IRC log for #koha, 2009-07-01

← Previous day | Today | Next day → | Search | Index

All times shown according to UTC.

Time Nick Message
12:02 hdl_laptop1 hi |Lupin|
12:06 |Lupin| hi hdl_laptop1
12:07 hdl_laptop1: many thanks for the link you provided last time on "recommandation 995". Actually it was exactly what we were looking for.
12:07 Amit hi hdl_laptop
12:11 hi jdavidb
12:11 jdavidb Hello, Amit.  :)
12:20 hdl_laptop1 |Lupin|: np
12:50 |Lupin| gmcharlt: here ?
12:57 gmcharlt hi |Lupin|
12:58 |Lupin| hi gmcharlt
12:59 gmcharlt: last time I suggested to talk about the accessiblity of Koha's Intranet during tomorrow's meeting and you said a message on the list would be nice
12:59 gmcharlt which you sent, but feel free to bring it up during tomorrow's meeting as well
12:59 |Lupin| gmcharlt: since the message did not receive many replies, do you think it'd still be relevant to talk about this during the IRC meeting ?
13:00 gmcharlt yes
13:00 |Lupin| gmcharlt: ok. Just waned to check it won't overload the meeting.
13:00 gmcharlt: should it be added to the wiki page ?
13:01 paul_p hi gmcharlt & |Lupin|
13:01 (hi amit & jdavidb as well)
13:01 gmcharlt |Lupin|: go ahead and add it to the wiki page
13:01 hi paul_p
13:01 jdavidb Howdy, paul_p!
13:02 |Lupin| gmcharlt: ok
13:02 hi paul_p
13:19 gmcharlt: done
13:20 hmm if only those how have added their names to the page will attend the meeting, that's gonna be a very private meeting.. :/
13:20 owen ebegin: I thought that was a good response to the "nobody won't replay" Koha list message
13:20 ebegin owen :) thanks
13:20 gmcharlt |Lupin|: don't worry, many more will attend
13:21 ebegin++
13:21 ebegin Someone has to tell him
13:24 jdavidb gmcharlt:  I have the meeting tomorrow down for 4 PM Eastern US time...zat right?
13:24 gmcharlt jdavidb: 3 p.m. EDT
13:24 jdavidb *grumble*  
13:24 That's what I get for moving my computer between time zones.  Things get confused.
13:25 owen There wouldn't have been any donuts left by the time you got there
13:27 schuster Thanks for the reminder!  and the time..  I always mess up the time!  Late early etc... wrong day! HA...  
13:28 jdavidb All right, I've spent the last hour on this, and am utterly stumped.  What would cause fines.pl to throw this:
13:28 No branchcode argument to new.  Should be C4::Calendar->new(branchcode => $branchcode) at fines.pl line 134
13:29 I've checked, and all of the issues have a branchcode, as do all the borrowers, so it's not a stray NULL that doesn't belong.
13:29 |Lupin| so in France it will be at 9PM, right ?
13:30 gmcharlt jdavidb: $branchode is not undef, and there's no silly typo with the hashkey name floating around?
13:31 jdavidb Haven't spotted any obvious ones.  It's doing this on three different data sets, all running HEAD, so it makes me wonder...
13:34 At first, I suspected trashy data...imma insert some testing statements to see if I can find out.  Hard to imagine on three different servers, though.
13:38 |Lupin| When importing MARC records in Koha, where wold you store old records number you want to keep track off just in case, please ?
13:39 (it's Unimarc btw)
13:42 hdl_laptop1 009 for instance
13:43 jdavidb Hrm.  An issue, with an itemnumber that does not exist in items...there's my bandit.  Now, how did I end up with those in three different sets?
13:45 |Lupin| hdl_laptop1: thanks !!!
13:48 hmmm is there a method somewhere to read marc records in bathc mode from a scalar value rather than from a file ?
13:52 slef @seen fredericd
13:52 munin slef: fredericd was last seen in #koha 1 week, 1 day, 4 hours, 21 minutes, and 53 seconds ago: <fredericd> |Lupin|: \n in MARC records sound not good!
13:52 slef fredericd: are you there?
13:54 any librarians available?
13:54 (and yes, I know fredericd is a developer...)
13:54 |Lupin| slef: available but not librarian
13:54 slef |Lupin|: sorry, I need an insight into librarianship
13:54 jdavidb I'm not a librarian, and I don't play one on television, but I can try.  I've spent my whole career around them.
13:54 |Lupin| has someone noticed that unimarc.info is down ?
13:55 gmcharlt |Lupin|: you can tie a scalar to a file handle, or just split on \x1D and parse each blob into a MARC::Record object
13:55 collum slef: Librarian here
13:56 slef jdavidb, collum: thanks.  will PM collum
13:57 |Lupin| gmcharlt: ok. How do you do the parsing part ?
13:58 gmcharlt |Lupin|: my $marc = MARC::Record->new_from_usmarc($blob)
14:00 Sharon I'm a librarian, what's up?
14:00 |Lupin| gmcharlt: oh ok ! thanks !
14:05 can anyone help Re:unimarc.info ?
14:05 there was an electronic version of the unimarc manual there, in HTML, very handy
14:06 I wish I downloaded the whole site while it wwas still alive...
14:08 slef nah, it's social issues or psychology or something
14:09 how to approach librarians about particular issues
14:09 sorry not to do it in public, but the case studies involve, well, live cases :)
14:09 Sharon unrelated, we discovered last night that when you renew an item on loan from another library, the book renews with the owning library's policy, not the transacting library's policy.  
14:09 jdavidb If it's an academic librarian, just say, "I have a paper due tomorrow."  They'll sigh, stare at you like you're an idiot, and then give you the answer you need.
14:10 slef |Lupin|: there's something there for you
14:10 Sharon: is that a bug or a feature?
14:11 Sharon Bug.  It confuses a patron to have their book circ first for 2 weeks then renew for 3
14:12 one of those things that wouldn't be an issue in a stand-alone installation, but it doesn't work so well in a 30+ library consortium
14:12 slef Sharon: can you file it on bugs.koha.org along with a note of what it should do, please? (renew on original library's policy?)
14:12 Sharon will do.
14:12 slef sorry, transacting library
14:12 Sharon yep. circ and renewal policies need to match the transacting library.
14:13 |Lupin| slef: thanks. I tried but the page I was looking for was not there.; thanks however.
14:15 slef Sharon: don't file on bugs.koha.org
14:15 Sharon: looks like a local configuration error to me
14:15 Sharon: can you check your ItemHomeLibrary system preference?
14:15 Sharon can do
14:16 slef sorry, CircControl system preference
14:16 I misread a line :)
14:17 Sharon It's set for Pickup library
14:17 slef ok, so it might be a bug worth reporting after all
14:17 thanks for checking
14:17 Sharon no problem.  I hadn't seen that syspref before.
14:19 magnusenger Hi all!
14:19 i'm having a first look at the files in nb-NO/xslt and find a lot of english texts, like "Author" etc
14:20 gmcharlt magnusenger: right - the XSLT are not run through the translator at present
14:20 magnusenger should these XSLT-files be translated seperately from the other translations?
14:20 ok, so they should just be translated?
14:21 any plans for including then m in the general translation mechanism?
14:22 then m = them
14:25 slef I think paul_p or chris are the people to ask.
14:25 paul_p someone looking for me ?
14:26 mmm... I don't have any idea. /me beeps hdl_laptop1, but i'm afraid the answer is no (although it should be fixed ;-) )
14:27 hdl_laptop1 I think it has to be fixed.
14:27 magnusenger ok, thanks! I'll have a go at translating them into Norwegian then. And ponder any tweaks to make them work with our beloved NORMARC...
14:27 hdl_laptop1 But we have then to decode them.
14:27 and get them into po. Which is not so obvious a task
14:29 magnusenger ok
14:30 slef no getmsg for xml yet?
14:31 actually, can't we use xml:lang attributes and then extract/cache xsl of used languages or am I missing obvious complexities?
14:32 hdl_laptop1 slef: no.
14:32 you are not.
14:33 slef just needs sponsors, huh?
14:33 gmcharlt slef: yeah, that seems like it woudl be the easiest way to handle
14:33 hdl_laptop1 But the point is that it would then be totally different from what has been done for templates.
14:34 gmcharlt hdl_laptop1: different may be OK, as long as translation can be handled via PO files
14:34 using xml:lang woudl be a way of helping to manage extracting strings from the base XSLT files
14:34 hdl_laptop1 if you have one big xsl file containing all the translations... ?
14:35 Ah one language per xslt so only one lang ?
14:39 slef could do it various ways, but I was thinking one big xsl file, or split by modules and use some sort of include statements, then xsl the xsl to combine and split by language
14:39 I don't mind. I'll leave it to you l10n experts :)
15:03 |Lupin| 2-226-11325-8 can someone see what's wrong with this ISBN, please ?
15:04 Business::ISBN says it is not valid...
15:05 slef http://www.isbn-check.de/check[…]bmit=test&lang=en
15:08 ebegin |Lupin|, any error code?
15:09 |Lupin| ebegin: my script does not look for them... For the moment i was just counting how many of our records have valid isns..
15:09 this one was considered invalid as many many others, I'll try to figure out why
15:10 ebegin try by removing the dashes (-)
15:11 |Lupin| ebegin: Business::ISBN doesn't like them ?
15:11 ebegin The first test it does is about the lenght... if 10, if 13 else BAD-ISBN
15:12 |Lupin| ebegin: I checked, the one passd to the module is 2226113258
15:12 ebegin is 2226113258 tagged as invalid by B::ISBN?
15:15 slef wb brendan
15:15 brendan hello slef
15:17 |Lupin| ebegin: now I'm trying it and it says it is valid. However in my program it should have printed a line for all the vlid isbns and no line was printed for that record.
15:17 well I'll add some debugging info in the script
15:22 ebegin |Lupin|, I think that the problem is probably in your code. perl -MBusiness::ISBN -e 'print Business::ISBN->new( 2226113258 )->is_valid,"\n";' returns 1
15:23 schuster At checkin when you need to transfer something would it be hard to change the response from the code to description? ie Please return The princess and the god / to PROF
15:24 Instead of PROF to Professional Library?
15:24 slef wasn't there a patch about that recently?
15:24 it was showing categorycode instead of the description in error
15:24 branchcode even
15:25 schuster hmmm maybe...  I'm behind on my patches so I'll go be a pest someplace else.  Thanks slef!
15:25 owen Looks like that was just aimed at the opac: http://bugs.koha.org/cgi-bin/b[…]w_bug.cgi?id=3292
15:25 munin 04Bug 3292: enhancement, PATCH-Sent, ---, matthias.meusburger@biblibre.com, REOPENED, in opac branchcode should be replaced with name
15:25 slef anyway, not hard at all, it's probably a one-liner in something in circ
15:25 might be three lines if we're unlucky
15:26 schuster I'll revisit it when I have all my patches!
15:26 slef ok, thanks
15:30 owen $err{msg} = $branches->{ $messages->{'IsPermanent'} }->{'branchname'};
15:30 ..not sure why that isn't giving a library name already
15:32 |Lupin| ah actually there was a problem in my sql query so the record was not even examined... uh...
15:35 schuster Owen - which version of Koha?  I'm  3.01.00.014 - which I know is behind on some patches.
15:35 owen No, I can confirm that it *doesn't* show the name in the latest version
15:36 schuster hmmm - guess I go back to my original question...  How to get it to have the Name rather than the code Please return The princess and the god / to PROF
15:37 owen That is the question! Because to my unskilled eye the code should already produce the result you desire
15:38 schuster Fiddle.
15:39 owen I must be looking in the wrong place...
15:40 slef owen: file and line number, please?
15:42 owen I'm looking at the code starting with #Error Messages on line 340 of returns.pl
15:42 But I think I'm trying to match against the wrong section of the template
15:43 Yeah, instead I should be looking at what populates <!-- TMPL_VAR NAME="homebranch" --> on line 166 of returns.tmpl
15:44 slef yeah
15:45 owen Got it
15:45 Line 192: homebranch       => $biblio->{'homebranch'},
15:46 -->  homebranch       => $branches->{ $biblio->{'homebranch'} }->{'branchname'},
15:46 slef oh so we're on about after the transfer has been initiated?
15:47 the page returned by the dotransfer
15:54 atz owen: a biblio doesn't have a homebranch
15:54 an item does
15:54 not sure how that "$bibliio" is actually constructed though...
15:55 owen All I know about it is that when I pull that lever there the ice cream comes out.
15:55 schuster owen - thank you...;)  I love icecream...
15:56 owen atz: line 186, my $biblio = GetBiblioFromItemNumber($itemnumber);
15:57 atz yeah, that one has all 3 tables info in it
15:57 biblio, bibilioitem and item
15:57 gmcharlt fortunately, item information is only from that specified item
15:58 atz that function is a bit costly for several places it is used... would be nice to have the ability not to get the marc and marcxml blobs, for example
16:06 schuster OK along those same lines - some messages get the pretty yellow box and others don't...  What templates control the error message box in Checkin?
16:08 |Lupin| till soon, everybody
16:13 owen schuster: Yellow is for alerts/warnings, blue is for messages
16:14 schuster OK so Not checked out - is a?
16:14 owen Message.
16:15 schuster And there is a background with that which is supposed to be blue?
16:17 owen Yes
16:18 schuster hmmm on my system it shows as a very faint grey - you can hardly tell it is a different screen color than the background.
16:19 Are those controlled through a template? and which one?
16:19 owen Through the stylesheet. The background is this image: /intranet-tmpl/prog/img/message-bg.gif
16:22 schuster hmm hard to make that blink red...;)
16:23 owen schuster: Is the problem that you're trying to highlight certain events in a different way?
16:23 schuster Yes people are not watching the screen so they miss the events...  The Yellow has caught their eye...
16:24 Can't wait for the sounds with 3.2...
16:24 owen If you make everything yellow it will stop catching their eye ;)
16:24 So "not checked out" in particular is an event you want to highlight?
16:24 schuster As one librarian always told me "if I could only get that arm to reach out and smack them"...
16:24 CGI986 hello
16:24 hello every one
16:25 owen Hi CGI986
16:25 CGI986 ihy owen
16:25 schuster I'd like them all to highlight yes.  Guess I could change the message-bg.gif copy it to blue.bg.gif and then cp alert-bg.gif to message-bg.gif and they would all be yellow right?
16:26 CGI986 owen am from Etjiopia i gonna ask some question if possible
16:26 ?
16:26 owen?
16:26 owen schuster: Yes... but then your staff will acclimate to the yellow and stop noticing those just like they stopped noticing the blue!
16:27 CGI986: Ask away, someone will answer if they can
16:27 moodaepo quick question, where/how do I setup permissions say just circ and just cataloging?
16:28 owen anoop: When you have a patron record open (in circulation, for instance), look for the toolbar that includes Edit, change password, etc.
16:28 Click the "More" button and choose "Set Permissions"
16:28 CGI986 am student in university and i gonna work on library automation, so would you like to tell me what prociuder i could follow "owen"
16:29 especially on tutorial (koha)
16:30 anoop so can we make new category with special permission for cir/catalog etc separately?
16:31 never mind found that section
16:32 owen anoop: Permissions are set on a per-user basis, so you can't create a "category" of users with specific permissions.
16:33 schuster basically a cheating way ...  Never could tell there was a blue!
16:33 I could randomly change the color...  Ya know I have nothing else to do...
16:34 anoop owen: thanks
16:34 owen schuster: if you didn't see blue then you're missing that background graphic
16:34 CGI986: Please clarify your question
16:37 schuster but it is a very faint blue so it is there, just doesn't stand out like I need it to.
16:37 More of a grey...
16:44 owen schuster: There isn't an "intranetUserCSS" preference like there is an OpacUserCSS pref, but you can fudge it by using the intranetuserjs preference
16:44 </script>
16:44 <style type="text/css">
16:44 div.message {
16:44 background-image: url(http://zivotdesign.com/downloa[…]ing-warning.gif);
16:45 }
16:45 </style>
16:45 <script type="text/javascript">
16:48 slef anoop: did you find how to do it? I was looking for it yesterday.
16:49 anoop slef: I just found where the settings to select group (permissions) for each patron was...
16:49 slef anoop: per-patron not per-group?
16:49 anoop was that what you were looking for too? or did you want to create a new category with special permissions?
16:50 (11:32:59 AM) owen: anoop: Permissions are set on a per-user basis, so you can't create a "category" of users with specific permissions.
16:50 slef I'm setting permissions for a whole category.
16:50 anoop I don't see why that can't be done though
16:50 slef I'm adding features to Koha to do it.
16:51 wanted to check I'm not wasting time/money/effort :)
16:51 anoop nah I'm in the very beginning stage...supposed to make up mu mind between koh/evergreen soon : )
16:51 s/koh/koha
16:57 schuster @wunder plano, tx
16:57 munin schuster: The current temperature in Crooked Lane, Plano, Texas is 29.8°C (11:57 AM CDT on June 30, 2009). Conditions: Scattered Clouds. Humidity: 59%. Dew Point: 21.0°C. Pressure: 29.84 in 1010.4 hPa (Rising).
16:58 schuster Sorry OWEN!  I'd send you Ice cream but it would melt...
16:58 That works great!
16:58 jdavidb @wunder bethesda, md
16:58 munin jdavidb: The current temperature in Burning Tree Estates, Bethesda, Maryland is 29.1°C (12:55 PM EDT on June 30, 2009). Conditions: Scattered Clouds. Humidity: 43%. Dew Point: 15.0°C. Pressure: 29.26 in 990.7 hPa (Steady).
17:02 anoop @wunder 56001
17:02 munin anoop: The current temperature in MSU Physics Dept, Mankato, Minnesota is 16.4°C (12:00 PM CDT on June 30, 2009). Conditions: Overcast. Humidity: 72%. Dew Point: 11.0°C. Pressure: 29.80 in 1009.0 hPa (Falling).
17:06 schuster Owen - back to my original question branch code showing rather than Description...  That was not resolved right?
17:06 owen I submitted a patch
17:07 schuster Oh...  Thanks!  Now I just need to hurry up the patches!  ;)
17:11 These two small changes today were worth going to Koha!  have I said how much we love Open source!
17:12 What do others use for their editing software?  vi just isn't cutting it off the server...
17:13 Oh WOW !!!  I did just link to the background image...  That would probably freak people out!
17:14 I'm curious though since we are locally hosted what will happen if our internet goes out will it just load without the background?
17:14 owen Yes
17:15 What you can do is replace "background-image: url(http://zivotdesign.com/downloa[…]ing-warning.gif);" with "background: #FFFF99 none" if you just want a solid color
17:15 ...replacing #FFFF99 with whatever color you want.
17:17 schuster Thanks for the option, but I really like what you have provided so far.  I'm adding that to my custom notes.
17:22 jdavidb Howdy, gmcharlt!
17:22 gmcharlt hi jdavidb
17:33 brendan good mid-morning #koha
17:35 Snow_Fox morning
18:30 jdavidb Howdy, pianohacker.
18:38 pianohacker Good mornin'
18:40 "why won't you replay"
18:45 rhcl anybody know offhand what SIP can pull from the ILS database? For example, can you return the name and age of a patron from some app on a client machine?
18:47 pianohacker I'd recommend the SIP standards, but I'm not feeling quite that cruel
18:48 If you're comfortable with Perl, looking at the Koha SIP server in C4/SIP/ might be the easiest way
18:48 rhcl OK, TNX
18:48 atz rhcl: you can get name and sometimes date of birth, not age
18:49 rhcl Ah, yes DOB would be fine.
18:49 atz see Patron Information Response
19:00 owen opac-detail doesn't show my unapproved tags even though I'm logged in?
19:35 pianohacker hi
19:35 chris atz: nice work on the LDAP stuff :)
19:36 heya pianohacker
19:36 atz thx chris
19:38 it's understandable why everybody was griping about the old LDAP implementation....
19:39 chris *nod*
19:39 it was fairly haphazard
19:40 atz the initial implementation targeted the trickier approach, intending to copy data over every time
19:40 chris speaking of it, i saw ron last night, the library they were working is going live today
19:41 afaik, first prison running koha anywhere in the world
19:41 atz ?  
19:41 interesting
19:41 pianohacker Open source software in a jail is mildly ironic, but still cool
19:41 atz when ppl wedged in the simpler bind_as_auth stuff, that was poorly done meaning it would send 50% failed queries AT BEST
19:43 chris ahh
19:44 brendan hey goodmorning chris
19:45 hello atz and pianohacker -- i hope all is well
19:45 chris pianohacker: how far away from fort collins are you?
19:45 heya brendan
19:45 atz greets brendan
19:46 pianohacker chris: Probably a good 4 hours, but there's a library workshop in a nearby town that I go to every year that would be a a good time to visit
19:47 chris ahh, just someone i know (nat torkington) is sitting in a coffee shop there right now
19:47 pianohacker Ah, gotcha. Thought the prison you referenced might be there :)
19:48 chris pianohacker: nope its in ACT in australia
19:53 ok off i go, bb in a while
19:53 pianohacker bye
20:03 Snow_Fox i remember hearing about a project that dumps the overdue emails into a html file
20:03 where can i find information on that?
20:04 pianohacker Snow_Fox: PTFS did that work, it'll hopefully be released to the public within the next few months
20:10 Snow_Fox are there any similar projects in regards to managing over dues?
20:10 dumping stuff to a database etc
20:10 pianohacker Snow_Fox: Not to my knowledge
20:14 owen Snow_Fox: what did you have in mind?
20:16 Snow_Fox essentially trying to figure a way of streamlining the overdues for patrons that do not have emails, i know its a bit of a shocker but there is still a fair bit of kansasns that dont have emails :-P
20:16 owen Oh, you're a Kansan?
20:17 pianohacker I was kind of surprised to find out that you didn't have the same dialup/electric fences problem that HLT had ;)
20:17 owen We're in the same situation here in Ohio, and what we've done in the past is do a custom report to create printed overdues
20:18 The report spits out a CSV file which we upload to a USPS web service which prints and mails postcards
20:18 atz the easiest solution is probably to GIVE your patrons email addresses
20:18 owen And computers? :)
20:18 And internet service? :)
20:18 atz your library should have both of those
20:19 owen But the overdue notices are for people who have forgotten to come back to the library!
20:19 atz ... with their stuff
20:22 pianohacker At our library, at least, avoiding overdue fees by not coming in for a while happens fairly frequently
20:24 We currently deal with that using the overdues report and a custom print-bills-for-lost-items system
20:36 thd What could cause intermittent slow response from Koha when logged in as an authenticated user?
20:36 atz anything in the internet... but most commonly it's costly cronjobs or reports
20:37 thd atz: cronjobs and reports should be running overnight.
20:37 atz no, cronjobs are whenever you schedule them and reports are run per request
20:38 thd of course, so reports are not necessarily scheduled
20:39 atz you might try restarting mysqld with slow-query-log on
20:39 or implementing SNMP monitoring to detect system bottlenecks
20:39 traceroute can tell you if a network node between you and the server is particularly slow
20:40 thd The difficulty is that the problem is intermittent and not predictable
20:40 atz yeah, that suggests and overall monitoring scheme would be most useful
20:40 chris back
20:40 thd It does become bad enough to disconnect logged in users occasionally
20:40 atz otherwise, of course, check your logs to determine what was happening at the time of the failures
20:40 often you can find things like "OOM" (out of memory) errors
20:41 or w/ mysql slow query log, it will tell you when things took too long to complete
20:42 thd It should never take 3-4 minutes to post a cataloguing record or 2 minutes to check out a book.
20:42 atz check netstat for outsider connections (like DOS, hackers, or if there are bunch of undead connections suggesting switch failure)
20:43 traceroute might identify high-latency network notes
20:43 thd undead connections seems like just the right sort of evil
20:43 atz *nodes
20:43 w/ cataloging the bottleneck is sometimes local javascript
20:44 newer browsers have much improved javascript engines
20:44 thd even when saving the record?
20:44 atz yes, when saving the record, there are many (useful) javascript validation checks
20:44 thd oh yes, I had forgotten about those
20:45 pianohacker i.e., change the 'Save' button to 'Validating...'
20:45 or 'Checking...'
20:45 atz pianohacker: that would be a good idea... a little status bar above the status bar
20:45 pianohacker Yup
20:45 atz or the button itself would be good
20:45 thd what validation is done apart from required subfield checks?
20:45 atz that will tell you if it is local or server side problem
20:45 pianohacker Especially since, if it's slow javascript that's tripping things up, the browser won't be responding anyway
20:46 atz thd: whatever the plugins specify... there are dozens
20:46 checking the header for integrity, for example
20:48 thd the plugins are Perl code running on the server merely sending values back to the JavaScript in the client are they not?
20:48 pianohacker thd: The plugins actually generate JavaScript for each field
20:49 atz yeah, they are perl providing javascript... somewhat insane, but "it works"
20:50 thd Do they require communication with the server each time that they are invoked?
20:50 atz no
20:50 they provide it on the first page load
20:50 it's not ajax
20:50 thd The more plugins then the worse the performance would be.
20:50 pianohacker Yup
20:51 chris thats a fairly standard rule, the more something is doing, the longer it will take :-)
20:51 atz potentially, yes.    you can make a streamlined MARC framework so that fewer fields are displayed, if your catalogers would benefit from that.
20:52 schuster When we were having the intermittent problem(server is local) I would run show processlist in mysql.
20:52 thd chris: unfortunately in the case of the Koha MARC editor the plugins are being loaded when the page is first loaded even if the plugins are not used.
20:52 chris yep
20:53 schuster Often it showed me  we had lots of renewing of sessions happening at that time - and we discussed the session log.
20:53 chris i didnt think anyone was complaining about page load tho
20:53 atz the session log needs to die
20:53 chris i thought people were complaining about saving
20:54 i vote we shift the session handling out of the db
20:54 atz writing to a static location in /tmp is dumb... and makes multiple kohas on the same server all write to the same place
20:54 chris use memcached
20:54 atz chris: i'm all for as much caching as we can get
20:54 pianohacker That's actually something I've been wondering about
20:54 chris each server has its own keyspace
20:54 pianohacker If we integrate memcached, will it be a requirement?
20:54 chris and use that for session handling
20:55 you could make it optional, if you want to use the db to handle sessions that could still work
20:55 atz chris: the proper way to do that would be to make a memcached storage driver for CGI::Session
20:55 chris yep
20:55 im 99% sure it already exists
20:55 pianohacker If so, we'd want to make errors complaining about it's absence just a tad more transparent than than those occur when Zebra is missing
20:56 *those that
20:56 chris yep
20:56 all the session handling for stuff.co.nz is in memcached
20:56 we have at times a million or so concurrent sessions
20:56 the db would die in a fire if we tried to make it handle that
20:57 (we use Apache::Session tho)
20:57 http://search.cpan.org/~oinume[…]r-memcached-0.02/
20:58 it could be an optional requirement and set in the config during install i guess
20:58 pianohacker Yup
20:59 atz chris: dunno about that implementation... Oct 2008, no bugs reported, only 11 pass tests, 23 "unknown"
21:01 chris yep, might have to fix it
21:02 thd atz: what are you reporting with 23 test errors?
21:02 chris http://search.cpan.org/~oinume[…]emcached_mysql.pm
21:02 interesting idea
21:03 the perl is only about 150 lines long
21:04 thd chris:  Have you asked the sysadmins at Catalyst about what happens if you have 10 -15 thousand messages in a single maildir on ext3?
21:04 chris yeah they said 'dont'
21:04 thd chris: what did they say to do?
21:05 schuster So with the sessions thing is that something someone is working on or do we need to log that as an enhancement/bug?
21:05 pianohacker re memcached_mysql: Yeah. Seems a bit paranoid, since the worst-case situation is that your users have to log in again
21:06 chris theres was more its gonna make a mess for whatever MUA you use
21:06 not so much the filesystem
21:06 they said "tell him to organise his mail better"
21:06 split by year or something
21:07 thd I usually do but I have not had time to read AUTOCAT for a long time
21:07 chris procmail for the win
21:07 thd and long messages can get messy in a hurry
21:07 chris let it sort it for you
21:08 thd yes I could use procmail more proactively
21:08 chris you can even do crafty stuff so threads arent broken
21:08 but that all gets complicated fast
21:10 thd chris:  Did they recommend any particular file system in any case for use with maildir?
21:11 chris nope
21:13 thd Did they recommend not using maildir?  I assume that they are using maildir themselves.
21:13 chris their only recommendation was try not to get 15k files in one dir
21:13 but that wasnt a worry with the fs
21:13 atz yeah, that's bound to be troublesome unless it's on solid state
21:14 chris *nod*
21:15 wizzyrea maildir makes babies cry
21:16 thd yes, one thing I never understood with perfect clarity was whether that rule applied cumulatively to all subdirectories under a directory. I assume that it must not but am I really correct in my assumption?
21:16 atz thd: no, it's ok if you build a reasonable tree
21:17 thd wizzyrea: I have never used maildir and have used a mail DB in dbmail for years but I am thinking of switching.
21:18 wizzyrea: what does not make babies cry?
21:33 pianohacker brb
21:40 wizzyrea thd: food, naps. :)
21:47 joetho_2 naps. huh.
21:48 2 pm central US time tomorrow, right? I am terrible at counting backwards. I would fail a DUI bigtime.
22:05 schuster hmmm I have it as 4 pm central...
22:16 brendan schuster -- from pianohacker -- http://www.timeanddate.com/wor[…]&sec=0&p1=0&p2=75
22:25 schuster So you are telling me it should be at 2 pm central...  Last time I used this I was a day off!
22:25 chris hehe
22:25 all i know is its 7am my time tomorrow
22:25 schuster The first meeting I just happened to be in the IRC when it started!
22:26 Speaking of off it is time to head home...  I'm wipped!  Must have been all the excitement created by the two things Owen helped me with today.
00:20 pianohacker Good night
00:25 chris hmm rick is being disingenous when calling people disingenous
00:25 pot kettle ... rick
00:31 atz++
03:29 Amit hi chris, brendan
03:29 good morning oha
03:30 good morning #koha
03:30 brendan hello Amit
03:30 how's the weather ?
03:30 windy here
03:30 Amit weather is cool
03:30 here in bangalore
03:30 joetho hot
03:30 brendan :)
03:30 joetho US
03:31 Amit brendan: i m checking my patch comment given by Joe Atzberger
03:31 brendan cool --
03:32 going to look at it myself
03:32 Amit your point brendan
03:33 brendan none yet --
03:33 I'm behind on reading my email
03:35 Amit atz around?
03:37 brendan u there?
03:48 joetho SEKLS?
04:02 test666 SEKLS?
04:07 Amit hi
04:09 Gaurav Hi Amit, Savitra
05:32 ryan ahh, Minnesota, USA.  A professional wrestler as governor, now, after much ado, a comedian as senator...
06:56 Amit_Gupta hi koha
07:02 hdl_laptop hi
07:13 fredericd hi
07:14 hdl_laptop1 hi
07:45 Amit_gupta hi veki
07:46 veki Amit_gupta: hi
08:46 slef hi all.  That email "hypertext link" on koha-devel - why do we have URI in the item record?  Is it used in the templates?
08:49 hi paul_p - I just asked: That email "hypertext link" on koha-devel - why do we have URI in the item record?  Is it used in the templates?
08:49 paul_p: will you be staying around for the developer meeting at 2100 frenchietime?
08:52 Amit: remind me who you are (and maybe I will add it to http://wiki.koha.org/doku.php?id=irc_regulars )
08:52 hi btw :)
08:52 I think no-one else is awake yet :)
08:52 Elwell sssh. keep the noise down.
08:52 slef Elwell: :)
08:53 one of my top microblog "trends" is ":)" according to the analysis tool I used today. #coops and #coop09 are the only bigger ones lately
08:57 fallor hi all
08:59 slef hi fallor
08:59 fallor you wounldn't happen to know about migration scripts on koha3, would you?
08:59 slef a little
09:00 but it usually all rests on what you're migrating *from* and that's a big can of worms
09:00 fallor well, it's koha 2.2
09:00 i can't get either one of the index rebuilding scripts to run
09:00 slef what's the error?
09:00 fallor i.e. rebuild_zebra.pl and rebuild_nozebra.pl
09:01 we originally decided not to use zebra indexing, so zebra is probably not configured properly
09:01 but rebuild_nozebra is a complete mystery
09:01 it seems that it doesn't understand anything about the xml configfile
09:02 slef so you've upgraded to koha 3.0.3 with a copy of your 2.2 database, run through the upgrade?
09:06 fallor well, it's 3.0.0 actually
09:06 but yep, that's the way
09:06 this is what rebuild_zebra.pl says:
09:06 Can't use an undefined value as a HASH reference at ./rebuild_zebra.pl line 88.
09:07 slef line 88 of rebuild_zebra-pl is what in 3.0.0?
09:08 sed -ne 88p rebuild_zebra.pl to display it ;-)
09:08 fallor my $biblioserverdir = C4::Context->zebraconfig('b​iblioserver')->{directory};
09:09 actually the system seems to work otherwise
09:10 i.e. all the patron data and system settings are there
09:10 it's just that the database itself is not searchable
09:10 obviously because there is no index?
09:14 slef probably
09:14 is zebrasrv running?
09:14 pgrep zebrasrv should print its process id number
09:15 assuming you have the BSD procps tools installed
09:15 fallor nope, it's not running
09:15 i think the index has to be created before it starts?
09:15 slef does it?  That doesn't sound right to me
09:16 if zebrasrv isn't running, how can it be given the records to index?
09:16 hdl_laptop1 fallor marc21 unimarc ?
09:16 fallor marc21
09:16 hdl_laptop1 slef: zebrasrv doesnot have to run to be fed with index
09:16 fallor well...
09:17 slef oh wait
09:17 fallor this is what zebrasrv -f /etc/koha/koha-conf.xml says:
09:17 04:15:57-01/07 [warn] XML: I/O
09:17 04:15:57-01/07 [warn] XML: warning :
09:17 04:15:57-01/07 [warn] XML: failed to load external entity "/etc/koha/zebradb/explain-biblios.xml"
09:17 04:15:57-01/07 [warn] XML: I/O
09:17 04:15:57-01/07 [warn] XML: warning :
09:17 04:15:57-01/07 [warn] XML: failed to load external entity "/etc/koha/zebradb/retrieval-info-auth-grs1.xml"
09:17 04:15:57-01/07 [warn] XML: I/O
09:17 04:15:57-01/07 [warn] XML: warning :
09:17 04:15:57-01/07 [warn] XML: failed to load external entity "/etc/koha/zebradb/explain-authorities.xml"
09:17 04:15:57-01/07 [warn] chdir /var/lib/koha/zebradb/biblios [No such file or directory]
09:17 04:15:57-01/07 [log] zebra_start 2.0.32 /etc/koha/zebradb/zebra-biblios.cfg
09:17 04:15:57-01/07 [warn] Cannot open `/etc/koha/zebradb/zebra-biblios.cfg' [No such file or directory]
09:17 04:15:57-01/07 [fatal] Failed to read config `/etc/koha/zebradb/zebra-biblios.cfg'
09:17 hdl_laptop1 wow....
09:17 slef have you set KOHA_CONF?
09:17 hdl_laptop1 It seems you have a problem with your KOHA_CONF
09:18 slef: you beat me
09:18 fallor and there is no zebradb directory under /etc/koha
09:18 slef export KOHA_CONF=/path/to/your/koha-conf.xml
09:18 hdl_laptop1 maybe you should use perl Makefile.PL; make;make install
09:18 fallor err... nope, i haven't set that
09:19 slef hdl_laptop1: you've probably done more upgrades from 2.2 than me. I had very few 2.2 libraries.
09:19 hdl_laptop1: I'll drop you in it^W^W^W^Wleave this one with you
09:19 hdl_laptop1 :D
09:20 fallor ok, now I get a screen full of error messages
09:20 hdl_laptop1 fallor: all the files you are trying to use are in the source directory. But has to be adapted to your configuration.
09:21 fallor: first one ? or pastebin
09:21 fallor <opachtdocs>/usr/share/koha/opa​c/htdocs/opac-tmpl</opachtdocs> isn't a variable assignment, skipping it <intrahtdocs>/usr/share/koha/intrane​t/htdocs/intranet-tmpl</intrahtdocs> isn't a variable assignment, skipping it
09:21 that's not the first one
09:21 but the same seems to repeat for every line in koha-conf.xml
09:22 slef erm, is that koha 3.0's rebuild_zebra.pl or 2.2's?
09:22 (was there one in 2.2?)
09:22 fallor there wasn't one in 2.2
09:22 it didn't have zebra as far as i know
09:22 slef oh I know!
09:22 it's getting your 2.2 C4 modules
09:22 fallor oh?
09:22 slef you need to set PERL5LIB too!
09:23 ordinarily, if you haven't set PERL5LIB the script wouldn't even run
09:23 fallor ok... what should I set it to?
09:23 slef export PERL5LIB=/path/to/folder/above/your/C4
09:23 hdl_laptop1: don't we have $VERSION in C4::Context yet?
09:23 fallor so /usr/share/koha/lib?
09:24 slef where's your koha3 C4?
09:24 it sounds like  /usr/share/koha/lib would contain your koha 2.2
09:24 but I could be wrong
09:25 fallor oh
09:25 something's happening now
09:25 nope, /usr/share/koha is koha3
09:25 slef ok
09:26 fallor koha2 is in /usr/local/koha
09:26 slef right, I'm afk for about 60mins
09:26 hrm, once I've rerecorded my voicemail message
09:26 fallor so now i need to start zebrasrv
09:26 well, except that
09:27 there still is no /etc/koha/zebradb dir
09:27 where are those files?
09:29 found them
09:30 veki_ I installed Koha but, when I type http://1192.168.1.2:8080 I cannot connect with KOha. Apache and My sql are workin. I use opensuse 11
09:30 I installed Koha but, when I type http://192.168.1.2:8080 I cannot connect with KOha. Apache and My sql are workin. I use opensuse 11
09:30 sorry typo in http part
09:32 Amit hi slef
09:36 SelfishMan veki_: Can you connect to koha on http://bitl.in/6z0 ?
09:36 ugh...auto url shortening
09:39 veki_ SelfishMan: I got, what is problem. I fixed /koha/httpd-conf... thanks....
09:39 SelfishMan cool
10:00 Amit hi slef
10:01 slef : http://wiki.koha.org/doku.php?id=irc_regulars
10:08 hi kf
10:13 kf hi Amit
10:15 Amit heya kf
10:15 net problm in my office
10:22 slef Amit: sympathies. I've had one for months :(
10:22 Amit k
10:47 hi gaurav
10:47 Gaurav hi amit
11:28 Amit hi davi
11:29 davi hi Amit
11:29 soul9 has anyone used the matching feature of tools/manage-marc-import.pl?
11:33 gmcharlt soul9: I have, and wrote most of it - what's your question?
11:38 soul9 well, i'm having a hard time making it work...but i don't really know where to start explaining my question :)
11:38 so: i have imported two iso2709 files in the breeding reservoir
11:38 actually, i have imported the same file twice
11:39 imported the bibs from it in the first instance
11:39 then used the matching rule based on isbn
11:39 so it says it applied the new matching rule
11:40 the thing is that the match? column always has 'no_match'
11:41 gmcharlt well, it's possible that there are in fact no matches
11:41 you could try creating a new matching rule
11:41 and applying that one
11:42 also check to make sure that your indexes are up-to-date by running rebuild_zebra.pl
11:43 soul9 ah!
11:43 it uses zebra to match? ok
11:43 gmcharlt yep
11:43 soul9 i have looked into the C4::Matcher, but not that thoroughly
11:44 oki doc', thanks for the pointer

← Previous day | Today | Next day → | Search | Index

koha1