Re: Compiling Elisp to a native code with a GCC plugin

[David Kastrup]

"Stephen J. Turnbull" <address@hidden> writes:

> David Kastrup writes:
>
>  > > (defun is-the-string-following-point-equal-to-this-string-p (s)
>  > >   (string= s (buffer-substring (point) (+ (point) (length s)))))
>  > >
>  > > or
>  > >
>  > > (defun is-the-string-following-point-equal-to-this-string-p (s)
>  > >   (search-forward s (+ (point) (length s)) t))
>
>  > The former will signal an error when the string is longer than the
>  > rest of the buffer.  The latter won't.
>  > 
>  > You can't figure this out by reading the doc strings of the used
>  > functions.  You have to read their source code.
>  > 
>  > Since a user is not likely to pick the correct one-liner, it might
>  > make sense to define a function for that.
>
> It might, but I would prefer a documentation patch. ;-)

Can't really do the trick here.

"If START or END are outside the accessible buffer region, an error
is signaled.  In case you wanted to check for the presence of a string
str at buffer point, consider using `search-forward' with a LIMIT
of (+ (length str) (point)) and a NOERROR of t.

Or use
(condition-case nil
   (string= s (buffer-substring (point) (+ (point) (length s))))
  (args-out-of-range nil))

Or maybe
(string= s (buffer-substring (point) (min (point-max) (+ (point) (length s)))))
"

I mean, get real.  One can't write suggestions for all possible intended
uses into a DOC string.

Would make more sense to give buffer-substring a NOERROR argument of its
own.  Still ugh.

-- 
David Kastrup