My Windows Wish List

Christian Stimming stimming at
Tue Aug 8 08:05:44 EDT 2006

Hash: SHA1

Tom Browder schrieb:
> I would like to operate gnucash under Windows, not because I am a
> Windows fan but for ease of transitioning from Quicken to gnucash (I
> have too many years of data for a quick test).  I am currently trying
> to make gnucash work under the cygwin environment with the current
> version of gnucash, and there is hope yet (but slow going).

Cool. We (and I) are very interested to hear about your progress. As
I've explained in the wiki over and
over again, compiling is already possible, but several runtime issues
need to be resolved (guile runtime search paths, for example). Basically
it means it's not too far away from running on windows.

> However, in the long term I wonder if there is any interest in going
> another route:
>   -use a cross-platform solution GUI like Qt, fltk2, or wxWidgets
> (please, not java); or perhaps perl or python GUIs

No, won't work. Or more precisely: Way way too much work. If we'd decide
on the GUI toolkit today, we might very well decide on something
different than gtk, but now the project is stuck on the decision of
1999. If anyone intends to switch to a different GUI toolkit, you'd have
to rewrite roughly 60% of the 300000 lines of code. In part we have to
admit that we allowed much code to blur the line between GUI and program
logic, but for the most part this is just the outcome of having an
extremely large project. Rewriting in a different GUI toolkit is way
inefficient and won't happen.

>   -convert the whole C code base to C++

The programming language itself doesn't change any of our current
problems. This won't happen.

>   -eliminate scheme/guile

This, on the other hand, might in fact happen at some point in the
future. Some of the developers would like this to happen as soon as
possible, others are more indifferent. Note that the program *should*
have a scripting language anyway (because that's much more suitable for
report calculations than pure C code), so the question is rather what to
choose as a *replacement* for the scheme/guile code. Once someone starts
to write the necessary wrapper code from the C functions to the
scripting language (currently done via g-wrap; well-known alternative is
SWIG) this might actually happen within a few months.

>   -use a self-contained Windows package and installer for that version

*This* in turn is probably the most easiest part, once it is possible to
run it on windows. We all know that windows users expect less
requirements for the expense of larger packages, and this is what we
would intend to do anyway. For my windows projects I have been using
installers anyway, e.g. , so that's
what I would have proposed, too.

Version: GnuPG v1.4.2.1 (MingW32)
Comment: Using GnuPG with Mozilla -


More information about the gnucash-devel mailing list