Re: [PATCH 2/4] add -u: get rid of "treewideupdate" configuration

[Nguyen Thai Ngoc Duy <pclouds@gmail.com>]

On Sat, Apr 9, 2011 at 11:56 AM, Junio C Hamano <gitster@pobox.com> wrote:
> Instead, when we refactor get/init-pathspec API, we could expose an
> interface to turn one element from argv[] into a "struct pathspec_item".
> Then we can try to feed argv[i] to that string-to-pathspec_item function,
> and consider that argv[i] _is_ a proper pathspec only if it parses
> correctly *and* if it matches either an item in the current working tree.
>
> That would be a moral equivalent of the current verify_filename() check
> but is far more precise one; e.g. the current code rejects
>
>        git grep -e foo '*.c' ;# bad
>
> because '*.c' is not an object name, but lstat("*.c") fails, and you need
> to disambiguate with '--'.  If you rewrite the verify_filename() in the
> way I outlined above, you wouldn't have to.

I considered that, but we may need to go through the whole worktree
just to verify "*.c" matches something. The worst case scenario can be
expensive (eg. doing that on a-forest-with-no-c-file gentoo-x86.git).
Alternative approach is recognize "*.c" has wildcard and let it pass
without actually matching.
-- 
Duy