On Thu, 28 Sep 1995, Paul Marquess wrote:
> From: Raphael Manfredi <ram@hptnos02.grenoble.hp.com>
>
> > As for the non-blocking I/O not being in perl's Configure, the unit
> > is now part of the standard metaconfig release. So it must be that
> > the right symbols are not scanned by Andy's metaconfig. He may have
> > omitted the relevant files from the MANIFEST.new?
Hmm. No, the problem appears to be that none of the symbols
defined by the unit are actually used in the 5.001m sources.
Those symbols are:
VAL_O_NONBLOCK, VAL_EAGAIN, RD_NODATA, EOF_NONBLOCK
Was there a patch to POSIX.xs or FCNTL.xs or something I was supposed to
apply that used those symbols, or was I supposed to fake it and insert
the unit anyway for future use by extensions? I didn't deliberately
exclude it; I just didn't remember to explicitly do anything to include it.
> Can this get added next time Andy?
I'm making no promises about any hypothetical "next time". :-)
Andy Dougherty doughera@lafcol.lafayette.edu
> From: Raphael Manfredi <ram@hptnos02.grenoble.hp.com>
>
> > As for the non-blocking I/O not being in perl's Configure, the unit
> > is now part of the standard metaconfig release. So it must be that
> > the right symbols are not scanned by Andy's metaconfig. He may have
> > omitted the relevant files from the MANIFEST.new?
Hmm. No, the problem appears to be that none of the symbols
defined by the unit are actually used in the 5.001m sources.
Those symbols are:
VAL_O_NONBLOCK, VAL_EAGAIN, RD_NODATA, EOF_NONBLOCK
Was there a patch to POSIX.xs or FCNTL.xs or something I was supposed to
apply that used those symbols, or was I supposed to fake it and insert
the unit anyway for future use by extensions? I didn't deliberately
exclude it; I just didn't remember to explicitly do anything to include it.
> Can this get added next time Andy?
I'm making no promises about any hypothetical "next time". :-)
Andy Dougherty doughera@lafcol.lafayette.edu