cuneiform team mailing list archive

["Yury V. Zaytsev" <yury@xxxxxxxxxx>]

Hi, Jussi!

On Wed, 2009-05-06 at 23:12 +0300, Jussi Pakkanen wrote:
> 
> Since we managed to slay the last bug preventing Cuneiform from being
> run in many different processes at the same time, I think it is time
> to have a new release.

Sounds good to me! We need to remember the "Release Early, Release
Often" mantra in order to keep the project alive and kicking.

> If you have issues you would like to see done for the next release,
> please speak up.

Well, I would love to check whether I could compile the current codebase
with ImageMagick/Unix + MinGW (the ImageMagick + MSVC works fine for me)
and rework the readme.txt OR add a separate file like compile.txt or
whatever to describe the Windows build process with different tools and
helper scripts in greater detail. No one will really appreciate this
GUI-cmake-editing-way, especially when you have to care about the
correct sets of libraries to link with (thanks god we don't have this
problem on Linux)...

My last attempt got silently dismissed, so I would like to hear from
you, whether I should plan to put more efforts into this kind of
documentation/build automation or just forget about it.

Still, I am getting more and more busy with the forthcoming exams, so
probably I will have to put this off until the next release + 1.

> I suggest that we move all text files to UTF-8 encoding before the
> release. It works on all platforms, even Windows. See (the end of)
> https://bugs.launchpad.net/cuneiform-linux/+bug/369246 for more
> information about that.

Well, actually I have already raised this concern awhile ago:

https://lists.launchpad.net/cuneiform/msg00262.html

and I find it a very good idea, since otherwise the code is a pain to
work with on anything else apart from Windows (I mean otherwise I can
actually *read* the comments).

But I think that we need Dmitry's opinion on this matter. The potential
problems I can think of are:

1) Incompatibility with Visual Studio 6 (not that I care about this,
but... this needs to be confirmed or rejected)

2) I had an impression that Cyrillic characters are sometimes used in
the source code, not only in the comments and licence agreement.

Therefore it might be dangerous to perform a bulk iconv actions on the
source tree, because *.dat files would remain unconverted and nobody
knows how it will work out.

I need you and/or Dmitry to chime in.

Cheers,
 
-- 
Sincerely yours,
Yury V. Zaytsev