Cloning takes a long time about archaeology HOT 2 CLOSED

Is there some way to identify the biggest git commits?

When I am dealing with flat files inside of directories, I just do:

du -sch *|sort -h

I don't get that kind of tree sense of things out of git objects.

yebyen@bono:/home/john/urbit/.git$ ls -A|xargs du -sch|sort -h
4.0K    branches
4.0K    config
4.0K    description
4.0K    HEAD
4.0K    packed-refs
8.0K    info
28K refs
32K logs
44K hooks
44K index
99M objects
99M total
yebyen@bono:/home/john/urbit/.git$ cd objects/
yebyen@bono:/home/john/urbit/.git/objects$ ls -A|xargs du -sch|sort -h
4.0K    info
99M pack
99M total
yebyen@bono:/home/john/urbit/.git/objects$ cd pack/
yebyen@bono:/home/john/urbit/.git/objects/pack$ ls -A|xargs du -sch|sort -h
208K    pack-6748bcae51dc5382b70a1b811e48dd9760e8eaee.idx
98M pack-6748bcae51dc5382b70a1b811e48dd9760e8eaee.pack
99M total

Since the majority of the commit is in a pack file, I don't see any way to pare it down in that sense.

Or did someone repack the github.com repo in some way, to fix your issue in the last 13 days? I don't know if that is actually a thing. Does your .git/ directory look the same as mine above?

from archaeology.

You could always fork and squash all of the commits in your local branch

Staying in sync after that will require some git-fu

from archaeology.