Re: gnulib-tool should be interruptible

[Ralf Wildenhues]

Hello Paul,

* Paul Eggert wrote on Thu, Sep 21, 2006 at 12:18:31AM CEST:
> Ralf Wildenhues <address@hidden> writes:
> 
> >> That doesn't sound like much of a real problem, but if it is, this
> >> looks to me like a band-aid that doesn't solve things; it'd cut down
> >> the number of bogus messages without eliminating them.
> >
> > This I don't understand.  If I do the output with one `echo', with sane
> > shells
> 
> That's the part I'm worried about.  :-)

Oh.  Well, if the shell is not sane in this respect, then I don't see
what we should do about it.  Sure, one could try all kinds of different
methods (let `cat' produce the output, for example), but it does not
seem like it would get anywhere closer to being certain.  So yes, it may
not eliminate them on all systems.

> > that will cause exactly one `write',
> 
> Aren't they often buffered, if the string is long enough?

Well, the point was that --version output is short, typically shorter
than PIPE_BUF.

> How about
> 
>         * gnulib-tool (func_version): Create output all at once.  This
>         may help avoid triggering unnecessary SIGPIPEs, and at any
>         rate it doesn't hurt.

Yes, that sounds good.  I used that.

Cheers, and thanks,
Ralf