As I read the comments related to the configuration file parsing maybe the
initial focus should be on unifying the parsing code itself into a single
common set of functions or library package first. As I understand it, from the
reading the posts on this subject, many of the modules have their own parsing
code. Eliminating this redundancy would be a good thing for future
improvements and maintainability. And doing this first would be less likely to
introduce unexpected problems.
So, as a 20+ plus year user of fvwm, who has been doing development for more
than 30 years, I would say the first step should be unify the parsing code in
FVWM. I use a fairly simple and straight forward configuration for FVWM. Then
look at ways to improvement the configuration file system itself.
The approach could be the 2.6.x releases be maintenance of "normal" issues.
And then the primary change for 2.7.x releases would be a unified configuration
file parsing system. I would recommend 2.7.x since from the posts unifying the
parsing code appears to be a potential complex undertaking. Then later 2.7.x
releases could support some early desirable configuration file changes. Or
once the 2.7.x parsing code is stable the configuration file improvements could
be introduced in later 2.7.x releases, or 2.8.x could have improved
configuration files.
Don
Dan Espen wrote on 09/26/2016 09:33 AM:
> Ethan Raynor <ethanrayno...@gmail.com> writes:
>
>> On Mon, Sep 26, 2016 at 12:13 AM, Dan Espen <des...@verizon.net> wrote:
>>> Sounds to me like you are not subscribed to fvwm-workers.
>>> If you care about things like the repository, you should subscribe.
>>
>> ok - I will do this. How busy is the list?
>
> Not very, but when the conversation turns to development issues
> only, fvwm-workers@ is sometimes used without fvwm@ being included.
>
>>> Read the various TODO files.
>>
>> The only one I can see is this one -
>> https://github.com/fvwmorg/fvwm/blob/master/TODO.md
>
> The very fact that Thomas is publishing a write up instead of
> diving into changes tells me he's looking for these kinds of
> comments.
>
