On Thu, Feb 14, 2008 at 01:27:18PM +0100, Jim Meyering wrote:
Daniel Veillard <veillard@redhat.com> wrote: ...
In general I like the approach 'let's put what we consider fine first and discuss other parts later' . I guess it's the third round of review and I really find easier to comment based on existing stuff than big patches that gets iterated over time. So let's isolate what we consider okay and not needing further discussion, push them now, than in an upcoming fourth iteration,
Yes, monolithic patches are hard to manage, and it's not effective to re-review from scratch a 10,000-line diff when the incremental diff is say just 1/10 the size.
For that reason, I've been pulling from Dan's hg patch queue on an irregular basis, and importing the result into a git repository, putting each "pulled batch" on a new branch. Then, I can compare one batch to another with a simple
git diff t7..t8
If only I had pulled a little more frequently over the last week or two. My final incremental is summarized like this:
$ git diff --shortstat t8..t9-public 74 files changed, 2156 insertions(+), 9861 deletions(-)
FYI, the main changes that you'll have in that diff are - Adding of the virStoragePoolBuild & virStoragePoolDelete methods - Removal of the singleton pool (eg, 'loop' and 'volgrp' drivers) - LVM pool renamed from 'logvol' to 'logical' - More work on the 'disk' pool creation code. Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=|