Crushed by a giant packfile

[Matt Nordhoff]

Chris McCormick wrote:
> On Thu, Nov 26, 2009 at 02:59:35PM +1000, Ian Clatworthy wrote:
>> Chris McCormick wrote:
>>
>>> If you could point me to a tutorial or some documentation on how to do this, I
>>> would really appreciate it.
>> Very briefly:
>>
>> 1. bzr fast-export trunk > old.fi
>> 2. bzr fast-import-filter old.fi -x bad-file > new.fi
>> 3. bzr fast-import new.fi trunk.new
>>
>> See
>> http://doc.bazaar-vcs.org/plugins/en/fastimport-plugin.html#fast-import-filter
> 
> Wonderful, thanks Ian.
> 
> Is there some way I can see which revisions are involved in the packfile or
> does that make no sense? Is there some way to use bzr log to see how large a
> particular commit was? I've had a look at the man pages for log, log-format,
> and search, but I can't seem to find a way to work out the size of a particular
> commit.
> 
> Best,
> 
> Chris.

You could use "bzr dump-btree path/to/.bzr/repository/indices/$hash.tix"
on the .tix file corresponding to the large pack file. That should list
the IDs of the revisions contained in the pack, and the files they
modified. (The revision IDs are the ones that contain email addresses;
the file IDs contain file names. Usually.)

Then use "bzr log -v --show-ids -r
revid:foo at example.com-20091126055857-xxx" to see the revision's log
message, including the files modified in it, with their IDs.
-- 
Matt Nordhoff