On Mon, Jan 13, 2020 at 03:47:51PM +0100, Alexandr Nedvedicky wrote:
> Also kn@ was testing that change back in 2019 and found some glitches [2].
> IMO the issues pointed out by kn@ are present in code already, diff below
> just uncovers them. We can investigate/fix them once the change below will
> be in.
        # cd /usr/src/regress/sbin/pfctl
        # make
        # cd /usr/src/sbin/pfctl
        # pfctl -Fa -a regress\*
        Removing:
          regress/relative
          regress/one/two
          regress/one
          regress/foo/bar
          regress/foo/_4
          regress/foo
          regress/_1/foo/_3
          regress/_1/foo
          regress/_1/_3/_4/_5/_6/_7/_8/_9/_10
          regress/_1/_3/_4/_5/_6/_7/_8/_9
          regress/_1/_3/_4/_5/_6/_7/_8
          regress/_1/_3/_4/_5/_6/_7
          regress/_1/_3/_4/_5/_6
          regress/_1/_3/_4/_5
          regress/_1/_3/_4
          regress/_1/_3
          regress/_1/_2
          regress/_1
          regress

Perfect, everything else in my ruleset (anchors, tables, states) is left
untouched;  the previous diff[1] would ignore `-a anchor' and wipe
*everything* -- that was my biggest issue.

> I'll send partial diffs as a reply to this email to keep things organized.
Thanks, much appreciated.  I'd also be happy with committing diffs
separately so the log keeps clear and separated.

> [1] https://marc.info/?l=openbsd-tech&m=155545759616085&w=2
> 
> [2] https://marc.info/?l=openbsd-tech&m=155769978704563&w=2

Reply via email to