On Wednesday 15 February 2006 13:36, Vadim Gelfer wrote:
> New submission from Vadim Gelfer <vadim.gelfer@gmail.com>:
>
> i am trying to fix issue 126. when my fix does not work, i think it is my
> fault, but diff code is generating wrong output instead. i have attached
> broken diff. look for hunk at end. line with "-I" is only different with
> white space. but shows up in diff output as changed, even with -w option to
> diff.

It came as part of a larger hunk.  The current ws handling only ignores hunks 
where nothing but ws have changed.

-chris