Removed dependency on isort

[fpgmaas]

This PR removes the dependency on isort to get the standard library. The code to create the standard library files is borrowed from isort, which I think is permitted by their license. I did at a disclaimer to credit the project to the mkstdlibs.py file.

I have chosen to not use the built-in function for Python 3.10, but to eep the approach consistent for the various supported Python versions.

PR Checklist

• A description of the changes is added to the description of this PR.
• If there is a related issue, make sure it is linked to this PR.
• If you've fixed a bug or added code that should be tested, add tests!
• Documentation in docs is updated

Description of changes

[codecov-commenter]

Codecov Report

Merging #140 (77c9658) into main (ce81ea3) will decrease coverage by 0.0%.
The diff coverage is 40.0%.

@@           Coverage Diff           @@
##            main    #140     +/-   ##
=======================================
- Coverage   95.0%   94.9%   -0.1%     
=======================================
  Files         35      36      +1     
  Lines       1180    1176      -4     
=======================================
- Hits        1121    1117      -4     
  Misses        59      59             

Impacted Files	Coverage Δ
deptry/module.py	89.6% <25.0%> (-0.7%)	⬇️
deptry/stdlibs/py39.py	100.0% <100.0%> (ø)

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

[mkniewallner]

Since the script from isort relies on sphinx that we don't use in the project, we might want to use another source of truth for the modules.

https://docs.python.org/3/py-modindex.html also lists all the stdlib modules, so based on that, we could write the following script (that I also renamed to generate_stdlibs.py since it's IMO clearer than mkstdlibs.py):

#!/usr/bin/env python3

# This script is inspired by isort: https://github.com/PyCQA/isort/blob/4ccbd1eddf564d2c9e79c59d59c1fc06a7e35f94/scripts/mkstdlibs.py.

import urllib.request
from html.parser import HTMLParser
from typing import List
from typing import Optional
from typing import Tuple


OUTPUT_PATH = "deptry/stdlibs/py{}.py"
STDLIB_MODULES_URL = "https://docs.python.org/{}/py-modindex.html"
PYTHON_VERSIONS = (("3", "7"), ("3", "8"), ("3", "9"), ("3", "10"))

# Modules that are in stdlib, but undocumented.
EXTRA_STDLIBS_MODULES = ("_ast", "ntpath", "posixpath", "sre", "sre_constants", "sre_compile", "sre_parse")

DOCSTRING_GENERATED_FILES = """
DO NOT EDIT THIS FILE MANUALLY.
It is generated from `scripts/generate_stdlibs.py` script and contains the stdlib modules for Python {}.
You can generate it again using `poetry run scripts/generate_stdlibs.py`.
"""


class PythonStdlibHTMLParser(HTMLParser):
    def __init__(self) -> None:
        super().__init__()
        self._is_in_code_tag = False
        self.modules: List[str] = []

    def handle_starttag(self, tag: str, attrs: List[Tuple[str, Optional[str]]]) -> None:
        if tag == "code":
            self._is_in_code_tag = True

    def handle_endtag(self, tag: str) -> None:
        if tag == "code":
            self._is_in_code_tag = False

    def handle_data(self, data: str) -> None:
        if self._is_in_code_tag:
            self.modules.append(data)


for python_version in PYTHON_VERSIONS:
    dotted_python_version = ".".join(python_version)

    with urllib.request.urlopen(STDLIB_MODULES_URL.format(dotted_python_version)) as response:
        html_content = response.read().decode()

    parser = PythonStdlibHTMLParser()
    parser.feed(html_content)

    modules = {module.split(".")[0] for module in parser.modules}.union(EXTRA_STDLIBS_MODULES)
    modules.remove("__main__")

    with open(OUTPUT_PATH.format("".join(python_version)), "w") as python_stdlib_file:
        python_stdlib_file.write(f'"""{DOCSTRING_GENERATED_FILES.format(dotted_python_version)}"""\n\n')
        python_stdlib_file.write("stdlib = {\n")

        for module in sorted(modules):
            python_stdlib_file.write(f'    "{module}",\n')
        python_stdlib_file.write("}\n")

The slight difference is that this will add __future__, but since it is in the stdlib, I don't see any reason for excluding it.

[fpgmaas]

We could have also addedSphinx as a separate dependency group, e.g. under tool.poetry.group.scripts. But I agree with you that less dependencies is better, so I have changed the PR according to your suggested script.

P.S. You might want to open a PR in isort as well, since I don't think they have Sphinx in their environment either: link.