[RFC] Bundles as repositories

[Aaron Bentley]

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

Aaron Bentley wrote:
> I propose we have two variants of the format.
> 
> One variant is for human consumption, has a prelude and a base64 wrapper
> on the data, and has its prelude checked by default.

I'm actually reconsidering this.  I'm thinking that merge directives
should take the place of human-readable bundles.  The prelude and the
base64 wrapping would then be the responsibility of the merge directive.

The thing is the prelude implicitly reflects the merge that is being
requested.  If merge directives grow the ability to specify a base
revision, as contemplated in my email re: Re: [MERGE] Better progress
reporting in commit, then the prelude should reflect the base revision
specified by the merge directive.

Current bundle formats could be treated as primitive merge directives
for this purpose.

So this
1. allows us to avoid an additional bundle format
2. clarifies the purposes of merge directives vs bundles
3. hopefully clarifies the rationale for verifying preludes.
4. simplifies code paths where bundles and merge directives are handled
   similarly.

Aaron
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFGd/wU0F+nu1YWqI0RAkFNAJ9d+dRX0s3w0nOJ4eyzuhudT0UrDACfYPWc
XsPxixOLve76c/Y1Fyu4M5E=
=KJeL
-----END PGP SIGNATURE-----