Gjiten needs dictionary files to function. You will need the kanjidic dictionary file for kanji lookups for KanjiDic and other dictionary files for the word dictionary in edict format. You can get these from ftp://ftp.cc.monash.edu.au. You should download kanjidic.gz and edict.gz. There are some other dictionary files also you might be interested in. (j_places, compdic, etc.) gunzip the dictionary files to the Path To Dictionary Files set in . This should be the same directory where the radkfile and vconj files got installed. [default: /usr/share/gjiten/]. The source tarball contains a script dicfiles/getdics that will download a bunch of dictionary files with wget.
The Debian distribution contains three dictionary packages that you can install via apt-get: enamdict, edict and kanjidicic
As of v2.0 Gjiten uses the GTK2 and GNOME2 libraries. These handle text as unicode internally unlike their previous versions. Gjiten now needs the dictionary and data files in UTF-8 encoding instead of EUC-JP. Please convert the dictionary and data files to UTF-8. You can use iconv for this.
% iconv -f EUC-JP -t UTF-8 dictfile -o dictfile.utf8If the dictionary file is not UTF-8, Gjiten will complain after you add them in the preferences.
The datafiles radkfile.utf8 and vconj.utf8 are needed by Gjiten. Please copy these into the directory of your dictionary files.