On Sat, 2 Sep 2006, Junio C Hamano wrote:
repack without -a essentially boils down to:
rev-list --objects --all --unpacked |
pack-objects $new_pack
which picks up all live loose objects and create a new pack.
If rev-list had an extention that lets you say
rev-list --objects --all --unpacked=$active_pack |
pack-objects $new_pack
I have to say, that adding another pack-objects-specific flag to rev-list
doesn't necessarily strike me as very nice.
It might be much better to just teach "git pack-objects" to walk the
object list by hand. The whole "git-rev-list | git-pack-objects" approach
made more sense back when git-rev-list was the _only_ think listing
objects, but now that we've librarized the whole object listing thing,
maybe it's time to avoid the pipeline and just do it inside the packer
logic.
Of course, having git-pack-objects be able to take input from stdin is
still useful, but I'd rather start moving the obviously packing-specific
flags out of git-rev-list, and into git-pack-objects instead.
[ That would also potentially make packing more efficient - right now the
"phase 1" of packing is literally to just figure out the type and the
size of the object, in order to sort the object list.
So when you do a big repack, first "git-rev-list" ends up doing all this
work, and then "git-pack-object" ends up _redoing_ some of it.
Especially for tree objects (which are one of the most common kinds), we
already opened the object once when we traversed it, so opening it again
just to look at its size is kind of sad.. ]
We also used to have a bug with repacking, where a git-rev-list that died
with a failure would silently cause a "good" repack to happen because it
had fed git-pack-objects an incomplete list of objects. Junio fixed that
with a hack, but if we did all the object listing inside git-pack-objects,
we'd not need that kind of things at all..
Hmm? Comments?
Linus
--
VGER BF report: U 0.500053