git-annex/doc/todo/support_fsck_in_bare_repos.mdwn

What is says on the tin:

    22:56:54 < RichiH> joeyh_: by the way, i have been thinking about fsck on bare repos
    22:57:37 < RichiH> joeyh_: the best i could come with is to have a bare and a non-bare access the same repo store
    22:58:00 < RichiH> joeyh_: alternatively, with the SHA* backend, you have all the information to verify that the local data is correct
    22:58:41 < RichiH> and verifying that would already be a plus. if there  really _is_ a problem, having the SHA is enough to track issues down
    23:09:50 < joeyh_> oh, I think I have code that fsck could use on bare repos already.. just a matter of wiring it up
    23:10:42 < joeyh_> feel free to reopen a bug or whatever so I remember.. the unused command's branch content enumeration could be used in a bare repo
    23:14:51 < joeyh_> unused/dropunused could work in bare repos too btw

> Also `status`'s total annex keys/size could be handled for bare repos. --[[Joey]] 

>> Fsck is done. Rest not done yet. --[[Joey]]

>>> all [[done]]! --[[Joey]] 

[[!meta title="support unused, dropunused in bare repos"]]
2011-10-25 21:40:30 +00:00			`What is says on the tin:`

			`22:56:54 < RichiH> joeyh_: by the way, i have been thinking about fsck on bare repos`
			`22:57:37 < RichiH> joeyh_: the best i could come with is to have a bare and a non-bare access the same repo store`
			`22:58:00 < RichiH> joeyh_: alternatively, with the SHA* backend, you have all the information to verify that the local data is correct`
			`22:58:41 < RichiH> and verifying that would already be a plus. if there really _is_ a problem, having the SHA is enough to track issues down`
			`23:09:50 < joeyh_> oh, I think I have code that fsck could use on bare repos already.. just a matter of wiring it up`
			`23:10:42 < joeyh_> feel free to reopen a bug or whatever so I remember.. the unused command's branch content enumeration could be used in a bare repo`
			`23:14:51 < joeyh_> unused/dropunused could work in bare repos too btw`
status too 2011-10-27 16:59:55 +00:00
			> Also `status`'s total annex keys/size could be handled for bare repos. --[[Joey]]
fsck: Now works in bare repositories. Checks location log information, and file contents. Does not check that numcopies is satisfied, as .gitattributes information about numcopies is not available in a bare repository. In practice, that should not be a problem, since fsck is also run in a checkout and will check numcopies there. 2011-10-29 21:49:37 +00:00
			`>> Fsck is done. Rest not done yet. --[[Joey]]`

unused, dropunused: Now work in bare repositories. Turned out I had already done all the work needed to support this when unused started checking all branches. 2011-10-29 23:16:45 +00:00			`>>> all [[done]]! --[[Joey]]`

status: clean up for bare repositories The backend usage graph shows present keys as well as keys found in the repository tree, so it will also be populated for bare repositories. Changed wording to "visible annex keys", which explains why it's 0 in a bare repository (no keys visible as no tree), and also why it varies depending on which branch is checked out. This seemed better than doing something expensive to look up keys from the git-annex branch. 2011-10-29 23:03:43 +00:00			`[[!meta title="support unused, dropunused in bare repos"]]`