Hi all,
Jabok Library currently tries to become a part of the Czech Libary
Gateway for Visually Impaired which uses Z39.50 protocol to access
library data. However, when testing the results from Jabok Z39.50
server, we have encountered encoding problems. These have been confirmed
by further tests conducted via Tellico (http://tellico-project.org/)
which we use as an Evergreen-independent tool capable of serving as a
Z39.50 client.
From Evergreen documentation (although pertaining to an older release;
Jabok itself currently uses Evergreen 2.6.2) available from
http://docs.evergreen-ils.org/2.1/html/Z3950serversupport.html it seems
that the server uses marc-8 encoding for the output. But even when we
have this marc-8 encoding set up when downloading records to Tellico,
the records seems to come up with many incorrect characters (those which
use diacritics), not to mention that when words with diacritics are used
for searching, no results show up at all.
Would it be enough to change the default charsets mentioned in steps 3
to 4 at http://docs.evergreen-ils.org/2.1/html/Z3950serversupport.html
to utf-8? Or are some other tweaks necessary? (I am rather afraid that
the second option is closer to reality because even when marc-8 encoding
is used, the results are incorrect – of course, supposing that Tellico
itself is not buggy ;-).)
Maybe Dan would know more about this issue - I remember that when we
were discussing Zotero encoding issues back in April, there was a note
in the reply about fixing a similar problem with the Z39.50 output
(please see
https://www.mail-archive.com/open-ils-general@list.georgialibraries.org/msg10874.html)?
(When I tried to download records from Laurentian University Library –
using the host name and other data from
http://staff.library.mun.ca/staff/toolbox/z3950hosts.htm -, it worked
okay when encoding was set to utf-8 :-).)
Thank you in advance for any hints!
Linda
