[Koha-devel] Problems with diacritics

[Joe Atzberger]

Constantine --

I don't seem to have any problem with the steps you enumerate.

For example, in my version 3.01.00.006 system step three produces these
results:
http://atz.dev.kohalibrary.com/cgi-bin/koha/opac-search.pl?q=%CE%9C%CE%BF%CF%8D%CF%83%CE%B7%CF%82+

Perhaps you can update your codebase and try again?  Please also mention
whether you (or your browser) represents "Μούσης" with any combining
characters (as opposed to their normalized unicode representations).

In any case, please also verify the utf8-ness of your setup as described
here:
http://wiki.koha.org/doku.php?id=encodingscratchpad

Thank you for providing a concrete set of steps with data so that we at
least have something to test!

--Joe Atzberger,
LibLime

On Mon, Oct 6, 2008 at 5:45 PM, Constantine Tsardounis <
costas.magnuse at gmail.com> wrote:

>
> $ uname -a
> Linux asymmetry.*. 2.6.26-1-686 #1 SMP Wed Sep 10 16:46:13 UTC 2008 i686
> GNU/Linux (Debian - Lenny)
> US keyboard - I did not type the diacritics (I think the problem is with
> the
> character "ϊ" of the eu2.marc file)
>
> I uploaded, staged and imported a file with records that contain amongst
> other MANY titles the word "Ευρωαπαϊκή" (it means European) as a title of a
> book. After that, any catalog search that contains a word that was in that
> file
> gives me the error page:
>
> Error:  Can't call method "as_usmarc" on an undefined value at
> /usr/share/koha/lib/C4/Search.pm line 2133.
>
> I undid the import but nothing happened.
>
> The BIG problem is that it contained many many widely used words like
> "economics" or "environment" or "statistics". So now that I search for
> these
> words I get the same error page.
>
> Could you please provide me with a work around in order to remove this
> staged
> file please?
>
> ########
> Way to reproduce:
> ########
> Download: http://costis.name/eu2.marc : this file contains various entries
> amongst them
> there is a book written by McCann, and in a later line another book that
> contains the string "Ευρωπαϊκ" in its title, written by "Μούσης".
>
> 1. I have a clean installation of koha and import eu2.marc
> 2. I search for McCann, and I get the result I should
> 3. I search for Μούσης and I get the error (Error:  Can't call method
> "as_usmarc" on an undefined value at /usr/share/koha/lib/C4/Search.pm line
> 2133.)
> 4. I search for McCann, and I get the same Error!... Isn't it weird that I
> did
> not get it in step 2?
>
> How could i resolve this?
>
> ########
> Unsuccessful solution:
> ########
> In the database, I removed from the import_batches table the entries of
> the import_batch_id 13,14,15 (these correspond to the files that contained
> the
> entries with the diacritics and are like the eu2.marc file), but the
> problem
> persists. What else should I do ?
> --
> View this message in context:
> http://www.nabble.com/Problems-with-diacritics-tp19547458p19847757.html
> Sent from the Koha - Dev mailing list archive at Nabble.com.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: </pipermail/koha-devel/attachments/20081121/a289a147/attachment-0003.htm>