|
From: | John Meinel |
Subject: | Re: [Gnu-arch-users] Re: File naming conventions |
Date: | Mon, 18 Oct 2004 23:36:09 -0500 |
User-agent: | Mozilla Thunderbird 0.8 (Windows/20040913) |
Dustin Sallings wrote:
On Oct 18, 2004, at 17:19, John Meinel wrote:
[...]
I always work around it with: find . ! -path "*{arch}*" -name .... | xargsThat does seem to help some, but it's just another workaround.
[...]
This is another workaround with a different syntax. Perhaps I should've used gnu grep as an example instead, since (at least in my manpage), there is no --exclude functionality. I'm not convinced that having the --exclude functionality in every application would even be all that desirable, but having to use it all the time seems like it would.
The one thing that I wanted to point out, is that if you have nested projects, or just start the find somewhere above your top directory, it doesn't matter if the directory is hidden or not. find/grep/ctags/etc will find it. Even if you just do "find ." instead of "find *". The only thing that doesn't match it is a shell glob. At that point you *have* to use one of the workarounds. For me, I have a very nested tree, so I don't have the option, I'm not sure what the common case is, though. I will also say that ".arch" can be pretty easily filtered, and it happens to also look like ".arch-ids" and ".arch-inventory".
Actually, I really wish grep had the workaround, because otherwise I have to do this little hack:
grep -rn "what I want" * | grep -v "{arch}/"which means it still has to grep all of those files, it just doesn't tell me about it.
All this being said, if I were to vote for one, I'd probably vote for ".arch". It's more consistent with other important files used by tla, and it works nicer with globs, etc, etc. I don't really see a definite advantage for "{arch}", other than it is unique. I also don't feel it is a big burden.
John =:->
signature.asc
Description: OpenPGP digital signature
[Prev in Thread] | Current Thread | [Next in Thread] |