patchPrint help output when no command is given

The following patches improve the help output by TopGit. They are not all related, but they conflict.

I would just push them to the master branch, but I don't want to risk my chances of Petr getting angry with his new release assistant...

The user ought to be able to call `tg help` from anywhere in the filesystem, not just Git repositories, so the help parsing has to happen before the calls to git git binary.

Debian bug: #501982

Signed-off-by: martin f. krafft <madduck@debian.org>
---
 tg.sh |   17 ++++++++++++++---
 1 files changed, 14 insertions(+), 3 deletions(-)

Printing --help is not an error, but a successful operation, if the help output could be printed.

Signed-off-by: martin f. krafft <madduck@debian.org>
---
 tg.sh |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

Without cmddir, tg is basically useless, even do_help() needs it, so check it first and die hard if not found

Signed-off-by: martin f. krafft <madduck@debian.org>
---
 tg.sh |    6 +++---
 1 files changed, 3 insertions(+), 3 deletions(-)

As much as I love your message (although you really ought not be throwing around ducks!), this is more user-friendly.

Signed-off-by: martin f. krafft <madduck@debian.org>
---
 tg.sh |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

On Wed, Nov 19, 2008 at 05:03:40PM +0100, martin f. krafft wrote:

Huh, why do you actually need $args_saved at all? :-) This is bound to do horrible things with space-containing arguments etc., I think. You don't need to do the outer shift and then drop $args_saved altogether, no?

also sprach Petr Baudis <pasky@suse.cz> [2008.11.19.1924 +0100]:

I figured I need to restore $@ for others to consume, e.g. when you check for -r later.

The challenge is to identify help|--help|-h anywhere on the command line. Thus, you need to iterate, or do some weird matching against $*.

When you iterate in a for loop, it's not easy to get at the next argument, except if you use a state machine. I wanted to avoid that.

But you are right, I am wreaking havoc with space-containing arguments. I will have to go back and rework this.

I take it you agree with the general principle though?

On Wed, Nov 19, 2008 at 07:32:34PM +0100, martin f krafft wrote:

Oh, oops - I missed the loop, I should get some lunch. ;-) Thus, something like this?

	check_help() { while [ -n "$1" ]; do ...; shift; done }
	check_help "$@"

also sprach Petr Baudis <pasky@suse.cz> [2008.11.19.1939 +0100]:

Why didn't I think of that???

I just sent an updated patch without threading, sorry. tg-mail just fires without confirmation... :/

Hi,

On Wed, Nov 19, 2008 at 17:03, martin f. krafft <madduck@debian.org> wrote:

I really second this. Plus, I think its crucial to check that we are not inside the .git directory before setting up topgit. Because git rev-parse --git-dir only works in the top .git dir, not deeper (i.e. inside .git/refs) and will always return ".".

	$(git rev-parse --is-inside-work-tree) ||
		die "Not in a git working directory"

Bert

I think you wanted to CC Petr, not Junio. It's also useful to CC Debian bugs if you know of them. :)

also sprach Bert Wesarg <bert.wesarg@googlemail.com> [2008.11.21.1019 +0100]:

Thanks, noted. I am still working on this patch, since evaluating $argv/checking for --help at this early stage basically requires a rewrite of the CLI parsing, ideally using POSIX getopt or something similar, which takes much of the load away.

Anyway, Patch for your proposal forthcoming.

Bert Wesarg suggests to check that we are not inside the .git directory before setting up topgit. Because git rev-parse --git-dir only works in the top .git dir, not deeper (i.e. inside .git/refs) and will always return "."

This patch thus causes topgit to die early on when it's called from outside of a Git repository, or from underneath .git and outputs error messages accordingly.

This temporarily makes it even more impossible to call tg --help, but I expect to have that fixed soon too.

Signed-off-by: martin f. krafft <madduck@debian.org>

---
 tg.sh |   16 ++++++++++++++++
 1 files changed, 16 insertions(+), 0 deletions(-)

On Fri, Nov 21, 2008 at 13:06, martin f krafft <madduck@debian.org> wrote:

I CC'ed Junio, because I think this git rev-parse --git-dir is a bug. For example you can try starting git gui inside a .git dir.

A patch would depent on your --help patch, Because help should work everywhere. So either I wait for you or do do it.

Bert

On Fri, Nov 21, 2008 at 13:15, Bert Wesarg <bert.wesarg@googlemail.com> wrote:

Oh, I send a patch only to petr, but forgot to cc the list ;-)

But your patch should be fine, thanks.

Bert

also sprach Bert Wesarg <bert.wesarg@googlemail.com> [2008.11.21.1318 +0100]:

Maybe you can try to remember to CC me on topgit stuff in the future too, since I am helping Petr out a bit as his release apprentice. :)

also sprach Bert Wesarg <bert.wesarg@googlemail.com> [2008.11.21.1315 +0100]:

You might want to let him know of your intention. :)

No, not depend, but the patch makes the --help situation slightly worse, for now you cannot call --help inside .git anymore either. This will force us to fix the --help situation quicker.

I've run out of time, for a while at least. If you want to pick up the pieces and convert argument parsing to POSIX getopt, or maybe even better, git-rev-parse --parseopt, then please go for it. You can find my current TopGit branch at

  http://git.debian.org/?p=collab-maint/topgit.git;a=shortlog;h=refs/heads/fixes/independent-help
  git://git.debian.org/git/collab-maint/topgit.git, branch fixes/independent-help

If you use Debian: debcheckout topgit (requires devscripts 2.10.40)

Thanks,