Hello,
I have a simple question. I often meet in mrxvt's code some parts where
there is maybe not an error but a goofiness. For instance some actions are
made several times unecessary (eg. a test on a buffer to check whether an
action has to be made whereas there is no possibility that this buffer has
been modified since last time we checked!), etc. Other times, some variables
are set several times in a row to the same value (but this may be "hidden"
in the code), or it is set and immediately set to some other value, etc.
I am not always 100% sure, and I would need to make more extensive tests.
Some other times this is obvious. As an example yesterday (or 2 days ago, I
am not sure), I saw a loop which define variables, always at same value (so
independant to the incremental variable). The logic would be to simply get
this variable definition out of the loop because this is unecessary variable
creation (I guess each loop would create a different variable, unless the
compiler can "see" it and optimize, but even though, I find it not good
programming).
This was in rxvt_scr_expose (screen.c):
for (i = ...; ...; i++)
{
register int j = rc[PART_BEG].col;
register int k = rc[PART_END].col - rc[PART_BEG].col + 1;
MEMSET (&(PVTS(r, page)->drawn_text[i][j], 0, k);
}
here I would obviously get j and k def out of the loop.
The question is then:
when I see "details" like this, can I change it and commit just for this? Or
do you want only big and meaningful commits? Because I write them on my
notebook but I fear to forget them all the same (and as I run tests, I often
reset my local code to original state).
Bye.
Jehan
-------------------------------------------------------------------------
SF.Net email is sponsored by:
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services
for just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
Materm-devel mailing list
Materm-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/materm-devel
mrxvt home page: http://materm.sourceforge.net
