Let’s begin this session by removing debug statements from both the Xvc library and the Python bindings.

Another issue is the restart script. The if in that script that restarts the server doesn’t work; it always restarts the notebook server.

I removed some println! statements from xvc.py. There doesn’t seem to be anything in the library related to outputs.

Let’s search for how to check if a jupyter-lab command with the port 7979 runs in the background.

It looks like the bug is in the condition; it’s not -s, it’s -z.

Oops, yeah.

Let’s do a bit of tidying and check if the script is fixed.

I have a clippy warning with a new function that says these usually don’t take self as a parameter. This is for the pipeline new command, and it receives a self as an XvcPipeline object. It seems best to turn off the clippy warning for this.

I allowed two clippy warnings, and the script seems to work fine.

Let’s go on to copying the content from the Xvc README to the notebook.

There is an issue with the run-after-commit script. When Xvc commits the changes, the command we give is run again.

The issue is that git initializes the directory in the test-data/ directory, while xvc works in the current directory. I think we can either git init in the current directory or xvc init in test-data.

We’re already deleting .git and .xvc directories in the start-readme script. I think it may be easier to update git init to just initialize in the current directory.

Yep, let’s do it that way.

I see there are still extra outputs from the commands. We need to deal with this first.

I removed them, but there are still pink outputs. These are from dbg! statements, it looks like, or we’re initializing the output thread incorrectly.

Fixed those as well. I wrote up the xvc file list command examples as well. Now we have issues with xvc storage new s3 not running, and not even showing any debug output. We’ll deal with it in the next devlog, though.