gimage Reader will be crash by Chinese OCR about gimagereader

[manisandro]

I suspect the reason for the crash is that the traineddata might be too new for the version of tesseract the current version of gimagereader uses. If you use the older traineddata files from [1], do you still get the crash?

[1] https://code.google.com/p/tesseract-ocr/downloads/list

[Leroy-X]

you are right, the version of 3.02 is ok, it's work, but the new version is 3.04, hope the giamge will support that, thanks : )

[manisandro]

On Linux, gImageReader uses whichever version of tesseract which is currently available in the distribution - some already have version 3.04 (i.e. Fedora). On Windows, I need to bundle tesseract with the application, so it uses whichever version was the newest one when doing a release. So yes, the Windows build of the next version will use 3.04 (or a newer version if one will be released by then).

[manisandro]

FWIW, I've added some Windows installers for gImageReader built against tesseract-3.04.00 in the releases page until I manage to get a new release out.