Junio C Hamano [off-list ref] writes:
* ds/sparse-checkout-requires-per-worktree-config (2021-12-21) 5 commits
- sparse-checkout: use repo_config_set_worktree_gently()
- config: add repo_config_set_worktree_gently()
- worktree: add upgrade_to_worktree_config()
- config: make some helpers repo-aware
- setup: use a repository when upgrading format
"git sparse-checkout" wants to work with per-worktree configration,
but did not work well in a worktree attached to a bare repository.
Expecting a redesign?
cf. [off-list ref]
source: [off-list ref]
FWIW, this topic by itself passes its self test, but when queued
near the tip of 'seen', it seems to break t1091.
For an example run that fails on win+VS test (8), see
https://github.com/git/git/runs/4629824103
On Fri, Dec 24, 2021 at 6:11 PM Junio C Hamano [off-list ref] wrote:
Junio C Hamano [off-list ref] writes:
quoted
* ds/sparse-checkout-requires-per-worktree-config (2021-12-21) 5 commits
- sparse-checkout: use repo_config_set_worktree_gently()
- config: add repo_config_set_worktree_gently()
- worktree: add upgrade_to_worktree_config()
- config: make some helpers repo-aware
- setup: use a repository when upgrading format
"git sparse-checkout" wants to work with per-worktree configration,
but did not work well in a worktree attached to a bare repository.
Expecting a redesign?
cf. [off-list ref]
source: [off-list ref]
FWIW, this topic by itself passes its self test, but when queued
near the tip of 'seen', it seems to break t1091.
For an example run that fails on win+VS test (8), see
https://github.com/git/git/runs/4629824103
Unrelated to your comment, other than being about this same series:
I think "redesign" is too strong a word here. I think the patches are
a useful start...and might even represent the majority of the code
needed; it's just that they are focusing too narrowly. core.worktree
should also be handled, and I think we should include fixes at the
`git worktree add` level, not just `git sparse-checkout {init,set}`.
On Fri, Dec 24, 2021 at 9:11 PM Junio C Hamano [off-list ref] wrote:
Junio C Hamano [off-list ref] writes:
quoted
* ds/sparse-checkout-requires-per-worktree-config (2021-12-21) 5 commits
"git sparse-checkout" wants to work with per-worktree configration,
but did not work well in a worktree attached to a bare repository.
FWIW, this topic by itself passes its self test, but when queued
near the tip of 'seen', it seems to break t1091.
For an example run that fails on win+VS test (8), see
https://github.com/git/git/runs/4629824103
That's breaking because t1091.10 is incorrectly using:
git config core.sparseCheckoutCone true &&
rather than:
git config --worktree core.sparseCheckoutCone true &&
All the other instances in that script which set
`core.sparseCheckoutCone` correctly use `--worktree`.