Re: [gscan2pdf-help] gscan2pdf prerelease pushed to git

[Jeffrey R. <jef...@gm...>]

> 1) OCR works good but cyrillic text is brocken in saved PDFs, see
> attached exemple. Just select some text from the text layer in your
> pdf-viewer, paste it into a utf-compatible text editor (i.e. gedit)
> and compare it with the image. The text seems to be saved as
> escape-sequences instead of utf-8.

The problem here was that the PDF core fonts do not fully support
UTF-8. This therefore relies on whatever extra fonts the user has
installed.

I have just pushed what I hope are some sane defaults, which certainly
work on your example.

You will notice that I have added a font to use for non-ASCII text to
the save PDF dialog.

Things might get trickier with Japanese or something equally far
removed from ASCII.

Please test.

> 2) Minor bug: every time i try to save a PDF file there is a space
> before the file name, see attached screenshot.

This is almost certainly because the default PDF filename uses PDF the
metadata "%a %y-%m-%d"

If you do not specify the author in the metadata, then you will get a
blank, followed by the date. You can change the default PDF filename
in Edit/Preferences.