On Wed, Aug 28, 2002 at 11:07:29PM -0500, Dave Trollope wrote:
> Hi Domonik
>
> I was returning to the discussion we had at the start of the month where
> you wanted to replace the if/else chain with a bsearch. We had a
> detailed discussion about changing the data structures etc and I see you
> have modified the macros along some of the lines we discussed, but the
> bsearch problem hasn't been implemented yet and I had another perhaps
> simpler idea.
>
> Instead of trying to get the bitfield address of the field(s) to change
> and storing that in a bsearchable array, perhaps a simpler solution
> would be this - and this is flexible to complex styles as well as simple
> styles.
>
> For each style, create a simple static function to set the data
> appropriately, then in the array of names, use a function pointer. E.G.
>
> Instead of:
>
> if(StrEquals(token,"Sticky))
> { S_SET_STICKY(MCF(...),on)
> S_SET_STICKY(MCM...,1)
> S_SET ....
> }else if(StrEquals.....
>
> You could do this:
>
> static void f_Set_Sticky(int on) {
> S_SET_STICKY(MCF(...),on)
> S_SET_STICKY(MCM...),1)
> S_SET ....
> }
>
> struct setstyle { char *name; void (*func)(int on); } = {
> ..... { "Sticky", f_Set_Sticky }, ...
> }
>
>
> ss = bsearch(setstyle,token)
> if(ss) ss->func(on)
> else { /* do other style checking */ }
>
> This retains the whole data model, and avoids the whole bit mask address
> problem while also being flexible to more complex processing for some
> styles. This does create a lot of functions, but if they are static,
> they should be minimum overhead.
>
> What do you think?
I've thought about this approach too. It has one big weakness:
it doesn't reduce the amount of code. Instead, it even creates
more code. I'd grudgingly take the approach that requires
generating the sources - if I could think of a way to get names on
the individual bits that are displayed by a debugger.
Bye
Dominik ^_^ ^_^
--
Dominik Vogt, [EMAIL PROTECTED]
Reply-To: [EMAIL PROTECTED]
--
Visit the official FVWM web page at <URL:http://www.fvwm.org/>.
To unsubscribe from the list, send "unsubscribe fvwm-workers" in the
body of a message to [EMAIL PROTECTED]
To report problems, send mail to [EMAIL PROTECTED]
