Encapsulate code cells with triple backticks with +1 backtick in Markdown

[mwouts]

This PR will close #712 .

@rsokl I am considering taking this opportunity to remove the parsing of Python strings inside code cells (since now the cell marker is non-ambiguous), but that may break some of your notebooks, cf. #419 . Do you have many of them?

Also @chrisjsewell I have not yet ported this to md:myst but I intend to do so before merging.

[mwouts]

I am considering taking this opportunity to remove the parsing of Python strings inside code cells

FYI I changed my mind, and finally prefer to keep the parsing of the code cell, for a while at least, to provide backward compatibility.

[chrisjsewell]

Heya yeh that looks good to me 👍

[codecov]

Codecov Report

Merging #726 (1449249) into master (41b7e9c) will decrease coverage by 0.00%.
The diff coverage is 99.66%.

@@            Coverage Diff             @@
##           master     #726      +/-   ##
==========================================
- Coverage   99.12%   99.12%   -0.01%     
==========================================
  Files          95       97       +2     
  Lines        9502     9591      +89     
==========================================
+ Hits         9419     9507      +88     
- Misses         83       84       +1     

Impacted Files	Coverage Δ
jupytext/metadata_filter.py	99.13% <ø> (ø)
tests/test_auto_ext.py	100.00% <ø> (ø)
tests/test_cells.py	100.00% <ø> (ø)
tests/test_config.py	100.00% <ø> (ø)
tests/test_hide_remove_input_outputs_rmarkdown.py	100.00% <ø> (ø)
tests/test_ipynb_to_py.py	100.00% <ø> (ø)
tests/test_ipynb_to_rmd.py	100.00% <ø> (ø)
tests/test_jupytext_errors.py	100.00% <ø> (ø)
tests/test_jupytext_nbconvert_round_trip.py	100.00% <ø> (ø)
tests/test_jupytext_read.py	100.00% <ø> (ø)
... and 80 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 75cba9e...1449249. Read the comment docs.

[mwouts]

Heya yeh that looks good to me +1

@chrisjsewell you're too fast! You did not even leave me the time to ask 😆

So just to be sure you're also OK with the version bump on the MyST format (to 0.13)? Thanks!

[rsokl]

@mwouts thank you for keeping me posted on this.

I don't think that #419 pops up too much. That being said, with the present change, would upgrading the jupytext-markdown version of my document (i.e. old-md -> ipynb -> new-md) ensure that everything is ok in the future?

[mwouts]

I don't think that #419 pops up too much. That being said, with the present change, would upgrading the jupytext-markdown version of my document (i.e. old-md -> ipynb -> new-md) ensure that everything is ok in the future?

Indeed! The markdown code blocks with triple backticks that encapsulate a code cell with triple backticks in a string will become markdown code blocks with quadruple backticks, like in this test (look at line 784)

jupytext/tests/test_read_simple_markdown.py

Lines 768 to 785 in 1449249

	def test_markdown_cell_with_code_inside_multiline_string_419(
	text='''```python
	readme = """
	above
	```python
	x = 2
	```
	below
	"""
	```
	''',
	):
	"""A code cell containing triple backticks is converted to a code cell encapsulated with four backticks"""
	nb = jupytext.reads(text, "md")
	compare(jupytext.writes(nb, "md"), "`" + text[:-1] + "`\n")
	assert len(nb.cells) == 1