def check_grammar()

in cdk-project/lib/images/codebuild-image/python/src/notebooks/lint.py [0:0]

• 29 lines of code
• 14 McCabe index (conditional complexity)

def check_grammar(notebook):
    """Run LanguageTool against the given notebook.

    Args:
        notebook (Path): The notebook filename to run the formatting check against.

    Returns:
        [language_tool_python.Match]: A list of spelling and grammar rule violations found in the notebook.

    """
    tool = language_tool_python.LanguageTool("en-US")

    report = []

    cells = parse.markdown_cells(notebook)
    for cell in cells:
        code_block = False
        for line in cell:
            stripped_line = line.rstrip().strip(" #*")
            if stripped_line in ("```python", "```bash", "```"):
                code_block = not code_block
            if code_block:
                continue
            code_substituted_line = re.sub(
                "(`)\1{2,}[^`]*(`)\1{2,}|`[^`]*`", "[code]", stripped_line
            )
            # Strip down any anchor tags or URL markdown from the text before proceeding
            code_substituted_line = re.sub('<a.*?>|</a>', '', code_substituted_line)
            code_substituted_line = re.sub(r'\[(.*?)\]\((.*?)\)',r'\1',code_substituted_line)

            matches = tool.check(code_substituted_line)
            report.extend(matches)

    is_correctly_spelled = lambda rule: rule.ruleIssueType == "misspelling" and (
        rule.matchedText in dictionary.allow_list
        or "-" in rule.matchedText
        or "_" in rule.matchedText
        or "$" in rule.matchedText
    )
    report = [rule for rule in report if not is_correctly_spelled(rule)]

    is_ignored_rule = lambda rule: rule.ruleId in dictionary.rules_to_ignore
    report = [rule for rule in report if not is_ignored_rule(rule)]

    return report