I'd like to see both here, if the timestamps are exactly the same assume
its the same, otherwise take the time to open the local file and run it
through MD5 and compare it to the checksum in the local CVS/Entries to make
sure. We might need an option to force one of three possible modes
(timestamp only, MD5 only, and hybrid)
----- Original Message -----
From: "Pavel Roskin" <[EMAIL PROTECTED]>
To: "Jonathan M. Gilligan" <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>
Sent: Wednesday, April 12, 2000 12:09 PM
Subject: Re: Could timestamps be replaced with MD5?
> Hello, Jonathan!
>
> > Is a one-way hash such as MD5 really superior to a CRC here? I believe
that
> > CRC's are substantially cheaper to compute, which could be relevant in
the
> > case of large projects. The nonrepudiability and resistance to forgery
that
> > recommend one-way hashes for cryptographic protocols seem kind of
> > irrelevant here.
>
> I agree that MD5 is an overkill here, and a simple CRC should be Ok.
> However, timestamps are already used by "make".
>
> Although it's often not the right thing to do, makefiles may contain rules
> involving "touch". Files are touched so that the dependent files are
> rebuilt. There is no intention to confuse CVS. The intention is only to
> affect the behaviour of make.
>
> Linux kernel uses this technique. Also automake-generated makefiles may
> want to rebuild files under the source control. Again, this is not
> perfect, but even CVS itself has "configure" under the source control.
>
> If a file is rebuilt and is equal to the original version except the
> timestamp, why send it to the server for comparison? Slow connections do
> exist. Big files exist too.
>
> Regards,
> Pavel Roskin
>
>
