Importing PDB file with Chinese (GB18030) enc

  • Daniel Kirkwood

    Daniel Kirkwood - 2010-06-16


    I've got the latest public release of karatasi-java and I'm having trouble importing PDB files that have Simplified Chinese (GB18030) encoding.

    In the karatasi-java settings I have set the PDB encoding to GB18030. If I set this encoding, then karatasi-java can read the database name successfully and display it correctly (in Chinese characters) in the database list on the left-hand side of the program window. However when I "list" the database, the Chinese characters do not appear in the database columns, instead all I get is little question-mark symbols on a diamond-shape icon.

    If I try a different encoding, then not even the Database Name is correct. So I know that this encoding setting is having some effect, it just seems like it's only being applied to the database name and not the card fields.

    I hope someone can help me, thanks very much.

    Thank you also for doing so much work on the PDB conversion feature! I have so many SuperMemo for Palm cards and being able to import them into Karatasi is going to be great :)


  • Christa Runge

    Christa Runge - 2010-06-17

    Hi Daniel,

    are you sure that all needed fonts are installed on your PC?
    Which OS are you using?
    Try to setup the synch and transmit the database to the iPhone, so you can see how your cards are displayed at the iPhone.



  • Daniel Kirkwood

    Daniel Kirkwood - 2010-06-17

    Hi Christa,

    I sometimes run my machine (Mac OS 10.6) in Chinese so I definitely have suitable fonts.

    I tried uploading the converted database to my iPhone and I still see the question-mark symbols where the Chinese text should be, except for the database names. If I use a different (non-Chinese) encoding and upload it to my iPhone, then the database names are also wrong, just like on my Mac in Karatasi-java.

  • Christa Runge

    Christa Runge - 2010-06-19

    Hi Daniel,

    in a Palm database, the database name is managed by the Palm OS, while the records are managed by the application.
    In the Palm SuperMemo application, these records contain the cards and categories.

    It might well be that the character encoding of the records and the database name is not the same; especially in Asian languages with their rich character set. Note that this is pure guesswork at the moment - unfortunately we do not have a proper specification of the Palm SuperMemo database format.

    Did you by chance find an encoding which converts the cards and categories correctly, but not the database name?
    This could be an indicator for two different encodings. Candidates which I would try first are standard encodings like UTF-8, UTF-16, and UTF-32. But again, this is pure guesswork.
    If you like we can implement a karatasi Java version with selection boxes for two different encodings.

    Where did you buy your Palm? China, or Europe/America?
    Is it a native Chinese, or did you install additional software to get the Chinese character support? If yes: which software?



Log in to post a comment.