Time  Nick     Message
08:54 tumer    kados:around?did you get my commits to work?
07:15 chris    yeah i wonder why he's not releasing it
07:12 paul     I tried nothing. Although the sf/CVS version is 2.0 & the cpan one more than 2 years old.
07:11 chris    have we tried sending him a patch?
07:08 chris    yes
07:07 paul     seems that Andy stay silent when requested for new releases
07:07 chris    yeah, we might have to if no one is going to release a new version
07:06 paul     maybe fork a MARC::Record::Improved package ?
07:06 paul     but it seems we have a definite problem with MARC::Record
07:05 chris    right
07:03 paul     (ed summer is cpan maitainer, so mike will have to ask him)
07:02 paul     i have asked mike rylander for marc::file::xml, it should be done quickly
07:01 chris    i dont, joshua might
07:01 hdl      Becuas it can be blocking for us.
07:00 hdl      chris : Do we know when marc::* wil be released ?
07:00 chris    yeah the search.pl is quite straightforward, so you should be able to fix it there
06:59 chris    cool
06:58 hdl      So it can be solved.
06:57 hdl      But I fixed it with processz3950queue.
06:57 hdl      Big encoding problem.
06:57 hdl      I got my first result.
06:57 hdl      Thanks Chris.
06:57 hdl      No.
06:52 chris    hmm does the template you have installed have the Search internet button on it?
06:49 chris    the template looks the same as in dev_week
06:49 chris    hmm that looks right in 2.2
06:46 hdl      sorry searchresults.tmpl
06:45 hdl      search.tmpl should be the one.
06:45 chris    2 seconds
06:45 chris    looks like a template has been missed then
06:44 hdl      no.
06:44 chris    http://koha.kohadev.katipo.co.nz/cgi-bin/koha/z3950/search.pl?oldbiblionumber=&title=test
06:44 chris    does it look like this when you do a z3950 search?
06:42 chris    give me a couple of seconds to get it installed
06:41 chris    lemme take a look
06:41 hdl      Shoul I try to ask for a do_search ?
06:41 chris    maybe missing a template
06:41 hdl      So it just loads servers and then wait.
06:40 hdl      So it just loads servers and the wait.
06:40 chris    hmm
06:40 hdl      But there are no ops on first load.
06:40 chris    ahh hes committed there already cool
06:40 hdl      Joshua committed on rel_2_2
06:40 chris    without needing any refreshes
06:39 chris    it should just keep loading
06:39 hdl      needs some refresh.
06:39 chris    it works in dev_week
06:39 hdl      It doesnot work for me.
06:39 chris    hi bruno
06:39 hdl      chris : did z3950 search work when committed ?
06:38 btoumi   hi chris
06:37 chris    ahh cool
06:36 paul     helene passini
06:36 paul     yep.
06:36 chris    who is hpassini? another person from SAN-OP?
06:24 paul     hello chris (the only non-french native atm)
06:24 chris    im around now
05:46 osmoze   hello too
05:36 toins    hello world
05:33 paul     lot of  ppl on the chanel today,... even if it's quite
05:25 btoumi   chris sorry
05:25 btoumi   chrtis are u around?
05:20 btoumi   hi all
01:48 chris    and see if it answers
01:48 chris    (assuming ur on a *nix box)
01:47 chris    try telnet 63.109.217.181 2200
01:47 chris    could be
01:46 bureado  Or maybe it's just timing out
01:36 bureado  Maybe I made an sloppy ZOOM install, can anyone give me any pointers regarding this issue?
01:36 bureado  ZOOM error 10007 "Timeout" from diag-set 'ZOOM'9546/9 : Processing author=Dewey at CENTRAL TEXAS COLLEGE 63.109.217.181:2200 unicorn MARC21 (2 forks)
01:34 rhariram have gone thru perl unicode documentation once in the past. will do it again
01:30 kados    rhariram: check the perl unicode documentation
01:29 kados    rhariram: that is a very complicated question :-)
01:22 chris    is a good place to look
01:22 chris    http://www.kohadocs.org/usersguide/ch04s02.html
01:20 bureado  I'll check the docs. Thanks!
01:20 bureado  I see. That's the missing part indeed.
01:20 chris    which is much easier to use
01:20 chris    but we have just gotten a much nicer version working, that will be in 2.2.7 in the near future
01:19 chris    the long answer is its kinda tricky in 2.2.6 you need to start up the z3950 daemon (all documented at kohadocs.org and in the mailing lists)
01:18 chris    is the short answer :)
01:18 chris    yes
01:14 bureado  (in Koha, of course :))
01:14 bureado  Greetings. I have a question. Is it possible to "retrieve" information from a public Z3950 server (I'm using the LOC one) just by providing the ISBN or Author name? Is it that "automagical"?
01:09 rhariram kados: where  in perl could I start looking? hows the utf8 flag defined?
01:05 dewey    it has been said that marc_biblio is where you can find the relationship between marc_subfield_table.bibid and biblioitems.biblionumber in the 2.2 series of Koha
01:05 chris    marc_biblio?
01:05 chris    marc_biblio is where you can find the relationship between marc_subfield_table.bibid and biblioitems.biblionumber in the 2.2 series of Koha
01:05 chris    damn
01:04 chris    marc_marc_biblio is where you can find the relationship between marc_subfield_table.bibid and biblioitems.biblionumber in the 2.2 series of Koha
01:04 chris    it might have worked :)
01:04 chris    no, marc_biblio etc
01:04 chris    hehe if you had done
01:04 dewey    no marc_biblio is where you can find the relationship between marc_subfield_table.bibid and biblioitems.biblionumber in the 2.2 series of Koha
01:04 chris    no marc_biblio?
01:04 dewey    chris: I forgot marc_biblio
01:04 chris    dewey: forget marc_biblio
01:03 chris    just tell him to forget
01:03 kados    heh
01:03 dewey    marc_biblio is the old rel_2_2 table
01:03 kados    marc_biblio
01:03 dewey    i already had it that way, kados.
01:03 kados    dewey: no marc_biblio is where you can find the relationship between marc_subfield_table.bibid and biblioitems.biblionumber in the 2.2 series of Koha
01:03 kados    chris: is dewey broken?
01:03 kados    ahh
01:02 dewey    marc_biblio is the old rel_2_2 table
01:02 kados    marc_biblio
01:02 kados    good dewey
01:02 dewey    OK, kados.
01:02 kados    dewey: no marc_biblio is where you can find the relationship between marc_subfield_table.bibid and biblioitems.biblionumber in the 2.2 series of Koha
01:02 dewey    ...but marc_biblio is the old rel_2_2 table...
01:02 kados    dewey: no, marc_biblio is where you can find the relationship between marc_subfield_table.bibid and biblioitems.biblionumber in the 2.2 series of Koha
01:02 dewey    ...but marc_biblio is the old rel_2_2 table...
01:02 kados    dewey: no, marc_biblio is where you can find the relationship between marc_subfield_table.bibid and biblioitems.biblionumber in the 2.2 series of Koha
01:01 dewey    somebody said marc_biblio was the old rel_2_2 table
01:01 kados    marc_biblio
01:00 rhariram i just saw that the marc_subfield_table.bibid and biblioitems.biblionumber are not the same. So where does the info on the connection between the recs in the two tables get stored ?
00:52 kados    that could be really tricky :-)
00:52 kados    so next thing I'd check is the perl I guess
00:52 rhariram kados: yes. column type are the same. biblioitems.publishercode & marc_subfield_table.subfieldvalue are varchar
00:33 kados    rhariram: is the column type the same for marc_subfield_table as the column you'r ehaving trouble with?
00:31 rhariram kados: was just looking at the koha.mysql file. Table definitions look ok.
00:13 rhariram ok
00:12 rhariram that right only in the koha tables and not marc_subfield_table
00:12 kados    that would be burried deep in the code somewhere and proabbly hard to spot/test
00:12 kados    or it could be a perl problem, something might be setting or unsetting the utf8 flag to perl and causing it to double-encode the field
00:11 kados    it could be a problem with the table definition
00:11 kados    not the marc_subfield_table?
00:11 kados    if I recall, it's only happening in the koha tables, right?
00:09 rhariram ok. no problem. could you give pointers on how to pursue it ?
00:08 kados    rhariram: got a number of clients needing stuff done before I do my volunteer stuff ;-)
00:07 kados    rhariram: sorry, I haven't had a chance to pursue it
00:06 rhariram any clue on bug 1230
00:05 kados    rhariram: what's up?
00:05 kados    hi rhariram
00:04 rhariram hi kados
23:51 kados    ahh, my fault
23:51 kados    hmmm
23:51 kados    database unavailable
23:51 chris    makes sense
23:50 kados    yea
23:50 chris    so it should be $server->{host} insteadd eh?
23:50 kados    hehe, now I get a 500 error :-)
23:50 chris    ahhh
23:49 kados    nothing to connect to :-)
23:49 kados    server dataNELSONVILLE PUBLIC LIBRARY9999
23:49 kados    warn ("server data",$server->{name}, $server->{port});
23:49 kados    got it
23:47 kados    yea :-)
23:47 kados    and if we were smart about it it would still work in lynx
23:47 chris    but first we have to figure out why its not connecting :)
23:47 chris    right
23:47 kados    which would be much faster to load
23:46 kados    so we could have asynchronous results I mean
23:46 kados    and some javascript could keep the results in the right places
23:46 kados    we could use that technique for the federated search in koha
23:46 kados    if we wanted
23:46 chris    yep
23:46 kados    just keep printing out to the browser
23:45 chris    it has the link and username and password
23:45 kados    because I've done that myself
23:45 kados    but I know how it works
23:45 chris    if you scroll back
23:45 kados    ahh, no I haven't
23:45 chris    i meant the z9350 search
23:45 kados    ahh, yea I've seen the OPAC
23:45 chris    the way the results come back
23:45 kados    the template, or what?
23:44 chris    its pretty cool
23:44 kados    no, didn't
23:44 kados    yea, for 'no error' :-)
23:44 chris    did you see tumers koha?
23:44 chris    but we still arent connecting for some reason eh
23:44 chris    cool
23:44 kados    chris: if you cvs up I re-did the tabs so they're pretty
23:43 kados    $oConnection[$s]->connect($server->{name}, $server->{port}) || warn ("something went wrong: ".$oConnection[$s]->errmsg());
23:43 kados    that's this line:
23:43 kados    something went wrong: No error at search.pl line 130.
22:54 chris    good night
22:54 chris    yeah i know mine is firewalls, i cant test further than that until i change that
22:54 tumer    gnight for now
22:54 tumer    so pls chek firewals etc and our settings
22:53 chris    right
22:53 tumer    line by line same code
22:52 tumer    only thing different is after results i call for xml in head and for marc record in that version
22:52 chris    very cool
22:51 chris    sweet that works just fine
22:49 tumer    try http://library.neu.edu.tr:8100/cgi-bin/koha/cataloguing/addbiblio.pl with id:0123456 pass:arman if u like
22:49 chris    right so its not that then, after work ill do some more testing, its probably somethng simple
22:45 tumer    chris i added the user and password to mine and it still works
22:35 tumer    let me know so i can do something if necessary
22:35 kados    thanks tumer!
22:35 kados    right now I should get some dinner
22:35 kados    well, I can try to troubleshoot some more later
22:33 tumer    very weird!
22:33 kados    huh ... it doesn't connect
22:31 kados    chris: good thinking
22:31 kados    hmm, it is inside that for loop
22:31 chris    you hopefully will see it connect in the zebra logs?
22:30 tumer    put a warn to $numresults
22:30 chris    if you do it against just 1 server, ur one
22:30 kados    I'm not so sure
22:30 chris    so it gets past that, ie its doing the search
22:30 tumer    why dont you try your own zebra server ?
22:29 kados    nothing
22:29 chris    well thats what zoom says
22:29 kados    and I don't get it for loc strangely
22:28 kados    but the error I'm getting is a connection prob
22:28 chris    $oResult[$z] = $oConnection[$z]->search_pqf($query) || warn ("somthing went wrong: " . $oConnection[$s]->errmsg());
22:28 chris    into
22:28 chris    $oResult[$z] = $oConnection[$z]->search_pqf($query);
22:28 chris    could do
22:28 kados    tumer: that warn is uncommented already
22:27 chris    and then when it tries to search it dies
22:27 tumer    line 153 will give you more messages on connection
22:27 chris    but u get past that
22:27 chris    interesting thats what fails for me (cos of the firewall)
22:27 kados    no warning for the 'something went wrong'
22:26 kados    connection 0: event 10 (end)205.184.70.3 at search.pl line 158.0 205.184.70.3 error @attr 1=4 @attr 4=1   "mark": Connect failed (10000) PASADENA P.L.:210
22:25 chris    see if we can get some info
22:25 chris    $oConnection[$s]->connect($server->{name}, $server->{port}) || warn ("somthing went wrong: " . $oConnection[$s]->errmsg());
22:25 chris    try making it
22:25 chris    that line
22:25 chris    $oConnection[$s]->connect($server->{name}, $server->{port})
22:25 chris    see about line 138
22:24 kados    but no results returned
22:24 kados    don't get an erro for loc
22:24 kados    yep
22:24 chris    have you tried commenting out the user and password bits?
22:24 chris    right
22:24 kados    I get Connect failed (10000) PASADENA P.L.:210 for all of them
22:23 kados    weird
22:23 kados    yea, doing that now
22:22 chris    try uncommenting the warns in there and watch the logs
22:22 kados    huh
22:22 chris    nope kados, just stuck some warns in
22:21 tumer    currently i am searching,loc,ohio, MIT Oxford all working
22:21 kados    did you change anything chris?
22:21 chris    ill test later but it looks to me like it should work if it could connect
22:21 tumer    LIBRARY.MIT.EDU:9909 which is MIT is also working
22:20 kados    66.213.78.76:9999 is where it's actually at
22:20 kados    9999
22:20 chris    hmm lemme see
22:20 kados    chris: 443 maybe?
22:19 kados    chris: I could start it up on a port you aren't firewalled on
22:19 chris    no ports will
22:19 kados    chris: 66.213.78.76:9900 work for you?
22:19 kados    s I'd think one of them would work
22:19 tumer    chris:ROCK.LIB.ASU.EDU:210
22:19 kados    but I have 7 servers defined
22:18 kados    loc is very slow though
22:18 kados    well my test box isn't firewalled
22:18 kados    heh
22:18 chris    so i cant really test im afraid
22:18 chris    so its a firewall blocking that port outbound on my dev box
22:17 kados    right
22:17 chris    i can connect to it from home
22:17 kados    olc1.ohiolink.edu 210
22:17 chris    probably a local firewall
22:17 kados    wtf?
22:16 chris    that explains that at least
22:16 chris    telnet: Unable to connect to remote host: Connection refused
22:16 chris    Trying 140.147.249.38...
22:16 chris     telnet z3950.loc.gov 7090
22:16 chris    ah ha
22:16 tumer    in fact they all work with that but just incase for those UNIMARC
22:16 kados    not yet for me
22:16 tumer    and loc works
22:15 tumer    i have USMARC hardcoded
22:15 kados    I wonder if it should be MARC21 instead
22:15 tumer    replace with USMARC
22:15 kados    would that be USMARC?
22:15 kados    server->syntax?
22:13 tumer    chris:comment out line 131 and 132 for user and password i do not have those
22:11 chris    so now i have to figure out why :)
22:11 chris    cant connect at search.pl line 138.
22:11 chris     $oConnection[$s]->connect($server->{name}, $server->{port}) || warn "cant connect";
22:11 chris    i did this
22:10 chris    ah ha
22:09 chris    im trying to figure out why
22:09 chris    yep it only does that once
22:09 tumer    uncomment line 153 and you will see all zoom movements
22:08 kados    just that page
22:08 chris    ill put some warns in and see where its getting to, its only doing the first server
22:08 kados    Still 7 servers to search
22:08 kados    Nothing found
22:08 kados    Bibliographic Records Found in 205.184.70.3   Title  	Author  	ISBN  	   	
22:08 kados    IE for me acts the same
22:07 tumer    it works for IE for me but i cannot test it on dev_week
22:06 chris    right
22:06 tumer    chris:it just does a print $template->output after each server result
22:06 kados    nope, same deal
22:04 kados    hmmm
22:04 tumer    this works on IE but i am not sure on firefox
22:03 chris    ah ha
22:03 tumer    no ajax
22:02 tumer    it just adds to the page
22:02 kados    does it use ajax?
22:02 kados    no?
22:02 kados    and manual reload doesn't turn up anything
22:02 tumer    the page dows not reload
22:02 chris    yeah that page isnt reloading
22:02 kados    no page reload
22:02 kados    I just get a message 'still 7 servers to search'
22:01 kados    it was me mucking around again ;-)
22:01 kados    may have spoken too soon
22:01 chris    what was going on?
22:01 kados    wait
22:01 chris    heh
22:01 kados    sweet!
22:01 kados    hang on, it's working now
22:00 kados    will do
22:00 kados    voyager
22:00 kados    z3950.loc.gov:7090
22:00 tumer    kados:can you please uncomment the warns in search pl
22:00 kados    z3950.loc.gov:7090/
22:00 chris    lemme add some more
22:00 chris    hmm mine doesnt do that
22:00 kados    but the page never reloads
21:59 kados    and it still gives the 'Still ?? servers to search' message
21:59 kados    nothing interresting in the log
21:59 kados    tumer: and some pretty good servers defined
21:59 kados    tumer: nothing found for 'tom sawyer'
21:58 kados    musta been something I was mucking with :-)
21:58 kados    yea, fixed now
21:58 chris    theres not even an $i on line 73 for me :)
21:57 kados    maybe I'll synch up again
21:57 chris    im not
21:57 kados    lobal symbol "$i" requires explicit package name at /koha/intranet/modules/C4/Breeding.pm line 73.
21:57 tumer    i only added one line
21:57 chris    weird
21:57 kados    hmmm, I'm getting a ton of errors in Breeding.pm
21:55 tumer    try me on library.neu.edu.tr:9010
21:53 chris    whats a known working z3950 server?
21:52 tumer    kados: on Windows i cannot fork so the results appear one after the other you may change that
21:52 chris    yep
21:52 kados    ahh, see it now
21:52 kados    chris: did you commit?
21:51 tumer    thanks chris
21:51 kados    it's nice and simple for small libraries
21:51 kados    rel_2_2 will have a long life AFAIK
21:51 chris    seems to be working now
21:51 kados    rel_2_2 is the 'stable' 2.2 series (next version in this series is 2.2.7)
21:51 kados    tumer: there are four now:
21:50 tumer    right
21:50 chris    Global symbol "$breedingid" requires explicit package name at /usr/local/koha/intranet/modules/C4/Breeding.pm line 125
21:50 tumer    sorry i am really confuesed with all this releases
21:49 tumer    no alteration to db
21:49 kados    tumer: originally I meant rel_2_2 but dev_week will do ;-)
21:49 kados    tumer: ahh, to dev_week? do I need to alter the db at all?
21:46 tumer    yes please
21:46 tumer    i have just committed a new z3950 search to dev_week
21:46 chris    want me to test out your commits?
21:45 chris    sure do
21:45 tumer    do you have a dev_week running?
21:45 chris    good thanks and you?
21:45 tumer    how are you?
21:44 chris    sup tumer?
21:44 chris    yep
21:43 tumer    chris:around?
17:24 thd      kados: are you there?
16:45 hdl      But I think taht if it would be easier, more and more persons would come to Koha.
16:45 kados    yep
16:45 hdl      This is what we are paid for.
16:45 hdl      migrating to Koha is quite a big burden for ppl.
16:44 hdl      And maybe Koha could get some inspiration from it.
16:44 kados    very nice features
16:44 kados    yep
16:44 hdl      It present some cool features in integrating oai and word and any doc type through plugins.
16:44 kados    hdl: not carefully
16:43 hdl      did you ?
16:43 hdl      kados : I dont think you tried greenstone.
16:42 hdl      hi _paul
16:42 kados    hdl: I agree
16:35 hdl      But it needs some development planning.
16:34 hdl      Maybe buildQuery should also be built in this aim.
16:32 hdl      This qould allow us to magically implement some new search engines
16:32 hdl      Opensearch, rpn, ccl, and so on.
16:32 hdl      Opensearch, z3950, ccl, and so on.
16:31 hdl      IMO, we should have getRecords be splitted into parts so taht queries could be trated by types.
16:30 hdl      It really nedds som more dev.
16:30 hdl      So I gues this shouldnot be part of Koha rel_3_0.
16:29 kados    too bad for me, I'm likely to be too busy in Nov/Dec for any development :(
16:29 hdl      (on XML->XSLT->XHTML)
16:29 hdl      (I agree)
16:29 kados    if you can spend some time understanding how it shouhld work)
16:28 kados    hdl: (IMO, using XML -> XSLT -> XHTML is the best approach
16:28 kados    hdl: but we may want a more flexible approach such as conversion to MODS or something
16:28 kados    hdl: ie, what is mapped to koha fields in the template
16:28 kados    hdl: the framework would decide how to display the MARC record when retrieved
16:27 kados    hdl: the opensearch does not work at all except on an uncommitted version of Search.pm that I need to clean up
16:27 kados    hdl: it is highly experimental
16:24 hdl      But the fact is that some information is still missing.
16:23 hdl      With "position" (primary, 2dary, blank) and "type" (zed or opensearch)
16:22 hdl      kados : but I thought that servers could all get into z3950servers table.
16:22 hdl      The fact is that each ZOOM connection need some options to be set, such as cclfile for instance.
16:20 hdl      So this is HIGHLY experimental at present time is it not ?
16:20 kados    hdl: honestly, I didn't think much about it ... except to allow federated search in the getRecords sub
16:19 kados    hdl: we would need a new framework for each federated source I think
16:19 kados    hdl: the results display
16:18 hdl      You told me it was not finihsed yet. But what is to be done ?
16:17 hdl      kados : Did your primary and secondary server loops worked for you ?
14:29 tumer    hdl: you can see a sample at http://library.neu.edu.tr/cgi-bin/koha/union-search.pl if it helps i'll send you the code
14:27 kados    but it's not completely finished
14:27 kados    dev_week supports federated searching
14:27 tumer    hdl:i have a different code than kados's so its kados to answer
14:19 hdl      So I was wondering how to get them work.
14:19 hdl      But it seems it doesnot work yet on rel_3_0.
14:19 hdl      I thought this facility was provided by kados on devweek
14:18 hdl      kados, tumer : my question was to adapt kados intranet search facility to search multiple catalogs would it be z3950 servers or opensearch servers.
14:04 paul     maybe you should re-write your question for kados & tumer...
14:04 tumer    hdl:sorry I meant z3950/search.pl not isbnsearch.pl (sorry paul as well)
13:55 kados    tumer++
13:55 paul     huray for tumer !
13:55 kados    as it's definitely the most asked question on koha list
13:55 tumer     and OK i will do it on 2_2
13:55 kados    I think we would get many 'points' from the community on this
13:54 kados    tumer: so that we can release 2.2.7 with improved Z3950 client
13:54 tumer    that should be very easy to port to 2_2
13:54 kados    tumer: I am asking if you could merge the HEAD code to rel_2_2
13:54 kados    tumer: but in HEAD, it's not
13:54 tumer    in head there is no more Z3950 deamons
13:54 kados    tumer: in rel_2_2, processz3950queue is still used
13:54 kados    tumer: ok, I'll start over :-)
13:54 kados    tumer: to eliminate processz3950queue for rel_2_2
13:53 tumer    kados and paul i am getiing confused!
13:53 kados    tumer: I was asking if you could take your new stuff in HEAD and put it into rel_2_2
13:53 paul     is it modified in koha-zebra ???
13:53 kados    tumer: because many libraries will still use it
13:52 paul     ???
13:52 kados    tumer: I'm speaking of rel_2_2
13:52 tumer    you do not need procesz3950 anymore
13:52 paul     (isbnsearch : toins modified it recently to use zoom as everywhere else)
13:52 kados    in fact, if we did this, I think we could release 2.2.7
13:52 paul     it's processz3950queue
13:52 paul     tumer : it's not isbnsearch.pl
13:52 kados    for new users
13:52 kados    it's the most troublesome part of koha 2.2 I think
13:51 tumer    no problem; my isbnsearch.pl should just work with little modifiation
13:50 kados    I mean the Z39.50 client ...
13:50 kados    do you think you could port the new stuff you did with ZOOM to rel_2_2?
13:50 kados    I'm getting really sick of the old Z3950 daemon
13:49 tumer    kados:shoot!
13:49 tumer     a very similar code runs in our OPAC in Advanced serach/Union catalogue
13:49 kados    tumer: I've got a question for you
13:48 tumer    check the code catalogue/isbnsearch.pl in head
13:47 hdl      and I am nneither getting connected to outerservers nor getting reults from them with getRecords.
13:46 hdl      the problem is that I am trying to allow outer catalog search as kados allowed in koha-2.4
13:45 tumer    hdl:whats the problem?
13:38 hdl      ?
13:38 hdl      How does it work.
13:38 hdl      And what about opensearch servers ?
13:38 hdl      It should be the same as with processz3950queue is it not ?
13:37 hdl      ?
13:37 hdl      I get a loop. but how do I use ZOOM for outer catalogue and present ppl with results.
13:37 tumer    paul yes very similar
13:36 hdl      kados : sorry to interrupt.
13:36 paul     - patron1 renew through SMS => you get EUR0.02
13:36 paul     - I send a SMS to patron1 => cost EUR0.10
13:36 tumer    i managed to get them to accept for me to send that free as a contribution to thr library
13:36 paul     so, it's something like :
13:36 paul     (because for this, you pay, right ?)
13:35 tumer    i have that
13:35 paul     ah, ok, I understand. Don't you have a feature to send SMS when a return date is near ?
13:35 tumer    i get a share of that money
13:35 tumer    the cost to them
13:35 tumer    customers sending SMS
13:34 tumer    nope
13:34 paul     you will get money to send SMS ???
13:34 tumer    and are wiling to share profits
13:34 kados    cool!
13:34 paul     ???
13:34 tumer    they like to have SMS traffic
13:34 tumer    yes! i will even earn money on it
13:33 paul     to get access to operators gateway, you just ask them (& pay some money I think ;-) )
13:33 paul     wow ! it's clear & really really easy !
13:32 kados    neat
13:32 tumer    paul:is it clear?
13:32 tumer    koha automatcally sends overdue messages with SMS as well
13:31 tumer    answers back with new renewed date or Return the book its reserved and so on
13:31 tumer    RN meaning renew
13:30 tumer    it just parses a line like RN username password barcode
13:30 kados    tumer: woot!
13:30 tumer    so  i have written a module (currently only for renewals)
13:29 tumer    we can send and receive SMS messages by POST or GET methods
13:29 tumer    we have access to an operators gateway
13:28 paul     could you explain how you do this (technically)
13:28 tumer    we will officially launch next week
13:28 paul     great !
13:28 tumer    we have the ssytem running
13:28 tumer    kados:Koha is now mobile!
13:20 kados    that's what I'm going to suggest I think for WIPO
13:20 kados    and use the single byte equivilents
13:20 kados    one solution is to simply avoid all multi-byte characters for mysql
13:19 tumer    equivalent (aÌ?)a is even better
13:17 tumer    map (aÌ?) a will do it
13:16 tumer    you canu use map function in sort.chr file and it will return 'GaÌ?bor for gabor
13:15 paul     + we will have the same problem when searching for patron gabor in koha-zebra !
13:14 kados    tumer: this is a legacy client
13:14 kados    :(
13:14 tumer    kados why are you still using marc_word and not zebra?
13:14 kados    only returns 'gabor'
13:14 kados    select distinct word from marc_word where word like 'gabor%' collate utf8_polish_ci;
13:13 tumer    hi kados
13:13 kados    hey tumer
13:13 tumer    hi all
13:13 paul     hello tumer
13:13 tumer    nope only in zebra!
13:12 dewey    hi tumer is still strugling
13:12 hdl      hi tumer
13:12 hdl      maybe tumer can help
13:12 kados    takes a very long time
13:12 kados    ie, select distinct word from marc_word where word like 'ga%' collate utf8_polish_ci;
13:12 kados    so it's not much help
13:12 kados    and it seems that the 'collate' sql is quite slow
13:11 paul     ok, i'm afraid i won't be helpfull in this matter...
13:11 kados    I was just wondering if you encountered it before
13:10 kados    so I don't want to be too hasty
13:10 kados    well, I have just discovered this problem
13:10 paul     strange behaviour. It means there is no complete utf8 support in mySQL, isn't it ?
13:10 kados    (I have ... it's WIPO ;-))
13:09 paul     right.
13:09 kados    because what if you have a catalog with five languages ?
13:09 kados    but it's not ideal
13:09 kados    perhaps
13:08 paul     maybe utf8_slovak or utf8_polish could help here ?
13:07 kados    I have tried utf8_general
13:07 kados    (note, this only applies to multibyte chars
13:07 paul     there are many utf8_xxxx, can't one of them solve this ?
13:07 kados    so it seems
13:07 paul     ??? really ?
13:07 kados    so for instance, a search on 'gabor' does not return results for 'GaÌ?bor'
13:06 paul     (note that, as all of them are unimarc & french, i've char encoding latin1,)
13:06 kados    mysql doesn't support collation of single-byte characters like 'a' to multibyte characters like 'aÌ?'
13:05 kados    it seems we have one remaining UTF-8 problem in Koha 2.2 series
13:05 kados    so you have not yet encountered doublebyte characters :-)
13:05 kados    ahh
12:47 paul     in switzerland, the libraries are 1 public & 1 specialized
12:46 paul     - 1 english unimarc library in nederlands
12:45 paul     - 2 french unimarc library in switzerland
12:45 paul     nope, I don't have a french marc21 library, but I have :
12:45 paul     hi kados.
12:42 kados    paul: do you have a french marc21 library now? or one in switzerland?
12:42 kados    paul: are you present?