On Fri, May 26, 2017 at 09:24:32AM -0400, Jeff King wrote:
> The middle ground would be to replicate a simple subset of the rules in
> verify_filename(). If we assume that all arguments with ":(" are
> pathspecs (which seem rather unlikely to have false positives), then
> that leaves only a few short-magic patterns: :/, :!, and :^.
>
> We already specially handle :/ here. So it would really just be adding
> the other two (which are just aliases of each other). It's not that much
> code. The dirty thing is just that we're replicating some of the
> pathspec logic, and any new magic would have to be updated here, too.
>
> I'll see if I can work up a patch.
So here's what I came up with. TBH, I could live without patch 5. What I
care most about is making the ":^" work. But I don't really see a
downside to it.
[1/6]: t4208: add check for ":/" without matching file
[2/6]: check_filename(): refactor ":/" handling
[3/6]: check_filename(): use skip_prefix
[4/6]: check_filename(): handle ":^" path magic
[5/6]: verify_filename(): treat ":(magic)" as a pathspec
[6/6]: verify_filename(): flip order of checks
setup.c | 42 ++++++++++++++++++++++++++++++++----------
t/t4208-log-magic-pathspec.sh | 32 ++++++++++++++++++++++++++++++++
2 files changed, 64 insertions(+), 10 deletions(-)
-Peff