[Koha] OPAC searches fail with ICU indexing enabled

[Andreas Roussos]

Dear list,

We're running Koha 3.20.04 on Ubuntu 14.04, and recently enabled
ICU indexing as per the instructions on the wiki
(https://wiki.koha-community.org/wiki/Correcting_Search_of_Arabic_records)

Most searches work fine, but queries for certain Greek characters in OPAC
(for example "[χ.ό.]"), return the following message:

Error:
:8: parser error : Input is not proper UTF-8, indicate encoding !
Bytes: 0xCF 0x3C 0x2F 0x74 ος Τιμόθεον Α΄-προς
Τιμόθεον Β΄-προς Τίτον-π ^

If I use the command-line zebra client to perform the same search,
I get 335 hits:

$ yaz-client -c /etc/koha/zebradb/ccl.properties
unix:/var/run/koha/imp/bibliosocket
Connecting...OK.
Sent initrequest.
Connection accepted by v3 target.
ID     : 81
Name   : Zebra Information Server/GFS/YAZ
Version: 4.2.30 98864b44c654645bc16b2c54f822dc2e45a93031
Options: search present delSet triggerResourceCtrl scan sort
extendedServices namedResultSets
Elapsed: 0.000743
Z> base biblios
Z> f [χ.ό.]
Sent searchRequest.
Received SearchResponse.
Search was a success.
Number of hits: 335, setno 1
SearchResult-1: term=χο cnt=335
records returned: 0
Elapsed: 0.014453

So, it looks as if Zebra can actually perform the search but somehow
the results cannot be displayed in OPAC.

Does anyone have any clues as to why this is happening?

Kind regards,
Andreas