On 3/6/2011 10:19 PM, Jelmer Vernooij wrote:
> Public bug reported:
>
> Branch.set_revision_history scales with the size of history - it's the
> main reason uncommit is slow on big trees.
>
> It also complicates the Branch interface unnecessarily for foreign
> implementations.
There is no reason to use it over Branch.set_last_revision_info() (if we
are missing that api, we should add it.)
I think there is even one that lets you not know the last revno, and use
a revision. Which then uses a Graph object to determine the revno from a
couple of starting points.
I'm fine nuking it, or at least deprecating it, etc.
John
=:->
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Cygwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 3/6/2011 10:19 PM, Jelmer Vernooij wrote: set_revision_ history scales with the size of history - it's the
> Public bug reported:
>
> Branch.
> main reason uncommit is slow on big trees.
>
> It also complicates the Branch interface unnecessarily for foreign
> implementations.
There is no reason to use it over Branch. set_last_ revision_ info() (if we
are missing that api, we should add it.)
I think there is even one that lets you not know the last revno, and use
a revision. Which then uses a Graph object to determine the revno from a
couple of starting points.
I'm fine nuking it, or at least deprecating it, etc.
John
=:->
-----BEGIN PGP SIGNATURE----- enigmail. mozdev. org/
0m6MACgkQJdeBCY SNAAM0qQCfSbWgP RaNcOawrIz0OVZz o8WK pXlaof4jKj6b08t Gq
Version: GnuPG v1.4.9 (Cygwin)
Comment: Using GnuPG with Mozilla - http://
iEYEARECAAYFAk1
rPoAnR7nPnnvGF/
=Yn3X
-----END PGP SIGNATURE-----