bzr branch should be resumable if interrupted
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Bazaar |
Confirmed
|
Wishlist
|
Unassigned | ||
Breezy |
Triaged
|
Medium
|
Unassigned |
Bug Description
If you do "bzr branch http://
At the moment, 'bzr branch' doesn't create a target branch until after the repository is copied. The current workaround is:
bzr branch http://
^C
cd local
bzr init
bzr pull http://
This has some small issues if the source is not the default format, etc. But it does let you re-use whatever data has been downloaded.
Alternatively, if you have a local shared repository, the data is stored there, so doing:
bzr init-repo --trees repo
cd repo
bzr branch http://
^C
rm -rf local
bzr branch http://
Should only copy the extra bits that it needs.
As an aside, because of how things are copied, all of inventory.knit is downloaded again. It would be nice if it could download data from file knits in an order that in knew when inventory revisions had been finished. (download in revision order, rather than file-id order). But that has its own problems.
Changed in bzr: | |
importance: | Undecided → Wishlist |
status: | Unconfirmed → Confirmed |
Changed in brz: | |
status: | New → Triaged |
importance: | Undecided → Medium |
tags: | added: performance resume |
Any news on this?