IRC log for #koha, 2008-01-08

All times shown according to UTC.

Time	Nick	Message
12:31	lloyd_	mm whre's realy bot gone
13:08	kados	paul: are you around?
13:13	hdl	hi kados
13:13	kados	hey hdl
13:16	owen	Hi guys
13:16	kados	hey owen
13:29	paul	hi back kados & hi owen.
13:30	paul	kados : we have 2 solutions : either a syspref to choose what languages to display (simpler), or let the user choose in a tools/ which translations to generate & let him DL .po files from translate.koha.org (much more complex)
13:32	owen	The best interface for language selection would probably be a set of checkboxes, which you couldn't do with systempreferences the way its set up now
13:33	owen	...Unless there was an interface for installing/uninstalling translations
13:41	kados	paul: I think all of the translations should be pre-generated for every release
13:42	kados	paul: and we just need a syspref to say which ones to display
13:42	kados	or as owen says, a set of checkboxes
14:42	hdl	kados : question about z3950 search and ImportBreeding.
14:43	hdl	Both are using FixEncoding. I think that from a performance point of view this is not the best solution.
14:44	hdl	Is there a reason to do FixEncoding in ImportBreeding rather than having it in z3950 search ?
14:59	gmcharlt	hdl: for some Z39.50 servers, that would be an improvement, yes
14:59	gmcharlt	hdl: but in the general case, you don't always know the encoding of a MARC record returned by a Z39.50 server without checking it first
14:59	hdl	what would be an improvement ?
15:00	gmcharlt	hdl: using YAZ to try to convert a MARC record the UTF8
15:37	kados	gmcharlt: bulkauthimport.pl dies with encoding issues with the sample data I tried ...
15:37	kados	gmcharlt: would FixEncoding be the solution to that?
15:38	gmcharlt	possibly -- I'll work on it
15:38	kados	k
15:57	hdl	kados : is your data with diacritics ?
16:04	kados	hdl: yes
16:05	hdl	I found that there were some serious issues with utf8 and unicode management.
16:05	hdl	For instance :
16:05	hdl	Ã© has 3 representations.
16:06	hdl	And we have to normalize data so that we produce the Ã© that will be recognized in Unicode XML.
16:07	hdl	Even the char_decode as it was could not help solving this for me.
16:07	gmcharlt	hdl: what is the name of the character you're talking about: not showing up in my IRC view?
16:08	hdl	Because it would produce Ã© as \xc3\xa9 (E ACUTE)
16:08	hdl	but \xc3\xa9 is known as hangul character and not E ACUTE.
16:09	hdl	I just sent a new FixEncoding
16:09	hdl	That uses encoding information (little change in z3950servers table)
16:10	hdl	And uses Text::Iconv for the most encoding problems.
16:10	hdl	ISO5426 is assigned by a new char_decode5426
16:48	owen	Hey pate
16:48	pate	hiya owen
16:48	hdl	hi pate
16:48	owen	I saw in the log you popped in the other day. How's it going?
16:48	pate	pretty well
16:48	hdl	Happy New Year
16:49	pate	thanks hdl, and the same to you
16:49	pate	I figured it was time I checked in and saw how koha is doing these days
16:49	pate	it looks like you guys are really getting stuff done with it
16:50	owen	You betcha
16:50	atz	yeah, it's been on high gear lately
17:10	kados	pate: we have our first Armenian translation :-)
17:12	pate	cool!