Re: Two Identical Revision Dates by Two Different Authors

[Mark D. Baushke]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Jon Froehlich <address@hidden> writes:

> This may sound like a stupid question, but I want a definitive answer
> and that is difficult to obtain through trial/error or
> experimentation.
> 
>  
> 
> Consider the following scenario: two authors are working on two
> separate source files in the same project. 

The two separate files would also need to be in two different
directories, otherwise the directory locking would force them int
commits on a different second.

> They both, coincidentally, commit changes at the EXACT same time to
> the CVS repository. 

Granularity for CVS is one second and it is very easy to have two
commits occur to the repository within one second by different authors.

> Will CVS timestamp both commits with the exact same time/date?

Yes.

> I realize that two commits occurring at the exact same time in CVS is
> unlikely because CVS granularity is at the millisecond level.

The RCS timestamp format is a the level of one second, not one
millisecond. However, locks intentionally wait until the next second to
give some delay to the proceedings so that now two versions of the same
file may occur at the same second.

> I guess my question is, can two equivalent revision dates (same exact
> revision timestamp) exist in a CVS repository by two different
> authors?

Yes.

        -- Mark
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (FreeBSD)

iD8DBQFBKPU13x41pRYZE/gRAltMAJ4uaAwpyDPiVG8JWw6MSdZmEmpSdACgpmeJ
dbYPuH21S/aUjGHX1SEwHdk=
=4g9E
-----END PGP SIGNATURE-----