Re: RFS: tesseract (updated package), third time lucky, I hope
On Tue, Sep 25, 2007 at 08:37:31PM +0200, Jeffrey Ratcliffe wrote:
> On 25/09/2007, Christoph Haas <haas@debian.org> wrote:
> > I may have missed something so bear with me. Currently there are two
> > packages in "unstable":
> >
> > tesseract-ocr - Command line OCR tool
> > tesseract-ocr-data - Command line OCR tool data
> >
> > >From #434152 I see that the tesseract-ocr* maintainer knows about the
> > new package. But why the name change? Is this the same software?
>
> The package name is still tesseract-ocr. v1 was English-only, hence
> only one data package. v2 is multilingual - and therefore one data
> package per language.
Alright, I was blind. Should have looked closer at debian/control. :)
Why are the include files packages in /usr/include/tesseract? Sounds
like that should rather go into a tesseract-ocr-dev package. Or is it
really needed to run tesseract?
Regarding the language packages: what do they need autoconfig files
(config.guess, config.sub) for?
IMHO the description of the language packages doesn't sound english
enough (although I'm not an englishman either). Perhaps instead of
"tesseract-ocr data for German" you could use something along
"Data files to make tesseract recognize german images"?
Cheers
Christoph
--
Peer review means that you can feel better because someone else
missed the problem, too.
Reply to: