Pycharm version control of Jupyter notebooks.

Answered

Pycharm 2019.1 has really powerful features for Jupyter notebooks. Are there plans for PyCharm to improve the version control of Jupyter notebooks since the editor mode is already pure Python code? It would be nice to have an option to automatically generate the markdown files. We can put the markdown files under git version control. When someone pulls the changes, the notebook is compared to the markdown file and the notebook is regenerated if older than the markdown file.

3 comments

Hi,

So by "improve the version control of Jupyter notebooks" you mean adding "automatically generate the markdown files", right? Why wouldn't you just put your Jupyter files under VCS instead?

0

When you rerun the files, the outputs will have a different timestamp even though the underlying code is unchanged. Also, the notebooks are often identified as binary which doesn't make it easy to keep track of changes.

0

You are right, each time you run the cell it changes the execution_count and other things.

Currently, we don't have a solution for that. Maybe we could somehow save the source code for each cell in order to compare it and therefore do not show it in commit dialogue.

Please feel free to submit a feature request to our issue tracker using the link https://youtrack.jetbrains.com/issues/IDEA

0

Please sign in to leave a comment.