Re: [PATCH] libposix: Add _HEADERS primaries to posix modules.

[Bruce Korb]

On 02/06/11 09:59, Jim Meyering wrote:
> Hi Bruce,
> 
> Giving an ultimatum like that

Hi Jim, et al.,

I really was not intending it as an "ultimatum".
Rather, it was on (bad?) advice from one of the primary gnulib
maintainers on which I was acting.

> ....  If you get desperate, you can always repost,
> telling any who have not responded that you interpret their silence as
> acceptance, and giving them, say, one day more.

That is precisely what I thought I was doing.
I would very much like to use libposix in the two projects I maintain.
I have been trying to get buyoff on the exact form for months.
Though untrue, it seemed clear that the gnulib-tool changes were
ironed out.  I was wrong.  I missed an exchange back in October
and the upshot was never incorporated into the libposix branch.
Oops.  But given the advice that a quick back-out is acceptable
in case of problems, it seemed going ahead was not lethal.
Anyway, This message was from Thursday:
http://lists.gnu.org/archive/html/bug-gnulib/2011-02/msg00044.html
I pushed on Saturday and I'll just never push again without
utterly explicit approval.  No more "repost, telling any ..." stuff.

> ......  In the future, you can help forestall objections or delays
> by ensuring that two or three projects do indeed bootstrap when using
> your proposed changes.

My two projects did.  They are obviously not projects as intimately
intertwined with gnulib machinery as are m4 or emacs.  I'll pull a
copy of the m4 sources and use it as a sanity check.  (To be explicit,
sharutils, autogen and libposix itself all used gnulib-tool successfully.
This proved to be insufficient.)