[BusyBox] Outright insanity from gnu sed.

[Rob Landley]

On Sunday 28 September 2003 20:04, Glenn McGrath wrote:
> On Sun, 28 Sep 2003 19:45:33 -0500
>
> Rob Landley <rob at landley.net> wrote:
> > Speaking of which, is there any real world test case where gnu sed's
> > "append a newline for N on EOF" behavior is actually depended upon?
> > (I.E. do we really _have_ to implement this?  It's amazingly ugly, and
> > those scripts are depending on an admitted and documented bug.)
>
> In practice i doubt it makes any difference.

Oh good.  It's toast then.  (Smaller code!  Yah!)

> Looks like your patch is going to be a major improvment, ill be away for
> the next 24-48 hours so i may not see your patch straight away, but i
> look forward to it.

Cool, that gives me more time to debug it. :)

I'm going to file attach the next thing I get that compiles, just in case I 
get hit by a bus or something, but I'll let you know when it passes my tests 
and is worth putting in the tree. :)

Rob

(P.S.  "Passes my tests" very well may involve configuring and compiling a 
working version of binutils.  That was, after all, the thing that started me 
down the primrose path.)

(P.P.S.  What's a primrose?)