Re: [PATCH 3/3] revision: handle pseudo-opts in `--stdin` mode

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Patrick Steinhardt <ps@xxxxxx> writes:

>  		if (sb.buf[0] == '-') {
> +			const char *argv[2] = { sb.buf, NULL };
> +
>  			if (!strcmp(sb.buf, "--")) {
>  				seen_dashdash = 1;
>  				break;
>  			}
>  
> -			die("options not supported in --stdin mode");
> +			if (handle_revision_pseudo_opt(revs, argv, flags))
> +				continue;
> +
> +			die(_("option '%s' not supported in --stdin mode"), sb.buf);
>  		}

The reason why we had the double-dash checking inside this block is
because we rejected any dashed options other than double-dash, but
with this step, that is no longer true, so it may make more sense to
move the "seen_dashdash" code outside the block.  Also unlike the
command line options that allows "--end-of-options" marker to allow
tweaking how a failing handle_revision_arg() call is handled, this
codepath does not seem to have any matching provision.  In the
original code, which did not accept any options, it was perfectly
understandable that it was unaware of "--end-of-options", but now we
do allow dashed options, shouldn't we be supporting it here, too?

Thanks.



[Index of Archives]     [Linux Kernel Development]     [Gcc Help]     [IETF Annouce]     [DCCP]     [Netdev]     [Networking]     [Security]     [V4L]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux SCSI]     [Fedora Users]

  Powered by Linux