On 23 Jun 1999, Rob Browning wrote:
>
> It looks like Linas has mostly finished the fork, and I'm not right in
> the middle of anything else GnuCash related, so I'd like to take this
> opportunity to step back for a few seconds and see what might be the
> next things we'd like to do. Of course this is all dependend on
> manpower and interest, but it'd be nice to have a list of things to do
> in some rough order of importance for us (and others) to look at when
> we've got some time to spare.
>
> My only fairly solid point is that I'm now convinced that using guile
> (even fairly heavily) is the right thing to do, so I'm not going to be
> too excited about sugestions like "(1) Rip out guile.", but other than
> that I'm pretty open. In fact, I'd like to see some of the things I'm
> listing below where speed isn't critical, done in guile rather than C.
> They'll be easier to write (if you know scheme of course), and easier
> to extend/customize.
>
> Here are some things I'd start the list with (note that I'm not
> volunteering for all these):
>
> TODO:
>
> * Fix the glaring problems with the GNOME register. Until that's
> usable, the GNOME version is *dead in the water*. Right now it
> does really dumb things like cycle through all the values in the
> transfer field every time you exit a cell. Further, it
> scrambles the values between transactions -- i.e. it's totally
> useless.
I would like to work on that, once I get familiar with the code. I already
made some modifications/corections to the account creation dialog.
>
> * Go through and remove all the dependencies on perl. Make perl a
> separate dir with a separate makefile that you only get if you
> call configure with something like --with-perl-module, and even
> then, it just gets built and set up as a MakeMaker (or whatever)
> install target that you invoke via "make perl-install". (Matt
> and I will probably handle this one.)
Actually I would suggest something else. The Gnucash engine is stable and
is independent from the rest of the code. It can be separates from the
rest of the Gnucash code and made a separate library. This way, interested
people can take the engine and create interfaces to whatever languages
they like.
I'm not saying to make the engine a separate project, just to add a
configure script for it, make an include directory with the public
include files, and perhaps some documentation. This way, one can just take
the engine directory out of the source code and use it.
The advantage is that the main Gnucash project does not have to maintain
bindings to other scripting languages.
>
> * Get basic stuff like adding accounts, etc. working in the GNOME
> version. This should be done along with testing of the other
> stuff that already exists (do we have account
> renaming/restructuring yet)?
>
> * Add smart account deletion - perhaps even a guile hook for
> deletion so that the action to be taken just before the account
> is removed can be customized. We can provide some standard
> options through the GUI and clueful users can extend it with
> their own.
What actions can be taken at an account deletion? I can only think of
asking the user three times if he wants to delete the account :)
>
> * Translate the reporting/quoting system(s) to guile.
>
> * Many other things, probably some more important than these, that
> I've left out right now.
>
Best Regards,
alex.
----- %< -------------------------------------------- >% ------
The GnuCash / X-Accountant Mailing List
To unsubscribe, send mail to [EMAIL PROTECTED] and
put "unsubscribe gnucash-devel [EMAIL PROTECTED]" in the body
