bin/77031: [patch] comm(1) unable to handle lines greater than LINE_MAX (2048)

[fergus]

On 04.02-21:16, Mario Hoerich wrote:
[ .. ]rgus Cameron:
> Ok, I must admit your patch _is_ far easier on the malloc(3)
> side than the one I posted to -current a while back. I've
> briefly considered to rewrite it, but since your's was the
> only feedback I got (thanks btw) there's probably not much
> interest in my solution anyway. Oh well.

personally i think you should post it 'as is' even if you
don't think it's worth putting extra effort into.  the neater
reference is good to have in the PR.

> In addition to that, how does fgetws_a() cope with files, which
> have no '\n' in their last line? (They should of course, but every
> once in a while some file has not.) I might be wrong here, but
> it seems the do { ... } while (line_p) doesn't terminate properly
> in that case. Just scanning through though, might have missed
> something.

that is the only case it terminates at.  i think my comments
must be ill concieved because the comment directly above
states

	"... stop at EOF (nb: will break from loop at end of line)"

in otherwords the loop will not exit until an EOF (and no
newline - which is standard in my tests).  the program will
'break' (line 385) the loop on an EOL.

-- 
: fergus cameron                :   [ .]        cobbled    :
: ^^^^^^@cobbled.net            : [ ~][ ]             .net :