IRC log for #koha, 2006-07-10

All times shown according to UTC.

Time S Nick Message
17:19 kados hey Burgundavia forgive the fact that the zoomsearch is down
17:19 kados Burgundavia: I'm working on an auto-update process :-)
17:19 Burgundavia kados: no worries. I have been working on something else (an edubuntu case study)
17:19 kados sweet
17:19 Burgundavia figured you might like that as well
17:20 Burgundavia
17:20 kados cool
17:21 kados Burgundavia: we could really use some design ideas for how to lay out the OPAC
17:21 Burgundavia I have a half finished email, I just need to create a mockup, because I realized talking about what I think is good is kind of useless
17:23 kados Burgundavia: feel free to submit HTML :-)
17:23 kados Burgundavia: well ...
17:23 Burgundavia heh
17:23 kados XHTML and CSS :-)
17:23 kados seriously
17:23 Burgundavia my brother is the HTML guru in the family
17:23 kados nice
17:23 kados well mockups are welcome too
17:23 Burgundavia luckily he is just as involved in Ubuntu as I am
17:23 kados :-)
17:23 kados excellent
17:23 Burgundavia he also works for a library
17:23 kados cool
18:37 Burgundavia kados: it occurs to me that the koha website doesn't mention where which library conferences you can see Koha at
18:39 kados Burgundavia: yea, we're bad about updating it
18:39 kados Burgundavia: usually the site does
18:49 Burgundavia well, the koha website does a copyright notice fo 2005, even if it does have more recent news on it
18:55 chris luckily since its based in nz, that makes utterly no difference, there is intrinsic copyright here, whether its asserted or not
18:56 kados yea
18:56 kados same in the US
18:57 chris ppl just get in the habit of sticking dates on things I guess :)
18:57 kados have you guys seen the hungarian translation of Koha?
18:57 kados
18:58 chris cool
18:58 kados it's committed to rel_2_2
18:58 kados I was hoping to have the  UN languages in the OPAC demo
18:58 kados before 2.2.6 was released
18:59 chris the finnish translation is nearly done afaik
18:59 kados but WIPO's been dragging their feet getting the translation done
18:59 kados excellent
19:02 kados chris: btw: I completely scripted the upgrade process from rel_2_2 to dev-week
19:02 kados chris: testing the script right now
19:02 kados chris: it takes probably 20 hours or so with NPL's data :-)
19:02 kados also, I discovered circulation isn't working in dev-week
19:03 kados ie, it's not updating zebra
19:03 kados haven't figured out why yet
19:03 chris sorted the utf8 worries
19:03 chris ?
19:03 chris cool about the scripting, once you commit the script ill give it a whirl
19:03 kados to figure that out I'd have to run the script on WIPO's data
19:04 kados but I know the cause of the probs
19:04 chris excellent thats 90% of the way to fixig them
19:04 kados there are several levels of encoding in mysql
19:04 kados if the server things the data is one encoding
19:04 kados and you convert it to another
19:04 kados so if you've got utf-8 data
19:05 kados in a latin1 db
19:05 chris right
19:05 chris you get the double conversion
19:05 kados it messes things up
19:05 kados so what you have to do
19:05 kados is convert the table to type blob
19:05 kados or binary
19:05 chris ahh of course
19:05 kados then, change the table encoding
19:05 kados then change it back to varchar
19:05 kados so it doesn't do any conversions
19:05 chris makes sense
19:05 kados that's written into the script
19:06 chris so for new koha installs its ok, we just set the tables up right
19:06 kados but to be sure if it works I'll have to test it on data with known encodings
19:06 kados yea
19:06 chris and for old ones we do this
19:06 chris cool
19:06 kados yep
19:06 kados hopefully
19:06 kados there's one more tricky bit
19:06 kados which is DBD
19:06 kados hopefully it's doing the right thing
19:06 chris right
19:06 kados but we could be in for a suprise
19:06 kados if it can't communicate in utf-8
19:06 kados without having the utf8 flag set
19:06 kados but at least we know about that :-)
19:07 chris yep
19:07 kados so we can fix it if we need to :-)
19:07 kados[…]ncodingscratchpad
19:07 kados I'm keeping that up to date with things I find
19:07 kados :-)
19:07 chris :)
19:09 kados this bit in mysql:
19:09 kados character_set_connection
19:09 kados hopefully DBD does the right thing when that's set
19:11 kados[…]006-4-786695.html
19:12 kados DBD::mysql will just pass
19:12 kados everything through unaltered. So if you use UTF-8 as connection charset,
19:12 kados you have to encode('utf-8', ...) all queries and parameters, unless you
19:12 kados are sure that they are either plain ASCII or already have the UTF-8 bit
19:12 kados set. And you will get raw UTF-8 strings back, which you have to decode()
19:12 kados explicitely.
19:12 kados However, I notice that on Debian Sarge (on which I did my testing),
19:12 kados libdbd-mysql-perl depends on libmysqlclient12. So there may be a problem
19:12 kados with mixing releases (The server is 4.1, but libmysqlclient12 belongs to
19:12 kados 4.0, which doesn't know about UTF-8).
19:12 kados yikes
19:13 chris hmmm so probably best to get the dbd from cpan
19:13 kados I think that post is from the maintainer
19:19 kados[…]nicode-mysql.html
19:19 kados MySQL has nice UTF-8 data which it passes along to MT’s Perl functions. Why is the data still not marked with the UTF-8 flag? Movable Type uses the Perl modules DBI and DBD::mysql to access the MySQL database. And guess what? They don’t have any Unicode support. In fact, forget marking the UTF-8 flag properly, according to this, DBD::mysql doesn’t even preserve UTF-8  flag when it’s already there.
19:19 kados that's the movable type guy
19:20 kados scary
19:20 kados here's a patch though:
19:26 kados CGI seems to have the same prob :(

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