Closed GoogleCodeExporter closed 9 years ago
This is patch does not make nonsense - language files (traineddata & cube
files) are in separated repository.
Original comment by zde...@gmail.com
on 24 Sep 2014 at 11:25
I just put back a fraction of code that was there before it got deleted
with the intention to help others that may have been using automated
scripts to build tesseract from its repository.
As I said, I had a script to build a binary package from svn version (which
was deprecated) and it stopped to work when I started to track the git
version, Also, if you ever cared to read the documentation about how to
install tesseract from svn version "INSTALL.SVN", you would see that it
used to recommend specifically what is stated below (it is actually a
fragment of its contents):
"...
So, the steps for making Tesseract are:
$ ./autogen.sh
$ ./configure
$ make
$ sudo make install
$ sudo make install-langs
$ make training
$ sudo make training-install
'sudo make install-langs' or 'sudo make install LANGS=' will install all
available language data files in tessdata directory.
If you want to install just few of them than run:
$ sudo make install-langs LANGS="eng ara deu"
It will install only English, Arabic and German language datafiles (if
they are present in tessdata directory)
..."
I fully comprehend that this is not what is recommended anymore on
"INSTALL.GIT".
That's all. So much for "nonsense" allegations.
Original comment by cbarros....@gmail.com
on 24 Sep 2014 at 9:48
There is no "INSTALL.SVN"[1]. Quoting outdated instruction is not relevant.
That fragment of code was presented in svn repository because it worked in past.
There is git repository and implementing your patch in git repository is
nonsense - it does not work.
[1] https://code.google.com/p/tesseract-ocr/source/browse/
Original comment by zde...@gmail.com
on 25 Sep 2014 at 8:39
Original issue reported on code.google.com by
cbarros....@gmail.com
on 24 Sep 2014 at 12:50Attachments: