On Thu, 2007-09-27 at 08:31 +0100, Heikki Linnakangas wrote:
> Kevin Grittner wrote:
> > <1190837675.4181.716.camel@ebony.site>, Simon Riggs <simon@2ndquadrant.com>
> > wrote:
> >> We should also document that this is designed to help compress files
> >> that aren't full because we switched early because of archive_timeout.
> >
> > Attached is a modified version to implement both of these. I also bailed
> > out if there was surplus input. I tried an optimization of allocating a
> > separate buffer for outputting the zeros, to avoid repeated memset calls.
> > It didn't seem to make a very big difference; do you think it's worth
> > cluttering the code with that?
>
> Would it work to just ftruncate the file?
At the source end, yes. At the destination one of the tests we perform
for a fully and correctly copied file is the file size; a smaller file
size we take to mean that the copy is still in progress. We check that
before we actually read the file, so it would complicate things
considerably to allow for variable file sizes.
-- Simon Riggs 2ndQuadrant http://www.2ndQuadrant.com