Re: [Gnu-arch-users] Re: cscvs--experimental--1.1 nearing doneness; call for testers

[Tupshin Harper]

Tom Lord wrote:

>    > From: Charles Duffy <address@hidden>
>
>    > Incidentally: bkcvs makes some guarantees [every revision in a given
>    > changeset at the same timestamp, no two changesets at the same
>    > timestamp] that make it ideal for doing a perfectly accurate cscvs-style
>    > reconstruction very easily... 
>
>
> I don't follow BK very closely.   I sample lkml from time to time is
> about it.
>
> One thing I saw go by sounded as if LM was saying their would be
> tar-bundled (traditional) patch-sets for each BK revision of every
> bkbits project.   Is that right?   If so, wouldn't _that_ be a very
> easy source to use for cross-revctl-system mirroring?
>
> -t
>  
Yes, that is almost certainly the "correct" way of doing any BK<-->Arch. 
The bkcvs route is (IMO) an undesirable hack as opposed to working with 
the BK changesets directly.

Rik van Riel makes available (via ftp, http and rsync) the broken out 
changesets for the 2.4 and 2.5 trees.
ftp://nl.linux.org/pub/linux/bk2patch/
rsync://nl.linux.org/pub/linux/bk2patch/

Anyone with a bitkeeper tree can easily export it in this format. 
Currently, the 2.5 patches are somewhere north of 1.7G and 100,000 patches.

-Tupshin