IRC log for #koha, 2013-01-11

All times shown according to UTC.

Time S Nick Message
00:02 jcamins Bother. I can't figure out how to get it to work.
00:15 rambutan[…]-shopping-trolley
00:16 rambutan left #koha
00:20 barriers joined #koha
00:21 wizzyrea 1
00:25 wizzyrea erm
00:36 NateC joined #koha
01:35 WaylonR joined #koha
01:35 WaylonR hiya all!
01:35 WaylonR jcamins, you about?
02:01 jcamins WaylonR: I am, yes.
02:02 WaylonR jcamins, okay, this search rewrite, how much extra work would it be to include full text (pdf) searching?
02:02 WaylonR or how would i go about that?
02:02 jcamins Hm.
02:03 jcamins It would require a fairly significant amount of work, as you would need to add a filter that parsed out the text from the PDF and inserted it into the records for indexing.
02:04 jcamins If you take a look at Koha/Filter/MARC/ you'll see an example of a search filter.
02:04 eythian that is the sort of thing that, in theory, solr is capable of.
02:04 jcamins eythian: so's Zebra.
02:04 eythian orly
02:04 WaylonR hmmm.
02:04 jcamins eythian: sure, we're just lazy.
02:04 eythian ah, fair enough then
02:05 WaylonR jcamins, ill forward you, works email to me.
02:05 jcamins eythian: actually, that pretty much describes every issue with Zebra.
02:05 jcamins WaylonR: thanks!
02:05 eythian this is true
02:05 WaylonR jcamins, also, what would be the total time for the two tasks you proposed in your proposal?
02:10 jcamins WaylonR: I actually included four items in the end for your mix-and-matching pleasure. 1(f) would take four weeks, 1(c) would take two weeks, 3(c) would take four weeks, and 3(b) would take six weeks.
02:10 jcamins However, the total for all of them would not be 16 weeks.
02:11 WaylonR hmm.. need to reread the item codes..
02:11 jcamins I'd say for all four of them together it would probably take about 9 weeks total.
02:13 jcamins In principle, Zebra can handle that volume of data.
02:13 jcamins However, I have never tried to load 1.25 billion words into a Zebra database.
02:14 WaylonR Better relevancy ranking  + Change the Advanced Search screen to use the new QueryParser + User-specified relevancy ranking  + Static relevancy ranking  = 9weeks? Considering our only media is pdf's, we may not need 3b.
02:15 jcamins About that, yeah.
02:15 jcamins When you initially asked me about it, I had assumed that you didn't need static relevancy ranking, but the e-mail you forwarded suggested that it might be needed.
02:15 WaylonR so, less 3b, what timescale then?
02:15 WaylonR hmm?
02:15 * WaylonR doublechecks?
02:16 jcamins Not the one you sent me just now. The previous one said something like "would it be possible to make Zebra sort so that fulltext came first?"
02:16 jcamins Probably 6-7 weeks.
02:16 WaylonR "entered keywords in the title come first and the rest come after, as a second sorting factor"?
02:16 jcamins No, that's 1(f) and 3(c).
02:17 jcamins Let me find the part in the e-mail that made me think you might need 3(b).
02:17 WaylonR ah.. fulltext first.. hmm...
02:17 jcamins Here: "as a second sorting factor –if possible-  hits will full-text available to come first."
02:18 jcamins I interpreted that as suggesting that you had non-fulltext hits as well.
02:18 WaylonR hmm... was thinking a ccl sort by 773u would work?
02:18 WaylonR as in, 773u true.
02:19 jcamins The problem is that if you want relevancy ranking, you can't sort by anything else.
02:19 WaylonR not the actual value of 773u.
02:19 WaylonR ahh.
02:19 WaylonR wow, this is a really advanced concept. anyway.. full text searching idea.. what you think about the email i sent you?
02:20 jcamins Static relevancy ranking would make it possible to use relevancy ranking _and_ that sort.
02:20 jcamins Also, question: *773*u? Not 856?
02:20 jcamins I don't know what to recommend here, actually.
02:20 jcamins It is absolutely possible to index the PDF fulltext.
02:21 jcamins If Mandumah is rekeying all the text, it is even trivial.
02:21 jcamins Well, "trivial."
02:21 jcamins In the sense of "I would only need to fill two whiteboards with scribbles to figure out how to do it."
02:21 WaylonR haha
02:22 jcamins And in terms of what would be best for your users, I think the option of fulltext search far and away wins.
02:22 WaylonR yeah, they are, it seems. Really impressive stuff.
02:22 WaylonR so the fulltext search would be part of the search module rewrite?
02:23 jcamins However, that is a huge amount of data, and massive records, and I am not certain what Zebra would do with that.
02:23 jcamins It could be, certainly.
02:23 bg joined #koha
02:23 WaylonR hmmm.
02:23 jcamins I didn't spec it out because no one one expressed an interest before, but if you want a spec and quote, I can definitely do that.
02:24 WaylonR well.. can you use the email i forwarded as spec?
02:24 jcamins According to the Zebra documentation, Zebra will happily handle "tens of millions of records, tens of gigabytes of data."
02:24 WaylonR or is that not what you mean? (okay, ive never actually written a formal coding proposal before))
02:25 WaylonR We do actually want Solr, after hearing good things about it.
02:25 WaylonR and the whole online updates thing.
02:25 jcamins I will take the e-mail you forwarded as a basis and write a detailed specification of what I propose to do.
02:26 jcamins Yeah, solr has a lot of promise, and would definitely be useful for you.
02:26 WaylonR Brilliant, thankyou very much. so.. write something that encompasses both emails, and also Solr as an option... need timeline, and cost.
02:27 jcamins Sounds good.
02:27 jcamins It'll probably take me a day or two to address all the possible permutations, but I guess it's just about the weekend for you anyway.
02:28 * WaylonR still needs to be able to specify search params other than keywords too, like branch:, fulltext: peerref: (which is in our current CCL setup)
02:28 bag let's randomly get koha to handle non-marc this weekend!
02:28 bag just kidding thought I could be funny for a sec :P
02:29 jcamins Right. My proposals are anti-regression, so you won't lose any current functionality.
02:29 WaylonR cool.
02:29 bag anti-regression is impossible :P
02:29 WaylonR only problem is, it'll work with only recent koha versions, which means i need to get back to my porting tasks.
02:30 jcamins Right. Unfortunately, I think you're probably stuck with that problem no matter what.
02:30 WaylonR yeah, and no one else can do it.
02:30 bag don't port - submit patches :P
02:30 WaylonR I will be doing both.
02:30 jcamins bag: nonsense. "Anti-regression" is what people picketing software developers' offices with the signs "NO MORE REGRESSIONS!" believe in.
02:30 WaylonR or at least, the ones that will be interesting enough for the koha community.
02:31 bag heh - agreed jcamins
02:31 WaylonR Im not sure that code that picks up a user's set virtualhost (in a borrower_attributes) and redirects the user to the right virtualhost, will be that useful for the community, would it?
02:31 bag maybe I have a different definition of anti-regression
02:32 jcamins You were thinking of it as a noun. :)
02:32 bag WaylonR: I could think of a few people that would like that
02:32 jcamins Like "here is a regression, here is an anti-regression."
02:32 WaylonR hmmm?
02:32 jcamins I think there would definitely be interest in that feature, WaylonR.
02:32 WaylonR okay, well.. i guess ill include it then.
02:32 bag and done correctly it couldn't hurt koha at all
02:33 WaylonR Anyway, im currently working on code to authenticate using any permutation of ip address/virtualhost/referer
02:33 bag there's the wrong word jcamins - "hurt"  my bad
02:34 WaylonR so.. if a admin is stupid enough to want to just use a referer as a authentication factor, they can.
02:34 bag and strip couldn't - that shouldn't be used either
02:34 bag it could help koha!
02:34 bag there we go
02:34 jcamins Exactly!
02:34 WaylonR but if they want to set a account to only authenticate from a ip address (or a ip address range), that will work too.
02:34 jcamins WaylonR: that concept horrifies me, but the patch seems like a very good idea.
02:35 jcamins (the concept of using referer for authentication)
02:36 WaylonR its basically koha's version of iptable rules, except this one matches any allow rule.. unless it matches a overriding deny rule
02:36 WaylonR well.. many places use referer's as the only authentication.
02:38 WaylonR[…]er-td7581917.html for instance.
02:38 jcamins Goodness.
02:39 jcamins Aren't referer headings easy to spoof?
02:39 WaylonR extremely.
02:39 jcamins Do you have a custom Zebra configuration?
02:40 jcamins I guess you must, since you mentioned a peerref: search prefix.
02:40 WaylonR yes, we do.
02:40 jcamins Would you be able to send that to me? That would help me make sure that my proposal covers everything you need.
02:41 WaylonR peer ref is just a bool to say if a biblio describes a peer reviewed/referee'd article.
02:42 WaylonR sure. oh, and we have  773 with journal, journal issue, article and page numbers, too.
02:42 jcamins Thanks.
02:42 WaylonR okay, hmm.. ill just tar up zebradb, sans databases?
02:43 jcamins Yeah, just etc/zebradb
02:49 robertNPBHS joined #koha
02:51 WaylonR sent, jcamins
02:51 jcamins Got it, thanks.
02:52 WaylonR so,  the same config is possible in Solr?
02:52 jcamins Well, the configuration files will be different, but in theory it should be possible to make an exactly equivalent configuration in Solr.
02:59 jcamins Okay, the Zebra customizations I'm seeing are fairly minimal.
02:59 jcamins Awesome ICU rules, though.
03:01 WaylonR jcamins, yeah, some stemming going on i think.
03:01 WaylonR but with the new stemming modules, and solr, they might become unneeded?
03:02 jcamins I don't think so.
03:03 jcamins I mean, you wouldn't need an ICU configuration file like that with Solr, but I think you would need an equivalent file.
03:03 WaylonR right.
03:03 * WaylonR wasn't the one who configured the icu rules.
03:04 jcamins Cool! Looks like Solr supports Light-10.
03:05 WaylonR Light-10?
03:05 jcamins It's a stemming algorithm.
03:05 WaylonR ahh
03:10 WaylonR yeah, that looks like it would make the ICU rules obsolete.
03:11 jcamins Unfortunately, Solr also introduces a whole other set of problems.
03:11 WaylonR hmm?
03:11 jcamins Actually, that's not fair.
03:11 jcamins Solr doesn't introduce problems any more than Zebra does.
03:12 WaylonR hmm.
03:12 jcamins The way we _use_ Solr introduces a whole other set of problems.
03:12 jcamins That's fair.
03:12 WaylonR ahh
03:13 jcamins Optimistically, there is a slight chance that the way Solr indexes are configured might work almost as well as GRS-1.
03:13 jcamins Since using GRS-1 is commonly acknowledged as one of the worst search-related decisions made in Koha's history, that's... not so good.
03:14 jcamins However, I'll address making sure that my proposed work on Solr will be an improvement in the proposal. :)
03:18 WaylonR Solr can't be DOM level?
03:20 jcamins It can, it's just that the current Solr implementation isn't.
03:23 WaylonR ahh
03:45 jcamins Right, time to head to bed I think.
03:45 jcamins Good night, #koha.
04:14 trea left #koha
04:51 mib_vczt1x joined #koha
04:51 mib_vczt1x hey guys
04:51 mib_vczt1x merry new year
04:52 mib_vczt1x any admin available?
04:53 mib_vczt1x i want to contribute some documentation
04:53 mib_vczt1x ive done so before but i dont remember to process
04:54 mib_vczt1x need some directions
05:08 thd-away joined #koha
05:10 thd-away joined #koha
05:13 trendynick joined #koha
05:44 bag evening
05:46 WaqarAzeem joined #koha
05:46 cait joined #koha
05:47 cait morning #koha
06:22 cait @wunder konstanz
06:22 huginn cait: The current temperature in Konstanz, Germany is 2.0°C (7:00 AM CET on January 11, 2013). Conditions: Partly Cloudy. Humidity: 95%. Dew Point: 2.0°C. Pressure: 29.92 in 1013 hPa (Rising).
07:10 BigRig joined #koha
07:29 * magnuse waves
07:29 magnuse @wunder boo
07:29 huginn magnuse: The current temperature in Bodo, Norway is -3.0°C (8:20 AM CET on January 11, 2013). Conditions: Mostly Cloudy. Humidity: 86%. Dew Point: -5.0°C. Windchill: -9.0°C. Pressure: 30.27 in 1025 hPa (Rising).
07:30 magnuse mib_vczt1x: still around?
07:37 reiveune joined #koha
07:38 reiveune hello
07:39 magnuse bonjour reiveune
07:40 reiveune salut magnuse :)
07:49 laurence joined #koha
07:55 jajm hello
07:55 asaurat joined #koha
08:09 lds joined #koha
08:11 gaetan_B joined #koha
08:11 gaetan_B hello
08:13 drojf joined #koha
08:13 drojf good morning #koha
08:13 drojf @wunder berlin, germany
08:13 huginn drojf: The current temperature in Prenzlauer Berg, Berlin, Germany is -0.2°C (8:43 AM CET on January 11, 2013). Conditions: Mostly Cloudy. Humidity: 78%. Dew Point: -4.0°C. Windchill: -4.0°C. Pressure: 29.86 in 1011 hPa (Steady).
08:17 kf joined #koha
08:17 kf druthb
08:17 kf druthb++ # translation manager work :)
08:17 lds joined #koha
08:18 sophie_m joined #koha
08:18 drojf hi kf
08:19 kf hi drojf
08:19 kf and good morning #koha :)
08:19 drojf we got snow :(
08:19 kf oh no
08:19 kf we got clear sky this morning, it's so nice after the foggy days
08:19 drojf i saw the sun for a minute, that was weird. has been a while :D
08:20 kf I can hear you
08:27 magnuse yeah druthb++
08:27 magnuse moahaha
08:28 francharb joined #koha
08:34 kf magnuse: evil laughter? :)
08:37 magnuse yup, because the sun has probably been gone longer here than for you guys :-)
08:40 kf we are weak
08:40 kf we need more sun :)
08:40 magnuse yes please :-)
08:43 magnuse if i import a batch of patrons with a username but no password, then set the passwords manually and then do a new import with possibly updated data but still no passwords, choosing to match on cardnumber and "Overwrite the existing one with this" - will the manually entered passwords be emptied?
08:44 kf yes
08:44 kf in my experience
08:45 kf as long as you have the field on your import file
08:45 kf it might not get overwritten if you don't have the field in your second batch
08:45 kf and a retest might be good to make sure (working with older versions here)
08:45 magnuse yeah, i guess i should test that - will do!
08:53 tajoli joined #koha
08:57 tajoli left #koha
09:26 Oak joined #koha
09:26 gerundio joined #koha
09:28 magnuse i just tested, and the manually created passwords are *kept*, both when the new import does not include a password column, and when the new import contains a password column but it is empty :-)
09:28 * Oak waves
09:28 Oak magnuse
09:41 magnuse Oak
09:41 magnuse :-)
10:08 Oak_ joined #koha
10:19 Oak__ joined #koha
10:26 gerundio hi, good morning
10:28 gerundio I just got to now that the search in Koha using Zebra is returning results with stopwords
10:28 gerundio is there any way to correct this behavior?
10:29 gerundio the only setting I find is QueryRemoveStopwords which only applies when Zebra is off
10:32 magnuse zebra is supposed to not use stopwords, i think
10:32 magnuse jcamins_away will know the details... :-)
10:32 vkm joined #koha
10:33 vkm hello, can i upgrade 3.6.4 to latest stable version directly
10:34 gerundio magnuse, when you say "zebra is supposed to not use stopwords", you mean that the behavior I'm describing shouldn't happen?
10:35 gerundio take this as an example:[…]sort_by=relevance
10:35 magnuse gerundio: i think zebra does not have a concept of stop words
10:35 gerundio ohh
10:36 gerundio looking at the URL of my search query I see the "and" being sent as the GET variable op
10:36 gerundio nevertheless the results show Fernanda and Fernandes with the "and" highlighted
10:37 magnuse yeah, i seem to remember a recent bug report about that
10:37 gerundio let me search bz then :)
10:37 vkm hello, can i upgrade 3.6.4 to latest stable version directly and which is most recommonded stable version at present
10:38 magnuse gerundio: there is bug 6149 - not so recent...
10:38 huginn Bug[…]w_bug.cgi?id=6149 trivial, P5 - low, ---, savitra.sirohi, In Discussion , Operator Highlighted in Search Results
10:39 gerundio they refer that the bug seems to be the highlight and not the results per se
10:40 magnuse gerundio: yup
10:40 gerundio the result I mentioned before with the highlighted "and" does have one search keyword in the subject
10:41 magnuse i think op=and means koha sees "and" as a boolean operator, not a search term
10:42 magnuse vkm: there are two stable versions: 3.8.8 and 3.10.1.
10:43 magnuse vkm: 3.8.8 has had a long time to stabilize, while 3.10.1 is quite fresh and might have unknown bugs still. i would recommend a test upgrade to 3.10.1 and then a thorough check that no problems affect the features you actually use
10:43 magnuse vkm: how did you install koha? packages, tarball, git?
10:45 vkm LIVE CD/DVD
10:47 vkm used live cd by vimal kumar for 3.6.4 on ubuntu 10.04 which is working fine
10:49 vkm please advise how to upgrade it, should i use ubuntu package installation on 12.04 and restore the database of 3.6.4
11:06 JesseM joined #koha
11:06 JesseM left #koha
11:09 thd-away` joined #koha
12:24 jwagner joined #koha
12:33 nengard joined #koha
13:17 oleonard joined #koha
13:22 oleonard Hi #koha
13:23 kf hi oleonard
13:23 kf :)
13:29 samuel joined #koha
13:29 samuel hi everybody
13:34 kf hi samuel
13:34 samuel is it possible to import in koha items which are in a csv file?
13:36 Dyrcona joined #koha
13:37 kf no, not using tools in Koha.
13:38 kf first question?
13:38 wahanui "What are you trying to do?" or "What is the goal?"
13:38 kf hi Dyrcona
13:38 NateC joined #koha
13:38 Dyrcona hi
13:40 samuel kf & wahanui: we have import bibliographic records without any problems. The items are in a separate file (it's a csv file).The goal is to link bibliographic records and items records. I've worked with marc edit, which was a bad idea.
13:41 kf oh
13:41 kf the problem is, when you write the items in the database directly, you will need the biblionumbers
13:41 kf for the imported records
13:41 kf there is no tool currently
13:41 kf either you add the items to your bilbiographic records
13:42 kf or you write a script to add them using koha internal functions I think
13:42 kf hm someone else might have a better idea tho
13:49 samuel ok
13:54 magnuse i think those are the main options
13:55 samuel yes, that's what i think,, maybe with marc-mir or marc-transform, i could solve the problem?
13:56 kf I don't know what those modules can do
13:56 kf marc-mir is not in Koha, but I have heard about it on code4lib I think
13:56 kf what you coudl also do
13:56 kf is create rudimentary marc records with 995 from your items csv
13:56 kf but you will need something to match on
13:57 magnuse druthb might have some tips too :-)
13:57 kf then you can tell the import add items / don't overlay record
13:58 samuel i've just created a marc file. The link can be done with the 001 field but i have to define the good matching rule
13:59 edveal joined #koha
14:01 * druthb perks
14:14 oleonard That's what we love about druthb, always perky!
14:20 kf hi jcamins
14:20 jcamins Hello.
14:22 * druthb beams at oleonard.
14:24 libsysguy joined #koha
14:29 barriers joined #koha
14:33 talljoy joined #koha
14:35 edveal joined #koha
14:44 samuel i'm trying to write a matching rule for my items. The matching point is the field 001 (it's a number), is it a problem if i've not subfield?
14:46 gerundio guys, can you help me find the translation line number for the following problem?
14:46 gerundio
14:46 gerundio there's a \t showing up where there should be just a tab
14:46 jcamins samuel: looks like you can just leave subfields blank.
14:47 gerundio I narrowed it down to line 226 of the file
14:47 jcamins gerundio: try searching for "\\t" in the po file.
14:48 samuel ok, i look the manual at the same time. For Match threshold, is it 100 by default?
14:48 gerundio jcamins, I've tried searching for \t* Field and nothing came up
14:48 gerundio with the addiotional \ it showed
14:48 gerundio jcamins++
14:49 gerundio weird part now is that the source text has \\t too
14:49 jcamins That is weird!
14:49 kf it's a known bug
14:49 gerundio but the EN file shows "\t* Field" in line 226
14:49 kf don't use things like \n and  \t in translatable text :(
14:49 gerundio kf for the win :)
14:49 kf because pootle does double escape them
14:50 kf automagically
14:50 kf I couldn't fix it, so I fixed every \n i found in the source code instead
14:50 kf but that's been a while ago and I didn't think of \t
14:51 kf @query all translation \n
14:51 huginn kf: Bug[…]w_bug.cgi?id=4042 enhancement, P5 - low, ---, bob, NEW , Public OPAC search can fall prey to web crawlers
14:51 huginn kf: Bug[…]w_bug.cgi?id=7808 normal, P5 - low, ---, gmcharlt, NEW , http://debian.koha-community.o[…]amd64/Packages.gz  file not found
14:51 huginn kf: Bug[…]w_bug.cgi?id=8366 major, P5 - low, ---, koha-bugs, NEW , Meaning of "All" in issuing rules
14:51 huginn kf: Bug[…]w_bug.cgi?id=8847 normal, P5 - low, ---, koha-bugs, NEW , Cataloguing doesn't work after installing ru-RU translation
14:51 huginn kf: Bug[…]w_bug.cgi?id=3013 normal, P5 - low, ---, salva, REOPENED , Value builder for 006 and 008 need choices for all format types
14:51 kf hm not waht i was looking for :)
14:53 gerundio jcamins++
14:53 gerundio kf++
14:54 qu-bit joined #koha
14:55 gerundio coolest thing though, the error wasn't reproducible in my Chrome or FF, just in IE
14:56 gerundio but the people who reported it to me sent me screenshots from their Chrome and FF with the issue showing up
14:57 kf gerundio: the bug is somewhere... but I can't find it
14:57 kf it's old so if you want to fix it what I did was
14:57 kf _(translated string) + "does not need to be translated part"
14:57 kf so pootle picks only up the first part, but not the \t
14:58 kf hm if it is the same bug that is
14:58 oleonard What file has the problem gerundio, and what version?
14:59 oleonard I see instances in authorities\ and cataloguing\ which look like they need to be corrected.
15:00 gerundio my issue was in cataloguing/
15:00 oleonard StrAlert += _("\t* Field ")
15:00 kf eew
15:00 gerundio so there's no fix right, just the workaround to replacing by hand?
15:00 kf gerundio: there is a fix - but it will not help you immediately
15:01 jcamins I don't see why you can't use kf's solution.
15:01 kf what you can do is translate without the tab
15:01 kf so it doesn't tab... but there will be nothing weird showing up
15:01 kf and we should fix all those up for the next version ideally
15:02 gerundio I didn't quite get kf's solution
15:03 oleonard StrAlert += "\t* " + _("Field ")
15:03 oleonard Hm, does that need parentheses?
15:03 kf gerundio: for javascript only those things will get into the po file that are marked with _()
15:03 oleonard StrAlert += ("\t* " + _("Field "))
15:03 oleonard ?
15:03 kf oleonard: not totally sure :(
15:03 * oleonard would have to test
15:03 gerundio oleonard, got it now :)
15:03 kf maybe look at some other places? there should be a \n somewhere in addbiblio
15:04 maximep joined #koha
15:23 rambutan joined #koha
15:23 magnuse gah, why does koha-conf.xml have placeholders in it when i create new instances? koha 3.8.8 running off the official packages
15:23 magnuse i see stuff like <xi:include href="__KOHA_CONF_DIR__/zebr​adb/explain-authorities.xml"
15:24 magnuse <host>__ZEBRA_SRU_HOST__</host>
15:25 magnuse oh well, maybe tomorrow...
15:25 chris_n joined #koha
15:25 * magnuse wanders off to make pizza
15:26 Viktor joined #koha
15:31 khall new edition of plugin system bug 7804 ready for testing! Note the updated test plan now has fewer steps!
15:31 huginn Bug[…]w_bug.cgi?id=7804 enhancement, P5 - low, ---, kyle.m.hall, Needs Signoff , Add Koha Plugin System
15:33 Viktor Hi!
15:34 Viktor Little bit of a long shot, but there anyone online that's familiar with the Pootle server?
15:35 Viktor Uploaded a (very nearly) complete .po of the Swedish Opac.
15:36 druthb I manage the Pootle server, Viktor. :
15:36 druthb :)
15:36 libysysguy that would be druthb Viktor
15:36 libysysguy dang I was almost able to announce her
15:36 druthb too slow, libysysguy.
15:36 libysysguy well I saw Swedish and got excited
15:36 Viktor Great druthb
15:37 Viktor Thanks anyhow libysysguy :)
15:37 Viktor druthb Problem is that it is still only 47% translated. The rest seems to be stuck as suggestions.
15:38 druthb You *did* do a nice big upload, Viktor; I see it!
15:38 Viktor I only upload every few years so I kind of forget how to do it right between times :)
15:38 druthb Yes, that's what it does on an upload, when there are differences.  Let me look at something.
15:39 druthb Viktor, there isn't anyone listed in the Swedish project to approve suggestions.  If you're willing to take on that role, I'm willing to grant it to you.
15:40 Viktor druthb Thank you! At the moment I'm the only one working on it so I guess it'll have to be me.
15:40 druthb (I can also make it where your uploaded file overwrites existing suggestions, by the way, which would speed things along for you.)
15:41 druthb Okay, that's set for all Swedish versions.  If you log out and back in, then re-upload your OPAC file, it should overlay.
15:41 druthb (there's a radio button option for that.)
15:42 druthb Viktor++  # I am pleased as pie to have another active translator, in *any* language!
15:43 Viktor druth Hm, seems I lost the rights to view the project at all.
15:43 druthb wup!  Lemme fix.  :)
15:43 Viktor :)
15:43 druthb try now?
15:44 Viktor druthb We are planning to translate the full staff interface too this spring. Just waiting for new funding that is approved but has yet to arrive.
15:45 druthb That is excellent news, Viktor! Please do let me know if you need anything from my end; it's ironic that the Translation Manager only speaks English--but any of the techie bits you need, I'm your girl. :)
15:45 libysysguy I should put Anna to work
15:45 druthb yes, libysysguy.  you should.
15:45 libysysguy she wants the funding too :p
15:46 Viktor Funding is always welcome :)
15:46 druthb Viktor:  I see your update; now only 527 words left in OPAC 3.10.  Great job!
15:47 Viktor druthb Sweet! I'll check those last ones myself. It's Norwegian lines coming from the use of NORMARC.
15:47 vfernandes joined #koha
15:47 vfernandes hi guys
15:48 vfernandes I'm having a big problem...
15:48 Viktor druthb But what does the pref.po files contain?
15:48 Viktor Hi vfernandes
15:48 druthb It's an addon file for the staff client--system preferences descriptions.
15:48 vfernandes can two different persons access to Staff client with the same IP?
15:49 vfernandes the librarians are claiming that the session is always falling and I suspect that is from access of two or more librarian into the staff client with the same IP
15:50 Viktor druthb Thanks - I'll save them for last but take care to get them done.
15:50 gmcharlt @quote random
15:50 huginn gmcharlt: Quote #38: "<owen> Holy Bugzilla activity explosion, Batman!" (added by ricardo at 02:00 PM, October 07, 2009)
15:50 Viktor vfernandes Intersting! I'll listen to any answers since I have no idea
15:51 druthb much good Karma, Viktor--I *love* seeing translations making big progress jumps like this.
15:51 Viktor vfernandes But I can't imagine same IP being a problem. Lots of people should be behind firewalls showing the same IP outwards.
15:52 druthb For 3.10, we currently have five languages listed as "Complete", and I have files from the Turkish team to update, which will complete that, I think.  I would *love* to see ten languages "Complete" for the 3.12 release.
15:52 Viktor druthb We have the most wonderful translators - I'll have to share that good Karma with them :)
15:53 kf vfernandes: i often log in with 2 different accounts in 2 different browsers
15:54 kf vfernandes: very handy for testing things as user and change configuration at the same time - so that should not be the problem
15:54 kf vfernandes: could it be browser settings? are they using IE?
15:54 Viktor druthb So Turkish and Swedish would make it 7 at least? Norwegian is coming along well also I think(?)
15:55 kf druthb++ :)
15:55 kf and German and Swiss German should be in your list too :)
15:55 druthb There are several that are in striking distance of complete--Swiss German, Italian, French, Spanish, Brazilian Portuguese.
15:55 druthb German is complete; Swiss German is *close*
15:55 kf druthb: ah, we won't reach complete for german
15:56 kf ah right
15:56 * kf won't complete translations for 3.4 and 3.6
15:56 kf hm 3.6 maybe, but not 3.4
15:56 druthb I'm just talking about the 3.10/3.12 level; going back to 3.8 might be useful, 3.6 and before, less so.
15:56 kf got you :)
15:57 Viktor druthb Ooh - that would be great. One of the advantages of Koha is being able to switch to patrons native language.
15:57 kf druthb: there are no translations for 3.12 yet :)
15:57 druthb yes! And there are a bunch of koha bugs out there that relate to better handling of more-than-one language at once--true multilingualism--that I'd like to see addressed.
15:58 druthb kf:  When I get back from vacation, I'm going to make the 3.12 project, and see where we stand.  I'd like to have a "String update" every little while, so that you don't have to hurry so much after string freeze.
15:58 kf :)
15:59 kf it will be good for starting tests earlier too
15:59 druthb yes.
16:00 druthb I'm also going to be working on kicking all the .PO files *out* of the main Koha repo, and into its' own.  But I still want it to be *super* easy to add new languages and manage languages in Koha.
16:00 Viktor I've been meaning to ask: How come the english translations isn't complete? It puzzles me
16:00 druthb en_UK, you mean?
16:01 kf different spelling and use of terms
16:01 druthb en_NZ is complete for 3.10, which is good.  We spell things differently, and there are a fair number of other differences.
16:01 Viktor Even NZ translation has a few items left.
16:01 kf yes, same thing - Koha uses en - those are variations of the standard translations
16:02 Viktor Which one is the "native" files?
16:02 kf they are not on pootle :) strings are in the files
16:02 kf no po files
16:02 druthb en_US is the presumptive "native" language.
16:02 kf druthb: it should be en_NZ yo uknow...
16:02 druthb I *know*....
16:02 kf ;)
16:02 druthb Freakin' American arrogance...take over the universe...
16:03 Viktor :)
16:03 Viktor Which is of course why I don't se any en_US files - they are not on Pootle but the stuff we translate from(?)
16:03 druthb Correc
16:03 druthb Correct, even
16:04 Viktor :)
16:04 vfernandes sorry kf i was at one call...
16:05 Viktor Understanding stuff makes me sleep better so thanks for the clarification kf druthb :)
16:05 vfernandes kf I use two different browsers and it work
16:05 druthb you bet, Viktor.  :)  Glad to have you and the Swedish translation team working!
16:06 vfernandes but is strange they are losing their sessions
16:06 thd-away joined #koha
16:06 kf vfernandes: itis :(
16:06 kf I think there is a known problem with IE
16:06 kf where the back button can make you lose your session
16:07 vfernandes they uses IE a lot :/ damn IE... the worst nightmare...
16:07 rambutan left #koha
16:13 reiveune bye
16:13 reiveune left #koha
16:18 Viktor Hm. I am migrating a library this year that is only allowed to use IE. Oh well. Good to know about the bugs at least.
16:19 Viktor druthb Glad to have someone to recieve the files and make the whole thing roll :)
16:19 druthb :)
16:23 Viktor Well - I'm heading out to the computer repair shop to give our laptop some love. Thanks for all the help!
16:23 gaetan_B bye!
16:44 kf have a nice weekend all!
16:45 kf left #koha
17:07 melia joined #koha
17:09 thd-away` joined #koha
17:13 gerundio joined #koha
17:17 asaurat left #koha
17:18 cait joined #koha
17:29 jatara joined #koha
17:38 nengard joined #koha
17:40 Oak joined #koha
17:47 laurence left #koha
17:51 talljoy1 joined #koha
17:53 nengard joined #koha
18:01 oleonard prove t/ fails for me too
18:01 oleonard t/00-merge-conflict-markers.t     (Wstat: 256 Tests: 4971 Failed: 1)
18:01 oleonard t/00-valid-xml.t                  (Wstat: 256 Tests: 141 Failed: 1)
18:03 jcamins The 00-valid-xml is probably a translation, and I'm not sure what's with the merge-conflict-markers.
18:04 jcamins Passes for me.
18:04 jcamins Are you sure you didn't apply a patch?
18:05 jcamins Ooh, it could also be a non-git file that has a merge conflict.
18:06 oleonard Oh could it? I'll check what I have lying around.
18:18 oleonard That must have been it jcamins
18:18 jcamins I've had that happen to me before.
18:19 * oleonard needs to be more careful about leaving things lying around after testing things
18:19 jcamins I go nuts looking for the conflict marker, then realize it's in a file that is no longer git controlled.
18:21 libsysguy joined #koha
18:33 * chris_n takes the plunge and moves his production box to 3.10.x
18:33 jcamins chris_n++
18:34 jcamins chris_n administratorsnack Xanax and hard liquor
18:35 * magnuse would have done that too if it were'nt for bug 9256
18:35 huginn Bug[…]w_bug.cgi?id=9256 critical, P5 - low, ---, koha-bugs, NEW , Search is broken for 3.10 package-installs when marcflavour != marc21
18:37 jcamins I thought there was a patch for that?
18:39 magnuse not that i have seen
18:39 jcamins Oh...
18:40 magnuse i have a vague feeling i know what you are thinkinbg about, but i can't recall it
18:40 jcamins was there a reason I didn't revert it?
18:40 magnuse yeah, you said that would break things for everyone else :-)
18:40 jcamins Oh.
18:40 jcamins Right.
18:41 jcamins That's a good reason.
18:46 rangi[…]3gTzarUJmq1a9qGQg
18:46 rangi 400km more to go today, then home
18:47 rangi thats brooklyn wellington btw, not ny :)
18:47 * rangi is waiting for the kids to get dressed
18:47 jcamins Aww.
18:48 jcamins I would've invited you for dinner, if you were that close!
18:48 rangi bgkriegel++
18:48 rangi hehe
18:50 jcamins And I take dinner invitations very seriously!
18:52 * chris_n dares to hope the upgrade really went as smoothly as it appears to have on the surface
18:57 libsysguy joined #koha
18:59 nengard left #koha
19:09 rambutan joined #koha
19:16 jcamins @karma microsoft
19:16 huginn jcamins: Karma for "microsoft" has been increased 0 times and decreased 3 times for a total karma of -3.
19:16 jcamins Apparently the Evergreen folks dislike Microsoft more than we do.
19:16 gmcharlt having to occassionally battle with Windows executables will do that to a project
19:17 jcamins Heh. Oooh, yeah, I forgot that Evergreen involves Windows executables.
19:21 cait ew
19:22 oleonard @wunder 45701
19:22 huginn oleonard: The current temperature in OHDOT 31-Athens County Garage, Athens, Ohio is 15.6°C (1:51 PM EST on January 11, 2013). Conditions: Mostly Cloudy. Humidity: 81%. Dew Point: 13.0°C. Pressure: 29.96 in 1014 hPa (Rising).
19:22 magnuse ^^ what she said
19:22 oleonard Freakishly nice outside today.
19:22 magnuse oleonard: too good to be true? ;-)
19:23 oleonard Too good to be normal anyway.
19:23 oleonard Okay so how to
19:24 druthb It's cloudy here, but very comfortable
19:24 oleonard handle db test failures?
19:24 oleonard t/db_dependent/00-strict.t                           (Wstat: 512 Tests: 580 Failed: 2)
19:24 oleonard etc
19:25 jcamins Probably that is an issue with missing dependencies.
19:25 cait have a nice weekend everyone :)
19:25 cait left #koha
19:28 oleonard missing perl modules jcamins?
19:28 jcamins oleonard: yeah, in the output of that test it'll say what files missed.
19:28 jcamins *failed
19:28 jcamins And generally what they tried to load and failed to.
19:28 oleonard only reports Data::Pagination missing
19:28 jcamins Hmmm... that's weird, then.
19:29 jcamins What's the error output from the test?
19:31 oleonard The "test summary report" or all the output?
19:31 pastebot "oleonard" at pasted "Test summary report" (29 lines) at
19:32 jcamins That's what you get on clean master?
19:32 oleonard Yes
19:32 jcamins Hrm.
19:33 jcamins If you run each of those individually, it will tell you which tests failed.
19:33 BigRig_ joined #koha
19:34 Callender joined #koha
19:49 oleonard Can't locate HTTPD/Bench/
19:50 jcamins Oh yeah.
19:50 jcamins Now I remember.
19:50 jcamins I reported a bunch of bugs about those various errors.
19:50 jcamins (in the 00-strict test)
19:50 oleonard Things which shouldn't be required, or dependencies which aren't listed?
19:59 nengard joined #koha
20:15 jcamins Sorry, got distracted.
20:15 jcamins Yeah, things that shouldn't be required.
20:19 rambutan joined #koha
20:39 drojf joined #koha
20:57 rambutan @seen oleonard
20:57 huginn rambutan: oleonard was last seen in #koha 1 hour, 6 minutes, and 56 seconds ago: <oleonard> Things which shouldn't be required, or dependencies which aren't listed?
20:58 oleonard I'm right here!
20:58 * oleonard points
20:59 rambutan seems the browser fix didn't
21:00 rambutan seems the problem occurs much more frequently than I had realized; the staff had simply stopped reporting the problem
21:00 oleonard If it happens so often it should be all the more easy to demonstrate
21:01 rambutan when my TS manager (Koha Queen) returns I presume she'll make the Koha syspref change, and we'll see what that does
21:02 rambutan I have one user, my "least technically-inclined" using a Linux computer, and while she uses it for checkout as needed, it's not one of the primary checkout computers
21:02 rambutan but that computer has never demonstrated the problem
21:03 rambutan One of my goals is to swap out all of our checkout computers running Windows to Linux computers
21:15 oleonard Plugins in Koha? Does that mean Koha can run Java now? :P
21:16 jcamins oleonard: if you really wanted to embed a server-side Java host I suppose you could, once bug 7804 gets pushed.
21:16 huginn Bug[…]w_bug.cgi?id=7804 enhancement, P5 - low, ---, kyle.m.hall, Signed Off , Add Koha Plugin System
23:07 BobB joined #koha
23:09 thd-away joined #koha
23:09 nengard joined #koha
23:13 maximep left #koha
23:47 rambutan joined #koha

| Channels | #koha index | Today | | Search | Google Search | Plain-Text | plain, newest first | summary