On Fri, Nov 19, 2010 at 06:36:13PM +0000, Peter Dennis wrote: > On 19/11/2010 18:23, [email protected] wrote: > >If so, that's what the atime and mtime should be set to. Otherwise, > >you're depending on implementation specific behavior. The code in > >action.file makes a temporary file, and copies the content of the > >file in the download cache into that tempfile. The tempfile is then > >renamed to the name of the file that's being installed. The > >transport will preserve the file times from the remote side on > >download, but the install code doesn't copy over the atime/mtime > >attributes right now -- unless there's a timestamp in the manifest -- > >because there hasn't really been any need. > > Okay - and that was the answer I was looking for - essentially this means > it is something that pkg(5) can handle (by setting a timestamp attribute > in the manifest) but it is something that has not been implemented > in the manifests - and so the question then translates should it be ? > But that I would guess is not within the scope of this forum.
I guess part of the question is should the installed files, if they have no timestamp in the manifest, get the timestamp from the downloaded files or the timestamp from the creation of the tempfile? Relying on the timestamp from download means that who knows what will show up on the files, since it's controlled by the server-side. Whereas using the time of tempfile creation is at least consistent. However, if there's a use case we're missing, we could re-consider how this is implemented. -j _______________________________________________ pkg-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
