git-annex/doc/todo/export.mdwn
Joey Hess f4be3c3f89
merge changes made on other repos into ExportTree
Now when one repository has exported a tree, another repository can get
files from the export, after syncing.

There's a bug: While the database update works, somehow the database on
disk does not get updated, and so the database update is run the next
time, etc. Wasn't able to figure out why yet.

This commit was sponsored by Ole-Morten Duesund on Patreon.
2017-09-18 19:21:41 -04:00

43 lines
1.7 KiB
Markdown

`git annex export` corresponding to import. This might be useful for eg,
datalad. There are some requests to make eg a S3 bucket mirror the
filenames in the git annex repository with incremental updates,
which seem out of scope (and there are many tools to do stuff like that
search "deploy files to S3 bucket"),
but something simpler like `git annex export` could be worth doing.
`git annex export --to remote files` would copy the files to the remote,
using the names in the working tree. For remotes like S3, it could add the
url of the exported file, so that another clone of the repo could use the
exported data.
Would this be able to reuse the existing `storeKey` interface, or would
there need to be a new interface in supported remotes?
--[[Joey]]
Work is in progress. Todo list:
* bug: export db update does not reash disk after Remote.Helper.Export calls
updateExportTree.
* tracking exports
* Support configuring export in the assistant
(when eg setting up a S3 special remote).
This is similar to the little-used preferreddir= preferred content
setting and the "public" repository group. The assistant uses
those for IA, which could be replaced with setting up an export
tracking branch.
Low priority:
* When there are two pairs of duplicate files, and the filenames are
swapped around, the current rename handling renames both dups to a single
temp file, and so the other file in the pair gets re-uploaded
unncessarily. This could be improved.
Perhaps: Find pairs of renames that swap content between two files.
Run each pair in turn. Then run the current rename code. Although this
still probably misses cases, where eg, content cycles amoung 3 files, and
the same content amoung 3 other files. Is there a general algorythm?