def test_alternate_tree_four_five_backticks(
no_jupytext_version_number,
nb=new_notebook(
metadata={"main_language": "python"},
cells=[
new_code_cell('a = """\n```\n"""'),
new_code_cell("b = 2"),
new_code_cell('c = """\n````\n"""'),
],
),
text='''---
jupyter:
jupytext:
main_language: python
---
````python
a = """
```
"""
````
```python
b = 2
```
`````python
c = """
````
"""
`````
''',
):
actual_text = writes(nb, fmt="md")
compare(actual_text, text)
actual_nb = reads(text, fmt="md")
compare_notebooks(actual_nb, nb)
def test_read_nbconvert_script(script="""
# coding: utf-8
# A markdown cell
# In[1]:
import pandas as pd
pd.options.display.max_rows = 6
pd.options.display.max_columns = 20
# Another markdown cell
# In[2]:
1 + 1
# Again, a markdown cell
# In[33]:
2 + 2
# <codecell>
3 + 3
"""):
assert jupytext.formats.guess_format(script, '.py') == 'percent'
nb = jupytext.reads(script, '.py')
assert len(nb.cells) == 5
def test_read_mostly_R_markdown_file(
markdown="""```R
ls()
```
```R
cat(stringi::stri_rand_lipsum(3), sep='\n\n')
```
""",
):
nb = jupytext.reads(markdown, "md")
assert nb.metadata["jupytext"]["main_language"] == "R"
compare_cells(
nb.cells,
[
new_code_cell("ls()"),
new_code_cell("cat(stringi::stri_rand_lipsum(3), sep='\n\n')"),
],
compare_ids=False,
)
markdown2 = jupytext.writes(nb, "md")
compare(markdown2, markdown)
def test_multiline_python_magic(no_jupytext_version_number):
nb = new_notebook(cells=[
new_code_cell("""%load_ext watermark
%watermark -u -n -t -z \\
-p jupytext -v
def g(x):
return x+1""")
])
text = jupytext.writes(nb, "py")
compare(
text,
"""# +
# %load_ext watermark
# %watermark -u -n -t -z \\
# -p jupytext -v
def g(x):
return x+1
""",
)
compare_notebooks(jupytext.reads(text, "py"), nb)
def test_indented_comment(text="""def f():
return 1
# f returns 1
def g():
return 2
# h returns 3
def h():
return 3
""", ref=new_notebook(cells=[new_code_cell("""def f():
return 1
# f returns 1"""),
new_code_cell('def g():\n return 2'),
new_code_cell('# h returns 3\ndef h():\n return 3')])):
nb = jupytext.reads(text, 'py')
compare_notebooks(nb, ref)
py = jupytext.writes(nb, 'py')
compare(py, text)
def test_escape_start_pattern(pynb="""# The code start pattern '# +' can
# appear in code and markdown cells.
# In markdown cells it is escaped like here:
# # + {"sample_metadata": "value"}
# In code cells like this one, it is also escaped
# # + {"sample_metadata": "value"}
1 + 1
"""):
nb = jupytext.reads(pynb, ext='.py')
assert len(nb.cells) == 3
assert nb.cells[0].cell_type == 'markdown'
assert nb.cells[1].cell_type == 'markdown'
assert nb.cells[2].cell_type == 'code'
assert nb.cells[1].source == '''In markdown cells it is escaped like here:
# + {"sample_metadata": "value"}'''
assert (nb.cells[2].source ==
'''# In code cells like this one, it is also escaped
# + {"sample_metadata": "value"}
1 + 1''')
pynb2 = jupytext.writes(nb, ext='.py')
compare(pynb, pynb2)
def test_read_mostly_R_markdown_file(markdown="""```R
ls()
```
```R
cat(stringi::stri_rand_lipsum(3), sep='\n\n')
```
"""):
nb = jupytext.reads(markdown, 'md')
assert nb.metadata['jupytext']['main_language'] == 'R'
compare(nb.cells, [{'cell_type': 'code',
'metadata': {},
'execution_count': None,
'source': 'ls()',
'outputs': []},
{'cell_type': 'code',
'metadata': {},
'execution_count': None,
'source': "cat(stringi::stri_rand_lipsum(3), sep='\n\n')",
'outputs': []}])
markdown2 = jupytext.writes(nb, 'md')
compare(markdown2, markdown)
def test_one_blank_lines_after_endofcell(pynb="""# +
# This is a code cell with explicit end of cell
1 + 1
2 + 2
# -
# This cell is a cell with implicit start
1 + 1
"""):
nb = jupytext.reads(pynb, ext='.py')
assert len(nb.cells) == 2
assert nb.cells[0].cell_type == 'code'
assert (nb.cells[0].source ==
'''# This is a code cell with explicit end of cell
1 + 1
2 + 2''')
assert nb.cells[1].cell_type == 'code'
assert nb.cells[1].source == '''# This cell is a cell with implicit start
1 + 1'''
pynb2 = jupytext.writes(nb, ext='.py')
compare(pynb, pynb2)
def test_magics_are_commented(fmt):
nb = new_notebook(
cells=[new_code_cell("%pylab inline")],
metadata={
"jupytext": {
"comment_magics": True,
"main_language": "R"
if fmt == "R"
else "scheme"
if fmt.startswith("ss")
else "python",
}
},
)
text = jupytext.writes(nb, fmt)
assert "%pylab inline" not in text.splitlines()
nb2 = jupytext.reads(text, fmt)
if "sphinx" in fmt:
nb2.cells = nb2.cells[1:]
compare_notebooks(nb2, nb)
def test_one_blank_lines_after_endofcell(pynb="""# +
# This is a code cell with explicit end of cell
1 + 1
2 + 2
# -
# This cell is a cell with implicit start
1 + 1
""", ):
nb = jupytext.reads(pynb, "py")
assert len(nb.cells) == 2
assert nb.cells[0].cell_type == "code"
assert (nb.cells[0].source ==
"""# This is a code cell with explicit end of cell
1 + 1
2 + 2""")
assert nb.cells[1].cell_type == "code"
assert (nb.cells[1].source == """# This cell is a cell with implicit start
1 + 1""")
pynb2 = jupytext.writes(nb, "py")
compare(pynb2, pynb)
def test_escape_start_pattern(pynb="""# The code start pattern '# +' can
# appear in code and markdown cells.
# In markdown cells it is escaped like here:
# # + {"sample_metadata": "value"}
# In code cells like this one, it is also escaped
# # + {"sample_metadata": "value"}
1 + 1
""", ):
nb = jupytext.reads(pynb, "py")
assert len(nb.cells) == 3
assert nb.cells[0].cell_type == "markdown"
assert nb.cells[1].cell_type == "markdown"
assert nb.cells[2].cell_type == "code"
assert (nb.cells[1].source == """In markdown cells it is escaped like here:
# + {"sample_metadata": "value"}""")
assert (nb.cells[2].source ==
"""# In code cells like this one, it is also escaped
# + {"sample_metadata": "value"}
1 + 1""")
pynb2 = jupytext.writes(nb, "py")
compare(pynb2, pynb)
def test_escape_start_pattern(ext, rnb="""#' The code start pattern '#+' can
#' appear in code and markdown cells.
#' In markdown cells it is escaped like here:
#' #+ fig.width=12
# In code cells like this one, it is also escaped
# #+ cell_name language="python"
1 + 1
"""):
nb = jupytext.reads(rnb, ext)
assert len(nb.cells) == 3
assert nb.cells[0].cell_type == 'markdown'
assert nb.cells[1].cell_type == 'markdown'
assert nb.cells[2].cell_type == 'code'
assert nb.cells[1].source == '''In markdown cells it is escaped like here:
#+ fig.width=12'''
assert (nb.cells[2].source ==
'''# In code cells like this one, it is also escaped
#+ cell_name language="python"
1 + 1''')
rnb2 = jupytext.writes(nb, ext)
compare(rnb, rnb2)
def test_read_simple_file(text="""println!("Hello world");
eprintln!("Hello error");
format!("Hello {}", "world")
// A Function
pub fn fib(x: i32) -> i32 {
if x <= 2 {0} else {fib(x - 2) + fib(x - 1)}
}
// This is a
// Markdown cell
// This is a magic instruction
// :vars
// This is a rust identifier
::std::mem::drop
""", ):
nb = jupytext.reads(text, "rs")
compare_notebooks(
nb,
new_notebook(cells=[
new_code_cell("""println!("Hello world");
eprintln!("Hello error");
format!("Hello {}", "world")"""),
new_code_cell("""// A Function
pub fn fib(x: i32) -> i32 {
if x <= 2 {0} else {fib(x - 2) + fib(x - 1)}
}"""),
new_markdown_cell("This is a\nMarkdown cell"),
new_code_cell("""// This is a magic instruction
:vars"""),
new_code_cell("""// This is a rust identifier
::std::mem::drop"""),
]),
)
compare(jupytext.writes(nb, "rs"), text)
def test_mark_cell_with_no_title_and_inner_region(
script="""# This is a markdown cell
# region {"key": "value"}
a = 1
# region An inner region
b = 2
# endregion
def f(x):
return x + 1
# endregion
d = 4
""", ):
nb = jupytext.reads(script, "py")
assert nb.cells[0].cell_type == "markdown"
assert nb.cells[0].source == "This is a markdown cell"
assert nb.cells[1].cell_type == "code"
assert nb.cells[1].source == '# region {"key": "value"}\na = 1'
assert nb.cells[2].cell_type == "code"
assert nb.cells[2].metadata["title"] == "An inner region"
assert nb.cells[2].source == "b = 2"
assert nb.cells[3].cell_type == "code"
assert nb.cells[3].source == "def f(x):\n return x + 1"
assert nb.cells[4].cell_type == "code"
assert nb.cells[4].source == "# endregion"
assert nb.cells[5].cell_type == "code"
assert nb.cells[5].source == "d = 4"
assert len(nb.cells) == 6
script2 = jupytext.writes(nb, "py")
compare(script2, script)
def test_read_md_and_markdown_regions(markdown="""Some text
<!-- #md -->
A
long
cell
<!-- #endmd -->
<!-- #markdown -->
Another
long
cell
<!-- #endmarkdown -->
"""):
nb = jupytext.reads(markdown, 'md')
assert nb.metadata['jupytext']['main_language'] == 'python'
compare(nb.cells, [
new_markdown_cell('Some text'),
new_markdown_cell("""A
long
cell""", metadata={'region_name': 'md'}),
new_markdown_cell("""Another
long
cell""",
metadata={'region_name': 'markdown'})
])
markdown2 = jupytext.writes(nb, 'md')
compare(markdown, markdown2)
def test_mark_cell_with_no_title_and_inner_region(
script="""# This is a markdown cell
# region {"key": "value"}
a = 1
# region An inner region
b = 2
# endregion
def f(x):
return x + 1
# endregion
d = 4
"""):
nb = jupytext.reads(script, 'py')
assert nb.cells[0].cell_type == 'markdown'
assert nb.cells[0].source == 'This is a markdown cell'
assert nb.cells[1].cell_type == 'code'
assert nb.cells[1].source == '# region {"key": "value"}\na = 1'
assert nb.cells[2].cell_type == 'code'
assert nb.cells[2].metadata['title'] == 'An inner region'
assert nb.cells[2].source == 'b = 2'
assert nb.cells[3].cell_type == 'code'
assert nb.cells[3].source == 'def f(x):\n return x + 1'
assert nb.cells[4].cell_type == 'code'
assert nb.cells[4].source == '# endregion'
assert nb.cells[5].cell_type == 'code'
assert nb.cells[5].source == 'd = 4'
assert len(nb.cells) == 6
script2 = jupytext.writes(nb, 'py')
compare(script, script2)
def test_read_less_simple_file(pynb="""# ---
# title: Less simple file
# ---
# Here we have some text
# And below we have some python code
# This is a comment about function f
def f(x):
return x+1
# And a comment on h
def h(y):
return y-1
""", ):
nb = jupytext.reads(pynb, "py")
assert len(nb.cells) == 4
assert nb.cells[0].cell_type == "raw"
assert nb.cells[0].source == "---\ntitle: Less simple file\n---"
assert nb.cells[1].cell_type == "markdown"
assert (nb.cells[1].source == "Here we have some text\n"
"And below we have some python code")
assert nb.cells[2].cell_type == "code"
compare(
nb.cells[2].source,
"# This is a comment about function f\n"
"def f(x):\n"
" return x+1",
)
assert nb.cells[3].cell_type == "code"
compare(nb.cells[3].source,
"""# And a comment on h\ndef h(y):\n return y-1""")
pynb2 = jupytext.writes(nb, "py")
compare(pynb2, pynb)
def remove_solution(input_py_str):
"""Removes solution from python scripts content str.
This is based on:
- cells having "solution" in their metadata tags when removing full cells
- a specific comment matching "# solution" that will keep only the content
before this comment and add "# Write your code here." at the end of the
cell.
"""
nb = jupytext.reads(input_py_str, fmt='py:percent')
cell_tags_list = [c['metadata'].get('tags') for c in nb.cells]
is_solution_list = [
tags is not None and 'solution' in tags for tags in cell_tags_list
]
# Completely remove cells with "solution" tags
nb.cells = [
cell for cell, is_solution in zip(nb.cells, is_solution_list)
if not is_solution
]
# Partial cell removal based on "# solution" comment
marker = "# solution"
pattern = re.compile(f"^{marker}.*", flags=re.MULTILINE | re.DOTALL)
cells_to_modify = [
c for c in nb.cells
if c["cell_type"] == "code" and marker in c["source"]
]
for c in cells_to_modify:
c["source"] = pattern.sub("# Write your code here.", c["source"])
# TODO: we could potentially try to avoid changing the input file jupytext
# header since this info is rarely useful. Let's keep it simple for now.
py_nb_str = jupytext.writes(nb, fmt='py:percent')
return py_nb_str
def test_read_markdown_IDL(
no_jupytext_version_number,
text="""---
jupyter:
kernelspec:
display_name: IDL [conda env:gdl] *
language: IDL
name: conda-env-gdl-idl
---
# A sample IDL Markdown notebook
```IDL
a = 1
```
""",
):
nb = jupytext.reads(text, "md")
assert len(nb.cells) == 2
assert nb.cells[1].cell_type == "code"
assert nb.cells[1].source == "a = 1"
text2 = jupytext.writes(nb, "md")
compare(text2, text)
def test_read_simple_file(
script="""// ---
// title: Simple file
// ---
// Here we have some text
// And below we have some code
System.out.println("Hello World");
""",
):
nb = jupytext.reads(script, "java")
assert len(nb.cells) == 3
assert nb.cells[0].cell_type == "raw"
assert nb.cells[0].source == "---\ntitle: Simple file\n---"
assert nb.cells[1].cell_type == "markdown"
assert (
nb.cells[1].source == "Here we have some text\n" "And below we have some code"
)
assert nb.cells[2].cell_type == "code"
assert nb.cells[2].source == """System.out.println("Hello World");"""
script2 = jupytext.writes(nb, "java")
compare(script2, script)
def test_ipython_help_are_commented_297(
text="""# This is a markdown cell
# that ends with a question: float?
# The next cell is also a markdown cell,
# because it has no code marker:
# float?
# +
# float?
# +
# float??
# +
# Finally a question in a code
# # cell?
""",
nb=new_notebook(cells=[
new_markdown_cell(
"This is a markdown cell\nthat ends with a question: float?"),
new_markdown_cell(
"The next cell is also a markdown cell,\nbecause it has no code marker:"
),
new_markdown_cell("float?"),
new_code_cell("float?"),
new_code_cell("float??"),
new_code_cell("# Finally a question in a code\n# cell?"),
]),
):
nb2 = jupytext.reads(text, "py")
compare_notebooks(nb2, nb)
text2 = jupytext.writes(nb2, "py")
compare(text2, text)
def test_mark_cell_with_vim_folding_markers(script="""# This is a markdown cell
# {{{ And this is a foldable code region with metadata {"key": "value"}
a = 1
b = 2
c = 3
# }}}
""", ):
nb = jupytext.reads(script, "py")
assert nb.metadata["jupytext"]["cell_markers"] == "{{{,}}}"
assert len(nb.cells) == 2
assert nb.cells[0].cell_type == "markdown"
assert nb.cells[0].source == "This is a markdown cell"
assert nb.cells[1].cell_type == "code"
assert nb.cells[1].source == "a = 1\n\nb = 2\n\nc = 3"
assert nb.cells[1].metadata == {
"title": "And this is a foldable code region with metadata",
"key": "value",
}
script2 = jupytext.writes(nb, "py")
compare(script2, script)
def read(cls, path):
resolved_path, fmt = cls.extract_format(path)
text = _resolved_handler(resolved_path).read(resolved_path)
# Read the document
nb = jupytext.reads(text, fmt=fmt)
# Set a kernel if there was none
if nb.metadata.get('kernelspec', {}).get('name') is None:
language = nb.metadata.get('jupytext', {}).get(
'main_language') or nb.metadata['kernelspec']['language']
if not language:
raise ValueError(
'Cannot infer a kernel as the document language is not defined'
)
kernelspec = kernelspec_from_language(language)
if not kernelspec:
raise ValueError('Found no kernel for {}'.format(language))
nb.metadata['kernelspec'] = kernelspec
# Return the notebook as a JSON string
return nbformat.writes(nb)
def test_read_simple_file(pynb="""# ---
# title: Simple file
# ---
# Here we have some text
# And below we have some python code
def f(x):
return x+1
def h(y):
return y-1
""", ):
nb = jupytext.reads(pynb, "py")
assert len(nb.cells) == 4
assert nb.cells[0].cell_type == "raw"
assert nb.cells[0].source == "---\ntitle: Simple file\n---"
assert nb.cells[1].cell_type == "markdown"
assert (nb.cells[1].source == "Here we have some text\n"
"And below we have some python code")
assert nb.cells[2].cell_type == "code"
compare(
nb.cells[2].source,
"""def f(x):
return x+1""",
)
assert nb.cells[3].cell_type == "code"
compare(
nb.cells[3].source,
"""def h(y):
return y-1""",
)
pynb2 = jupytext.writes(nb, "py")
compare(pynb2, pynb)
def test_fix_139():
text = """# ---
# jupyter:
# jupytext:
# metadata_filter:
# cells:
# additional:
# - "lines_to_next_cell"
# excluded:
# - "all"
# ---
# + {"lines_to_next_cell": 2}
1 + 1
# -
1 + 1
"""
nb = jupytext.reads(text, 'py:light')
text2 = jupytext.writes(nb, 'py:light')
assert 'cell_metadata_filter: -all' in text2
assert 'lines_to_next_cell' not in text2
def test_docstring_with_quadruple_quote(
nb=new_notebook(
cells=[
new_code_cell(
'''def fun_1(df):
""""
docstring starting with 4 double quotes and ending with 3
"""
return df'''
),
new_code_cell(
'''def fun_2(df):
"""
docstring
"""
return df'''
),
]
)
):
"""Reproduces https://github.com/mwouts/jupytext/issues/460"""
py = jupytext.writes(nb, "py:percent")
nb2 = jupytext.reads(py, "py")
compare_notebooks(nb2, nb)
def test_notebook_with_python3_magic(
no_jupytext_version_number,
nb=new_notebook(metadata={
'kernelspec': {
'display_name': 'Python 3',
'language': 'python',
'name': 'python3'
}
},
cells=[
new_code_cell('%%python2\na = 1\nprint a'),
new_code_cell('%%python3\nb = 2\nprint(b)')
]),
text="""---
jupyter:
kernelspec:
display_name: Python 3
language: python
name: python3
---
```python2
a = 1
print a
```
```python3
b = 2
print(b)
```
"""):
md = jupytext.writes(nb, 'md')
compare(md, text)
nb2 = jupytext.reads(md, 'md')
compare_notebooks(nb2, nb)
def test_read_write_script_with_metadata_241(
no_jupytext_version_number,
rsnb="""#!/usr/bin/env scriptisto
// ---
// jupyter:
// jupytext:
// text_representation:
// extension: .rs
// format_name: light
// kernelspec:
// display_name: Rust
// language: rust
// name: rust
// ---
let mut a: i32 = 2;
a += 1;
""",
):
nb = jupytext.reads(rsnb, "rs")
assert "executable" in nb.metadata["jupytext"]
rsnb2 = jupytext.writes(nb, "rs")
compare(rsnb, rsnb2)
def test_mark_cell_with_vim_folding_markers(script="""# This is a markdown cell
# {{{ And this is a foldable code region with metadata {"key": "value"}
a = 1
b = 2
c = 3
# }}}
"""):
nb = jupytext.reads(script, 'py')
assert nb.metadata['jupytext']['cell_markers'] == '{{{,}}}'
assert len(nb.cells) == 2
assert nb.cells[0].cell_type == 'markdown'
assert nb.cells[0].source == 'This is a markdown cell'
assert nb.cells[1].cell_type == 'code'
assert nb.cells[1].source == 'a = 1\n\nb = 2\n\nc = 3'
assert nb.cells[1].metadata == {
'title': 'And this is a foldable code region with metadata',
'key': 'value'
}
script2 = jupytext.writes(nb, 'py')
compare(script, script2)
def test_read_write_script_with_metadata_241(no_jupytext_version_number, pynb="""#!/usr/bin/env python3
# -*- coding: utf-8 -*-
# ---
# jupyter:
# jupytext:
# text_representation:
# extension: .py
# format_name: light
# kernelspec:
# display_name: Python 3
# language: python
# name: python3
# ---
a = 2
a + 1
"""):
nb = jupytext.reads(pynb, 'py')
assert 'executable' in nb.metadata['jupytext']
assert 'encoding' in nb.metadata['jupytext']
pynb2 = jupytext.writes(nb, 'py')
compare(pynb2, pynb)
