def test_split_at_heading_option(tmpdir):
text = """Markdown text
# Header one
## Header two
"""
tmp_md = str(tmpdir.join('notebook.md'))
with open(tmp_md, 'w') as fp:
fp.write(text)
cm = jupytext.TextFileContentsManager()
cm.root_dir = str(tmpdir)
cm.split_at_heading = True
nb = cm.get('notebook.md')['content']
# Was rst to md conversion effective?
assert nb.cells[0].source == 'Markdown text'
assert nb.cells[1].source == '# Header one'
assert nb.cells[2].source == '## Header two'
nb.metadata['jupytext']['notebook_metadata_filter'] = '-all'
text2 = writes(nb, 'md')
compare(text, text2)
def main(args=None):
parser = make_parser()
args = parser.parse_args(args)
infile = Path(args.infile).resolve()
in_dir = infile.parent
py_outfile = in_dir / f"{infile.stem}_nbsphinx.py"
nb_outfile = in_dir.parent / f"{infile.stem}_nbsphinx.ipynb"
print(f"writing:\n{py_outfile}\n{nb_outfile}")
with open(infile, "r") as input_file:
in_py = input_file.readlines()
# collect = ""
# for the_line in in_py:
# match = image_re.match(the_line)
# if match:
# text = match.group(1)
# soup = BeautifulSoup(text, "html.parser")
# out = soup()
# md_image = template.format_map(out[0].attrs)
# collect += md_image
# else:
# collect += the_line
# with open(py_outfile, "w") as output_file:
# output_file.write(collect)
# with open(nb_outfile, "w") as output_file:
# nb = jupytext.readf(py_outfile)
# jupytext.writef(nb, nb_outfile, fmt="ipynb")
orig_nb = jupytext.readf(infile)
split_cell_re = re.compile(
r"^(?P<front>.*?)(?P<img>\<img\s+src.*\>)(?P<back>.*?)", re.DOTALL)
need_display_import = True
new_nb_cells = list(orig_nb.cells)
for index, item in enumerate(orig_nb.cells):
print(f"at cell {index}")
item["metadata"]["cell_count"] = index
if item["cell_type"] == "markdown":
text = item["source"]
if text.find("pic") > -1:
print(f"found img for: {text[:20]}")
out = split_cell_re.match(text)
if out:
print(f"length of split is {len(out.groups())}")
print(f"splitting cell at index {index}")
cell_dict = dict()
for name in ["front", "back"]:
src = out.group(name)
if len(src) > 0:
cell_dict[name] = new_markdown_cell(source=src)
src = out.group("img")
match = image_re.match(src)
if match:
text = match.group(1)
soup = BeautifulSoup(text, "html.parser")
out = soup()
py_image = py_template.format_map(out[0].attrs)
cell_dict["img"] = new_code_cell(source=py_image)
count = 0
for key in ["front", "img", "back"]:
try:
if key == "front":
new_nb_cells[index] = cell_dict[key]
else:
new_nb_cells.insert(index + count, cell_dict[key])
count += 1
except KeyError:
pass
else:
item["metadata"]["cell_count"] = index
if item["source"].find("IPython.display") > -1:
need_display_import = False
print(f"found python cell: {item['source']}")
if need_display_import:
top_cell = new_code_cell(source="from IPython.display import Image")
new_nb_cells.insert(1, top_cell)
orig_nb.cells = new_nb_cells
# https://nbconvert.readthedocs.io/en/latest/execute_api.html
print(f"running notebook in folder {nb_outfile.parent}")
ep = ExecutePreprocessor(timeout=600,
kernel_name="python3",
allow_errors=True)
path = str(nb_outfile.parent)
path_dict = dict({"metadata": {"path": path}})
try:
out = ep.preprocess(orig_nb, path_dict)
except CellExecutionError:
out = None
msg = f"Error executing the notebook {nb_outfile.name}.\n\n"
msg += f"See notebook {nb_outfile.name} for the traceback."
print(msg)
raise
finally:
if "toc" not in orig_nb["metadata"]:
orig_nb["metadata"].update(toc_meta)
pdb.set_trace()
rearrange_jupytext_metadata(orig_nb["metadata"])
out = writes(orig_nb, "py", nbformat.NO_CONVERT)
pdb.set_trace()
with open(nb_outfile, mode="wt") as f:
nbformat.write(orig_nb, f)
jupytext.writef(orig_nb, py_outfile, fmt="py")
print(f"wrote {nb_outfile} and \n {py_outfile}")
