14:18:37dbseg: http://dev.gapines.org/opac/extras/docgen.xsl?service=open-ils.search&param=&all=on&offset=0&limit=25
#14:18:42natschiljeff: yeah, sorry, that first website was the wrong url
#14:18:53jeffnatschil: no problem.
#14:20:08phasefx_bott_: roger that
#14:20:17dbs_bott_: I'd love to pick your brain about the silent ingest failure you were experiencing; did you end up making it non-silent to figure out it was your custom field causing the problem? Because we're still having ingest problems like http://dwarf.cs.uoguelph.ca/ingest_fail.txt
#14:20:51jeffoh, that reminds me.
#14:22:21jeff_bott_: i was trying to force a break in the fixed fields editor by setting a record to an invalid record type (leader/06), and i couldn't get it to take on dev-tadl. Hadn't poked at it much, and it might just no longer be possible to break them like that, but fyi, if you're still having any ingest-related issues and that symptom helps.
#14:22:37jeffchanging another field in the record seemed to work okay.
#14:23:10jeff("couldn't get it to take" == marc editor reports record saved, then record refreshes without the changes that were being saved)
#14:23:16dbsnatschil: I think it's just that that's where the opensearch description points to in opensearch_osd() in OpenILS/WWW/SuperCat.pm - you might try poking around there?
#14:23:16_bott_dbs: I just went nuts with log_info messages throughout Ingest.pm until I found where it died. In my case on a custom 'type'. Around line 725, if I'm remembering correctly.
#14:24:24dbs_bott_: heh, okay - thanks
#14:26:29phasefxdownload page updated
#14:27:42dbsphasefx++
#14:28:59natschildbs: ok, I'll try and do that, thanks
#14:29:49dbsnatschil: 'twould be cool to have it search the dynamic OPAC, agreed
#14:30:10natschildbs: yeah, 'twould. is there a specific reason why it doesn't?
#14:30:38dbsnatschil: I don't know, it predates my involvement
#14:31:16cpeachdbs: I believe this is exactly what im looking for.. Is there a way i can find the accompanying WSDL's or a definitions of the service params. I checked http://evergreen-ils.org/dokuwiki/doku.php, but was unsuccessful in my search. Are there code samples of any kind?
#14:36:28miker_cpeach: there's no (big W, big S) Web Services interface, no SOAP. there is, however, XMLRPC
#14:46:57cpeachmiker_: okey thanks. I will work with that.
#14:58:08_bott_dbs: looked through that .txt -- maybe I'm not looking for the right thing, but it seems that your bre doesn't contain much of anything to ingest
#14:59:29dbs_bott_: there's a huge marc field in there
#14:59:51dbsbut yeah, null null null null
#15:00:38dbs'course, lots of those nulls just get default s
#15:10:43dchristenshey folks. Is there a list somewhere of the various scripts that should be cron'd? (reshelving, thaw expired holds, fine generation, ...?)
#15:13:04dbsnatschil wrote up some docs at http://www.mail-archive.com/open-ils-documentation@list.georgialibraries.org/msg00011.html but I don't know if there's anything in the wiki yet
#15:13:17dbsnatschil++
#15:13:27dchristensdbs: thanks
#15:13:33dchristensnatschil++
#15:13:51dbsnatschil: did you want a wiki account, btw? I'll bbiab, but just let me know if you want it
#15:14:59natschilbtw, I'm not sure whether they're entirely complete, especially what to do with the reshelving_complete.srfsh script, as I think the values I said you need to set there need to be set somewhere else, so use at your own risk :)
#15:16:47dchristensheh - duly noted :-)
#15:16:53natschildbs: I'm not sure if I need a wiki account, but if its not much work for you, and if you don't mind people with wiki accounts that don't really write many wiki articles, then it might be easier to just post stuff there if I come up with any more documentation
#15:17:32natschildchristens: also, I think some of those scripts aren't used too much anymore, but I'm not sure
#15:18:11rsinger has quit IRC
#15:18:40rsinger has joined #OpenILS-Evergreen
#15:19:51natschil feels honoured to be offered a wiki account :)
#15:20:26natschil (even though he hasn't really made any documentation)
#15:22:03jeffnatschil: namespaces are useful if you want to throw some things together. see some of the things under "Temporary Scratch Pad" on the main dokuwiki page, http://evergreen-ils.org/dokuwiki/doku.php
#15:25:20natschiljeff: yup, isn't that a bit like the open-ils-documenation mailing list?
#15:25:20dchristensWould anyone care to share their crontab? That would definitely make it easier to see... especially if it were from a "live" system.... ;-)
#15:51:10natschil has quit IRC
#16:00:56meliss has quit IRC
#16:02:02BING has joined #openils-evergreen
#16:07:44BING has left #openils-evergreen
#16:14:08brendan_gaHey All -
#16:14:39brendan_gaOne the latest server that I built I am having a search issue --- when I search for a one word keyword - I get
#16:14:46brendan_garresultFilterSearchResults(Object locale=en-US service=open-ils.search)rresult.js (line 306)
#16:14:46brendan_ga_remoteRequestCallback(0)RemoteRequest.js (line 177)
#16:14:46brendan_gaonreadystatechange()()RemoteRequest.js (line 194)
#16:14:46brendan_ga[Break on this error] if( result.count > 0 ) {
#16:14:49brendan_gaI get nothing
#16:15:00brendan_gabut if I search on two words - it works...
#16:15:20brendan_gaEG 1.4.0.4 and opensrf 1.0.6
#17:00:49_bott_ has quit IRC
#17:01:08_bott_ has joined #OpenILS-Evergreen
#17:04:27dbs has quit IRC
#17:26:40dchristens has left #OpenILS-Evergreen
#19:45:11grahamf has joined #openils-evergreen
#19:48:14djfiander has joined #OpenILS-Evergreen
#19:49:31rsinger has quit IRC
#20:03:06rsinger has joined #OpenILS-Evergreen
#20:12:04brendan_ga has quit IRC
#20:20:04djfiander has quit IRC
#20:22:14djfiander has joined #OpenILS-Evergreen
#20:26:18djfianderbecause I let something slip, I am now MPOW's expert on MFHD. I must now work hard to ensure that somebody else actually has to do the work.
#20:54:47dbs has joined #openils-evergreen
#21:11:30brendan_ga has joined #OpenILS-Evergreen
#21:13:54dbsso far, no complaints about the 1.4.0.4 vbox image. either it's perfect (hah!) or nobody has downloaded it (unlikely) or the masses are gathering to assassinate me.
#22:18:30dbsbtw, yaz 3.0.45 released today
#22:21:39grahamfhey dbs
#22:21:48dbsgrahamf!
#22:22:03grahamfyou happen to know if Net:Z3950 is a wrapper for libyaz, or is it pure Perl?
#22:23:02grahamfjust semi-curious. II; was having some issues with PyZ3950 (which is not yaz based); right now I'm running yaz-client using 'expect' which is, well, suboptimal
#22:23:56dbsNet::Z3950 is a libyaz wrapper
#22:24:42dbsseems that pretty much anything non-yaz-based has problems connecting to the Evergreen Z39.50 server at this point
#22:25:13grahamfok. I might check out the source, see what the cost of writing a Python wrapper might be like.
#22:25:37grahamfthough Expect really isn't hideous, especially for the little amount of Z39.50 we'd be doing in Syrup
#22:26:19grahamfso, I'm reading perlmods/Open-ILS/SIP, you know, just for fun
#22:26:56grahamfit looks like an Item Information request will return both holding/status info as well as a significant bib record (mvr), yes? that would be very good.
#22:29:19djfianderfuck I don't know
#22:30:18dbsgrahamf: djfiander is the SIP expert for Evergreen :)
#22:30:26grahamfdjfiander: heh
#22:30:52grahamfplus I hear he's quite handy with MFHD.
#22:31:00djfianderI'm buried deep in the middle of trying to figure out how to pull repeatable subfields out using MARC::Field.
#22:31:38grahamfI can only imagine the pain
#22:34:05djfianderfuck.
#22:36:00grahamfdjfiander: dare we ask?
#22:36:24djfianderutsl.
#22:36:40djfianderMARC::Field->subfields returns a list of array refs.
#22:36:58djfiandereach array ref is a list consisting of the subfield indicator followed by the values.
#22:37:07djfianderrather than every subfield coming out separately.
#22:37:26djfianderso, I get ONE return value for all of the subfields 'y' that appear.
#22:37:52grahamfnot good.
#22:38:07djfianderI think.
#22:38:08djfiandershit.
#22:39:58djfianderok. time to stop for tonight.
#22:40:25grahamf:-) better luck tomorrow, djfiander.
#22:40:26djfiandersubfield works that way, but now I think that subfields works the way it should, but the data I'm getting back is odd.
#22:40:35djfianderI'll have to write a test program to see
#22:42:06djfianderbtw. msauers reports that the Hilton lobby bar is out of beer on tap at CiL
#22:42:58djfiander has quit IRC
#22:55:39grahamf has quit IRC
#23:22:21dbs feels a bit more groovy getting z39.50 server support working with USMARC output and MarcEdit
#23:38:19jamesrf has quit IRC
#23:47:15sarabee has quit IRC
#23:52:48jamesrf has joined #openils-evergreen
< Monday, March 30th, 2009Raw Log FileWednesday, April 1st, 2009 >