compilation-next-error - extra `let' binding?

compilation-next-error - extra `let' binding?

[Drew Adams]

Can't tell for sure without checking all of the code called by the
body of the `let', but it looks like the `let' binding of `loc' in
`compilation-next-error' is superfluous:
 
(let* ((msg (get-text-property pt 'message))
       (loc (car msg)) ; ???????
       last)
 
If so, taking out the binding would make the code easier to
understand. If not, adding a comment for it might make the code easier
to understand.
 

In GNU Emacs 22.1.90.1 (i386-mingw-nt5.1.2600)
 of 2008-01-30 on PRETEST
Windowing system distributor `Microsoft Corp.', version 5.1.2600
configured using `configure --with-gcc (3.4) --cflags -Ic:/gnuwin32/include'
 

RE: compilation-next-error - extra `let' binding?

[Drew Adams]

It also looks like the test in the second branch of this `if' could be just
(> n 0), for clarity:

(if (zerop n)
    (unless ...)
  (setq last (nth 2 (car msg)))
  (if (>= n 0) ; <== Can't be zero here, anyway.
      (compilation-loop ...)
	...))

Re: compilation-next-error - extra `let' binding?

[Stefan Monnier]

> Can't tell for sure without checking all of the code called by the
> body of the `let', but it looks like the `let' binding of `loc' in
> `compilation-next-error' is superfluous:
 
> (let* ((msg (get-text-property pt 'message))
>        (loc (car msg)) ; ???????
>        last)
 
> If so, taking out the binding would make the code easier to
> understand. If not, adding a comment for it might make the code easier
> to understand.
 
It's actually used by the compilation-loop macro.
I've added a comment to this effect.


        Stefan

Re: compilation-next-error - extra `let' binding?

[Stefan Monnier]

> It also looks like the test in the second branch of this `if' could be just
> (> n 0), for clarity:

> (if (zerop n)
>     (unless ...)
>   (setq last (nth 2 (car msg)))
>   (if (>= n 0) ; <== Can't be zero here, anyway.
>       (compilation-loop ...)
> 	...))

Couldn't care less,


        Stefan