When DVC is only initialized experiments and plots break

[shcheklein]

Get project, uncommit or just copy all files to a new dir and initialize DVC there. Run experiments via CLI. Try to see them in the extension.

For me experiments panel was empty (and basis message with the links to docs). Plots got stuck on "Loading plots" and nothing in the terminal.

Extension version: 0.3.16

DVC version:

DVC version: 2.17.0 (pip)
---------------------------------
Platform: Python 3.9.12 on macOS-12.4-arm64-arm-64bit
Supports:
        azure (adlfs = 2022.7.0, knack = 0.9.0, azure-identity = 1.10.0),
        gdrive (pydrive2 = 1.14.0),
        gs (gcsfs = 2022.5.0),
        hdfs (fsspec = 2022.5.0, pyarrow = 9.0.0),
        webhdfs (fsspec = 2022.5.0),
        http (aiohttp = 3.8.1, aiohttp-retry = 2.7.0),
        https (aiohttp = 3.8.1, aiohttp-retry = 2.7.0),
        s3 (s3fs = 2022.5.0, boto3 = 1.21.21),
        ssh (sshfs = 2022.6.0),
        oss (ossfs = 2021.8.0),
        webdav (webdav4 = 0.9.7),
        webdavs (webdav4 = 0.9.7)
Cache types: reflink, hardlink, symlink
Cache directory: apfs on /dev/disk3s1s1
Caches: local
Remotes: None
Workspace directory: apfs on /dev/disk3s1s1
Repo: dvc, git

[mattseddon]

Yep, I hit this as well. Will need to put a safe guard in. It will be something to do with the experiments data not looking complete.

[shcheklein]

I also hit it again, now with an existing commit. Looks like if something is wrong with one of the commands (and I don't see actually what is happening in logs) based on the previous commit or an existing one- it fails and get stuck in some random state. Doing a few times git commit helps to cleanup the history.

[mattseddon]

I think this is a combination of a few things. The first two things that current jump out are

1. Before the initial git commit we get ERROR: unexpected error - Empty git repo.
2. The bug that I fixed yesterday with #2187.

Unfortunately, 1 is going to be a little more difficult to fix. I will start working on it now.

[mattseddon]

Found a 3rd issue where newly created deps have no hash/nfiles/size, again the code falls over under these circumstances. Working on another fix.

[mattseddon]

This should be fixed now.