Robert Collins wrote:
> Review: Needs Fixing
> measuring this would be good.... but I suspect there are too many confounding factors to get accurate results today. At least for dirstate trees it can avoid building a fileid<->path mapping, so I would be a lot happier if this passed the right path in, rather than not passing it in.
The only surefire way to get an accurate path is to call Tree.id2path,
so I don't think that would save anything.
> Tree transform should know that the rename has occured and be able to determine this easily.
The TreeTransform would need to check for renames and moves of the
current file and all its parents in order to give accurate results-- not
trivial. Not worth it, IMO.
Aaron
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Robert Collins wrote:
> Review: Needs Fixing
> measuring this would be good.... but I suspect there are too many confounding factors to get accurate results today. At least for dirstate trees it can avoid building a fileid<->path mapping, so I would be a lot happier if this passed the right path in, rather than not passing it in.
The only surefire way to get an accurate path is to call Tree.id2path,
so I don't think that would save anything.
> Tree transform should know that the rename has occured and be able to determine this easily.
The TreeTransform would need to check for renames and moves of the
current file and all its parents in order to give accurate results-- not
trivial. Not worth it, IMO.
Aaron enigmail. mozdev. org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://
iEYEARECAAYFAkq SzYoACgkQ0F+ nu1YWqI2mMACfaM zr0Z4akpF/ vYFxXbQPoujN j1K9D3wzvFgSDoC si
RokAnRpcNyYO0L+
=qZys
-----END PGP SIGNATURE-----