Ondrej Kokes

We're now rendering at most one cell output (https://github.com/kokes/nbviewer.js/commit/255fc3e9e44e4697d007738409ca8431c78608d0) and I've checked it all against the tests in nbformat and nbconvert, but I have yet to resolve the other points...

Not all the notebooks in S3 are valid UTF-8 encoded and proper JSON files, so I filtered them out: ```python import os import json from json.decoder import JSONDecodeError dirname =...

Also, in my testing, I only check notebooks that throw exceptions, but I still have: 1) that one console.error 2) extensions throwing errors (especially katex) Check those as well.

We could also leverage the nbformat and nbconvert tests - https://github.com/jupyter/nbformat/tree/master/nbformat/tests - https://github.com/jupyter/nbconvert/tree/master/nbconvert/tests/files

This commit https://github.com/kokes/nbviewer.js/commit/36c745d6a5a4fc6f8b25ba3a20695c1cdfb469c3 could be tested for regressions using https://github.com/jupyter/nbconvert/blob/master/nbconvert/tests/files/notebook1.ipynb

wip: https://github.com/kokes/nbviewer.js/tree/issue41

This one as well, since we don't support type "heading" https://nbviewer.jupyter.org/github/ipython/ipython/blob/1.x/examples/notebooks/Part%205%20-%20Rich%20Display%20System.ipynb

A simple POC is easy enough, but we need to take into account that we can have complex markdown in a cell, not just a single heading, but multiple.

Chrome extension in a what way? As a separate application or for the times when you're on Github and your notebook is not loading? For the latter case, you can...

OK, so we're rendering tracebacks, but notice that they might contain ANSI colours. For now, we're just stripping them. That's the state as of 1f7be42. [error.txt](https://github.com/kokes/nbviewer.js/files/1143080/error.txt)