Protocol of dataset component syncing in superds clone in project directory

[mih]

# change into (uninstalled) subdataset
cd derivative/cortical_surfaces_freesurfer

# obtain subdataset (datalad is configured to pull it from the internal RIA store via SSH)
datalad get -n .

# undo https://github.com/datalad/datalad/issues/5628
git annex enableremote inm7-storage url=ria+http://studyforrest.ds.inm7.de

# verify that is obtained `inm7-storage` from bulk1 vis SSH
git cat-file -p git-annex:remote.log

# add a GIN project (already created via the website), and configure as a type-git special remote
datalad siblings add -d . --name gin-update --pushurl git@gin.g-node.org:/studyforrest/cortical-surfaces-freesurfer.git --url https://gin.g-node.org/studyforrest/cortical-surfaces-freesurfer.git --as-common-datasrc gin

# undo undesired auto-config made by git-annex
git config --unset-all remote.gin-update.annex-ignore

# register github project too
git remote add github git@github.com:psychoinformatics-de/studyforrest-data-freesurfer.git

# get the state of all repos
git fetch --all --prune

# visual inspection -- sometimes the status quo is chaotic
tig --all

# merge pending contributions
git merge github/master

# pull all data locally for upload to GIN
datalad get .

# upload to GIN
datalad push --to gin-update

# sync availability with GitHub
datalad push --to github
datalad push --to origin

# verify availability info
git annex find --not --in gin-update
git annex find --not --in inm7-storage

# drop all files
datalad drop .

TODO

• [ ] go through and add publication dependencies on origin for any push to gin-update or github, to make sure we have all changes locally too.