Perhaps you have not declared your input format properly. Here are the essential PDIC instructions from Kur't README file (the whole file comes with the wadoku.zip download):
> Use dokuwa.txt for german-japanese
> and und wadoku.txt for japanese-german.
> The files are already ordered, so the sorting in will be fast.
> Start pdic and choose in the menu point
>
> Tools - 辞書の変換
> the following options:
> 転送元形式:
> 一行テキスト形式
>
> 転送先形式：
> PDIC 形式
>
> Load the files with 参照, e.g. wadoku.txt for the 一行 ,
> and choose wadoku.dic for the pdic file to be created.
> In the bottom menu choose
>
> 付け加える（区切り文字付き)
>
> --------------------
>
> Then move on to 詳細 and choose
>
> 一行テキスト形式の区切:
> ///
>
> i.e. with a space at the left and the right of the three slashes.
>
> In the left menu
> 訳 / 用列部について
> choose only the first point
> 訳 / 用列を区切りして取り込む
>
> and at the bottom choose
> 区切り文字
> /
> i.e. a space to the left and the right of the slash.
>
> In the right menu
> 出力に関して
> choose the first
> 格納しきれない場合のみ圧縮
> and the third point
> 用列を圧縮
>
> --------------------------------------------
>
> In the submenu
> 登録項目
> choose point 1,4 and 5.
> OK this menu as well as 詳細。
>
> --------------------------------------------
>
> OK once more.
It Still doesn't work!
Martin Boehme reported problem in ZPDVIEW with very short dictionary files. In particular, the microdict sample included with dico.pl does not seem to work correctly on the Zaurus. Solution: Make a longer dictionary file, or just use EDICT at once, not microdict. Thanks a lot, Martin!
... I applied for membership (to the mailing list) and noticed the source
code for ZPDVIEW is available here. Let's have a look at the source
code and see if that helps solve the problem, I thought. Well, looking
through the search code, I noticed the PDIC file is split into blocks
of entries. And, though I haven't yet understood how the search code
works exactly, it seemed to me that it might have a problem if there's
only one block in the file.
How can I add new entries to my Zaurus dictionaries?
You can do this on your PC with PDIC:
- Write the entries you want to add to a file in a file format that can be read by PDIC, for example CSV.
- Rename the dictionary you want to add to, for example call it target.dic.
- Convert the CSV file to PDIC format and call the new file target.dic. PDIC will prompt you whether you want to overwrite the old file or merge the dictionaries. (See also FAQ: Why does PDIC generate improper data format?)
- Rename the merged file according to Zaurus conventions.
Alternatively, you could use PDIC to convert a dictionary a format of your choice, edit the dictionary file and convert it back to PDIC format. Editing a dictionary in PDIC format is not feasible.
Why are Japanese entries wrong?
There was a problem with a previous version of the conversion script which confused the Japanese (2-byte) coding. I fixed the bug, so donwload the new version of the script.
Why do I get odd translations?
Ambiguities arise from using a dictionary the "wrong" way round, i.e. in the English-Japanese or German-Japanese direction for Edict and WadokuJT, respectively.
Why are some very easy words missing in the WadokuJT?
In my own script, I through out a lot of examples in order to save memory. It turns out some of the regular expressions in the script are too restrictive and throw out essential entries. Kurt's compilation does not have this problem (and the resulting dictionary file is more than twice as large). Consider my own compilation a slim version which you would want to supplement with other dictionaries, such as EDICT.
Does the Zaurus display exactly the same data as my PC?
Not always.
When you look at PDIC files using the PDIC program on your PC, it has 5 fields: Attibute, Engl, Pronunciation, Label, Translation.
ZPDVIEW will display only two of those fiels, Engl and Translation, and ignore the other ones. Therefore, we ususally lump the pronunciation into the translation field.
In the case of Eijiro, the original data as you can buy it on CD uses only these two fields. Therefore, Eijiro's data should be displayed by ZPDVIEW on the Zaurus exactly as by PDIC on the PC. (Note: Our own compilation of Eijiro slightly differs from the one on the original CD: We added pronunciation to some words, while the original generally is very spartanic with pronunciation. Moreover, our compilation is a bit outdated.)
How do I get the very latest version of each dictionary?
Do it yourself!
The internet dictionaries grow constantly, and all of our compilations are likely to be outdated - we won't bother compiling data very often. You must get the data from the source to have the latest version. In the case of Eijiro, just order their CD to get it precompiled and ready to use. For everything else, you must compile it yourself as explained.