Jupyter Debugger Protocol

[SylvainCorlay]

Votes for approval:

• [x] @afshin
• [x] @blink1073
• [x] @Carreau
• [x] @damianavila
• [x] @ellisonbg
• [x] @fperez
• [x] @ivanov
• [x] @jasongrout
• [ ] @jhamrick
• [x] @minrk
• [ ] @mpacer
• [ ] @parente
• [x] @rgbkrk
• [ ] @Ruv7
• [x] @SylvainCorlay
• [ ] @takluyver
• [x] @willingc

[bollwyvl]

Great work, all, on the stuff leading up to this!

+1 to DAP: it has a well-defined JSON schema, and the specification is generated from it.

Indeed, when integrating this into the documentation, it may be appropriate to formalize the Jupyter Kernel Messaging spec as a whole into a (set of) JSON schema vs the status quo of some .rst and a reference implementation, which similarly plagues LSP(#26). This would also be helpful for updating the aging jupyter/jupyter_kernel_test.

[SylvainCorlay]

Indeed, when integrating this into the documentation, it may be appropriate to formalize the Jupyter Kernel Messaging spec as a whole into a (set of) JSON schema vs the status quo of some .rst and a reference implementation, which similarly plagues LSP(#26).

Absolutely!

[afshin]

👍

I think this will open the way for many other kernel authors to opt into offering a debugging experience for users.

[SylvainCorlay]

Thanks for chiming in @captainsafia! I don't know which level of detail should go in the JEP vs the individual PRs but I wanted to get the ball rolling now that we have an implementation.

[meeseeksmachine]

This pull request has been mentioned on Jupyter Community Forum. There might be relevant details there:

https://discourse.jupyter.org/t/sustainability-of-the-ipynb-nbformat-document-format/3051/2

[SylvainCorlay]

ping @takluyver @Carreau @minrk @ivanov

I would love to have your vote on this since you worked so much on the original protocol.

[Carreau]

Approving.

[jasongrout]

This proposal has been open over 6 months, and has now earned 70% steering council approval plus several approvals from the wider community. @SylvainCorlay - do you want to address Fernando's comment about adding a remark about %debug? Then I'll merge this as approved (unless there are further objections raised).

[SylvainCorlay]

Do you want to address Fernando's comment about adding a remark about %debug

I just added a remark about the %debug magic in the introduction.

[Carreau]

Jason said:

Then I'll merge this as approved (unless there are further objections raised).

As he said it 6 day ago, I'm going to merge. Failure is unrelated.