On Jan 30 20:01, Corinna Vinschen wrote:
> On Jan 30 19:53, Corinna Vinschen wrote:
> > On Jan 30 11:47, Brian Inglis wrote:
> > > On 2019-01-30 10:31, Corinna Vinschen wrote:
> > > > On Jan 30 09:11, Brian Inglis wrote:
> > > >> On 2019-01-30 07:03, Corinna Vinschen wrote:
> > > >>> I uploaded a new Cygwin test release 3.0.0-0.2
> > > >>> It also changes the output of uname(2) for newly built applications.
> > > >>> Applications built so far (that includes uname(1) from coreutils)
> > > >>> will still print the old uname output. The new format allows for
> > > >>> longer
> > > >>> strings. Compare:
> > > >>> Upcoming new uname content:
> > > >>> sysname: CYGWIN_NT-10.0-17763 or CYGWIN_NT-10.0-17763-WOW64
> > > >>> release: 3.0.0-335.x86_64 or 3.0.0-335.x86_64.snap
> > > >>> version: 2019-01-29 19:23 UTC Build time in UTC
> > > >> Re: "(*) It would really be nice not having to ask for these
> > > >> infos every time." may want to append
> > > >> HKLM/SOFTWARE/Microsoft/Windows NT/CurrentVersion/UBR to uname
> > > >> -s sysname to show the patch levels of installed builds, as
> > > >> there appears to be substantial differences between editions
> > > >> and service models.
> > > [...]
> > > $ cmd /c ver
> > >
> > > Microsoft Windows [Version 10.0.17134.523]
> > >
> > > and save asking those who don't know that, in case the revision
> > > makes a difference. Insider build feature sets bump the builds,
> > > and patch sets bump those revisions; up to base releases with
> > > known feature sets, builds, and revisions; then patch Tuesdays
> > > bump those revisions higher; so you can tell if installs are
> > > Insider, base, or patched.
> >
> > I'm not so sure this makes sense from a Cygwin perspective. We're
> > interested in the major releases introducing changing and/or new
> > functionality. The monthly updates don't do that so they have no
> > meaning to us.
> >
> > I just wonder if we should replace the build number with the ReleaseId
> > (i.e. 17763 vs. 1809), but that excludes the fast lane updates from
> > being visible.
>
> On second thought there's also the format discrepancy. Right now the
> new uname crates the version string like "10.0-17763", but it might be
> better to use "10.0.17763", replacing the dash with a dot, to follow
> more closely the OS layout. On third thought it seems prudent to
> print either
>
> 10.0-1809{-WOW64}
>
> or
>
> 10.0.17763.253{-WOW64}
>
>
> Hmm. The second form appears to make the most sense, actually.But then again, no OS before W10 printed that info, e.g.: Microsoft Windows [Version 6.3.9600] We also have to make sure we're not breaking scripts, especially autoconf etc., so on forth thought, I'll rather stick to the current format. Corinna -- Corinna Vinschen Cygwin Maintainer
signature.asc
Description: PGP signature
