Re: [Nano-devel] Update

[David Lawrence Ramsey]

Just so everyone knows: There's some weirdness in the do_replace_loop()
code caused by some oversights on my part when I was porting over a few
bits of DB's changes to it.  I've gone ahead and fixed it in CVS in the
interest of not having hard-to-read code, and because the fixes are
minor; I've run all the test cases and canceled some replaces to make
sure that the fixes didn't break anything.  I just had to port a few
other minor bits of DB's changes, plus one of mine.

The changes are as follows:

* remove the second declaration and definition of match_len, which
shouldn't be there (DB)

* move the declaration of the char* copy from the beginning of the
function to the beginning of the if clause used when a replace is made,
since that's the only place where it's set and used (DB)

* move the code to conditionally break out of the loop upon canceling a
replace up to just after do_yesno() is called and the replace is
possibly done, which is when it should be used anyway (DB)

* change length_change from a long to an int, since the x-coordinates
are ints and a size_t won't work due to its being unsigned and the
chance that the length change is negative (me)


_____________________________________________________________
Sluggy.Net: The Sluggy Freelance Community!