Re: Q: does index-pack work in place on windows?

Nicolas Pitre [off-list ref] writes:

On Wed, 2 Feb 2011, Junio C Hamano wrote:

quoted

I am trying to see if an index-pack with slight modification would be a
good replacement for verify-pack.

...
index-pack _could_ be a replacement for verify-pack.  It certainly can 
validate a pack since it is its purpose, possibly faster than 
verify-pack.  You'd still have to compare the existing pack index 
against the one index-pack creates without overwriting that original 
index, taking into accound index version differences, etc.

We already know index-pack is a lot faster when you have a lot of deep
deltas, as it works from a base to its immediate delta children while
pinning that base, as opposed to verify-pack that verifies each and every
object in the pack in the index order, inflating and then applying
potentially long delta chains repeatedly---the only thing that could be
helping it right now is the in-core delta base cache.

However index-pack won't tell you what is broken in the pack when 
corruptions are to be found.

Yes, but at that point, you are pretty much lost anyway, as the only thing
you can do to salvage salvageable parts of the broken pack, if you still
trust its associated .idx file, is to walk the table of contents and ask
for each individual object; knowing where the .pack or .idx is broken with
the current verify-pack does not help you very much.

On Thu, 3 Feb 2011, Junio C Hamano wrote:

Nicolas Pitre [off-list ref] writes:

quoted

However index-pack won't tell you what is broken in the pack when 
corruptions are to be found.

Yes, but at that point, you are pretty much lost anyway, as the only thing
you can do to salvage salvageable parts of the broken pack, if you still
trust its associated .idx file, is to walk the table of contents and ask
for each individual object; knowing where the .pack or .idx is broken with
the current verify-pack does not help you very much.

Well... given that you do have to compare both the generated index and 
the on-disk index at some point, it is then possible to identify 
corrupted objects and their location.  And with pack index version 2 
including a CRC for the packed data, you should be able to tell if the 
corruption is actually in the pack or in the index (or both if the final 
pack/index SHA1 sum doesn't match).


Nicolas