TESSDATA_PREFIX issue?

[Nim Sudoh]

Hi,

On Arch Linux, installed Tesseract and tesseract-data-eng..


While I can use Tesseract from the command line sucessfully.

If I try and use gimagereader or gscan2pdf I get the same error:

Error opening data file /usr/bin/tessdata/eng.traineddata
Please make sure the TESSDATA_PREFIX environment variable is set to the parent directory of your "tessdata" directory.
Failed loading language 'eng'
Tesseract couldn't load any languages!
Could not initialize tesseract.

I have successfully set the environment variable as:

export TESSDATA_PREFIX=/usr/share/tessdata

(as shown by the command line success)

It may not be a Tesseract issue, but it seems odd both gimagereader or gscan2pdf  look in the wrong place for the eng.trainedata.

Can anybody give me any pointers?

Thanks!