Check git annex' error reporting about datalad HOT 5 CLOSED

do you have a minimal example to replicate? what version of annex/git?

from datalad.

The error itself may be due to the use of the submodule and not bleeding-edge git annex. My point is the exitcode!

cd wherever-youhave-datalad-repo/datalad/tests/testrepos/basic/r1
git annex init

results in:
init fatal: ref HEAD is not a symbolic ref
ok
(Recording state in git...)

Exitcode: 0

git annex version
git-annex version: 5.20141125
build flags: Assistant Webapp Webapp-secure Pairing Testsuite S3 WebDAV Inotify DBus DesktopNotify XMPP DNS Feeds Quvi TDFA CryptoHash
key/value backends: SHA256E SHA1E SHA512E SHA224E SHA384E SKEIN256E SKEIN512E SHA256 SHA1 SHA512 SHA224 SHA384 SKEIN256 SKEIN512 WORM URL
remote types: git gcrypt S3 bup directory rsync web webdav tahoe glacier ddar hook external
local repository version: 5
supported repository version: 5
upgrade supported from repository versions: 0 1 2 4

The "fatal" on stderr when exiting with 0 seems to be a strange combination. It's either not so fatal at all or should return non-zero, right?

Edit:
This leads to the question whether checking git-annex' exitcode is enough or some parsing of stderr is needed.

from datalad.

what systems it that on? there is now more relevant moving gears in annex than before, so it would be worth checking first with recent annex before we inquire help from joey

from datalad.

It's jessie.
By now, I don't want joey to fix something. We should have a closer look
(or just ask joey), whether or not catching the exitcode is supposed to be
sufficient for our needs.

2015-03-19 4:01 GMT+01:00 Yaroslav Halchenko [email protected]:

what systems it that on? there is now more relevant moving gears in annex
than before, so it would be worth checking first with recent annex before
we inquire help from joey

—
Reply to this email directly or view it on GitHub
#69 (comment).

from datalad.

Joey said that it is most probably due to that submodule handling in
Windows, which was recently fixed up. So I will close this issue, but
if it reappears with the new version of annex -- feel free to reopen.

Otherwise, git annex must spit out the error code if operation didn't
succeed as a result. There might be intermediate failures which could
cause it to fall back to alternative way to accomplish the same thing.

from datalad.