Initial implementation

- Package set up
- docxplain command-line entrypoint that is usable as a pre-commit hook.
- Integration with pandoc for document conversion
- Hashing of the plain text artifact to determine if it has changed.

Author: jonathansick

.pre-commit-config.yaml

@@ -0,0 +1,38 @@
+repos:
+  - repo: https://github.com/pre-commit/pre-commit-hooks
+    rev: v3.4.0
+    hooks:
+      - id: check-yaml
+      - id: check-toml
+      - id: check-json
+      - id: trailing-whitespace
+
+  - repo: https://github.com/Lucas-C/pre-commit-hooks-markup
+    rev: v1.0.1
+    hooks:
+      - id: rst-linter
+        files: (README\.rst)|(CHANGELOG\.rst)
+
+  - repo: https://github.com/PyCQA/isort
+    rev: 5.7.0
+    hooks:
+      - id: isort
+        additional_dependencies:
+          - toml
+
+  - repo: https://github.com/psf/black
+    rev: 20.8b1
+    hooks:
+      - id: black
+
+  - repo: https://github.com/asottile/blacken-docs
+    rev: v1.10.0
+    hooks:
+      - id: blacken-docs
+        additional_dependencies: [black==20.8b1]
+        args: [-l, "79", -t, py38]
+
+  - repo: https://gitlab.com/pycqa/flake8
+    rev: 3.9.0
+    hooks:
+      - id: flake8

CHANGELOG.rst

@@ -0,0 +1,7 @@
+Change log
+==========
+
+Unreleased
+----------
+
+Initial release.

README.md README.rstREADME.md renamed to README.rst

@@ -1,2 +1,5 @@
-# pre-commit-docx-plain
+#####################
+pre-commit-docx-plain
+#####################
+
 Pre-commit hook for converting Office (docx) files into plain text (using Pandoc).

pyproject.toml

@@ -0,0 +1,33 @@
+[build-system]
+requires = [
+    "setuptools>=42",
+    "wheel",
+    "setuptools_scm[toml]>=3.4"
+]
+build-backend = 'setuptools.build_meta'
+
+[tool.setuptools_scm]
+
+[tool.black]
+line-length = 79
+target-version = ['py38']
+exclude = '''
+/(
+    \.eggs
+  | \.git
+  | \.mypy_cache
+  | \.tox
+  | \.venv
+  | _build
+  | build
+  | dist
+)/
+'''
+# Use single-quoted strings so TOML treats the string like a Python r-string
+# Multi-line strings are implicitly treated by black as regular expressions
+
+[tool.isort]
+include_trailing_comma = true
+multi_line_output = 3
+known_first_party = "docxplain"
+skip = ["docs/conf.py"]

setup.cfg

@@ -0,0 +1,63 @@
+[metadata]
+name = docxplain
+description = Convert Office (docx) files to plain text using pandoc.
+author = Jonathan Sick
+author_email = hi@jsick.codes
+long_description = file: README.rst, CHANGELOG.rst, LICENSE
+long_description_content_type = text/x-rst
+url = https://github.com/jsickcodes/pre-commit-docx-plain
+project_urls =
+    Change log = https://github.com/jsickcodes/pre-commit-docx-plain/blob/master/CHANGELOG.rst
+    Source code = https://github.com/jsickcodes/pre-commit-docx-plain
+    Issue tracker = https://github.com/jsickcodes/pre-commit-docx-plain/issues
+classifiers =
+    Development Status :: 4 - Beta
+    License :: OSI Approved :: MIT License
+    Programming Language :: Python
+    Programming Language :: Python :: 3
+    Intended Audience :: Developers
+    Natural Language :: English
+    Operating System :: POSIX
+keywords =
+    precommit
+
+[options]
+zip_safe = False
+include_package_data = True
+package_dir =
+    = src
+packages = find:
+python_requires = >=3.7
+setup_requires =
+    setuptools_scm
+install_requires =
+    importlib_metadata; python_version < "3.8"
+    pypandoc
+    py-pandoc
+
+[options.packages.find]
+where = src
+
+[options.entry_points]
+console_scripts =
+    docxplain = docxplain.cli:main
+
+[options.extras_require]
+dev =
+    pytest
+
+[flake8]
+max-line-length = 79
+# E203: whitespace before :, flake8 disagrees with PEP 8
+# W503: line break after binary operator, flake8 disagrees with PEP 8
+ignore = E203, W503
+
+[mypy]
+disallow_untyped_defs = True
+disallow_incomplete_defs = True
+ignore_missing_imports = True
+show_error_codes = True
+strict_equality = True
+warn_redundant_casts = True
+warn_unreachable = True
+warn_unused_ignores = True

setup.py

@@ -0,0 +1,3 @@
+from setuptools import setup
+
+setup(use_scm_version=True)

src/docxplain/__init__.py

@@ -0,0 +1,24 @@
+from __future__ import annotations
+
+import argparse
+import sys
+
+from docxplain.converter import convert_file
+
+
+def main() -> None:
+    """Command-line entrypoint."""
+    parser = create_parser()
+    args = parser.parse_args()
+    changed = convert_file(args.source)
+    if changed:
+        sys.exit(1)
+    else:
+        sys.exit(0)
+
+
+def create_parser() -> argparse.ArgumentParser:
+    parser = argparse.ArgumentParser(description="Convert docx to plain text.")
+    parser.add_argument("source")
+
+    return parser

src/docxplain/cli.py

@@ -0,0 +1,43 @@
+from __future__ import annotations
+
+import hashlib
+from pathlib import Path
+
+import pypandoc
+
+__all__ = ["convert_file", "get_hash"]
+
+
+def convert_file(filename: str) -> bool:
+    """Convert the docx file to plaintext.
+
+    Returns
+    -------
+    changed : bool
+        True if the converted file is different
+    """
+    docx_path = Path(filename)
+    if not docx_path.is_file():
+        raise RuntimeError(f"Source file {docx_path} does not exist.")
+
+    plain_path = docx_path.with_suffix(".txt")
+    if plain_path.is_file():
+        exists = True
+        initial_hash = get_hash(plain_path)
+    else:
+        exists = False
+
+    pypandoc.convert_file(str(docx_path), "plain", outputfile=str(plain_path))
+
+    if exists:
+        final_hash = get_hash(plain_path)
+        return final_hash != initial_hash
+    else:
+        return True
+
+
+def get_hash(path: Path) -> str:
+    """Get the SHA256 hash diget of a file."""
+    m = hashlib.sha256()
+    m.update(path.read_bytes())
+    return m.hexdigest()

src/docxplain/converter.py

@@ -0,0 +1,33 @@
+[tox]
+envlist =
+    typing
+    lint
+    packaging
+isolated_build = True
+skip_missing_interpreters = True
+
+[testenv:lint]
+description = Lint codebase by running pre-commit (Black, isort, Flake8).
+skip_install = true
+deps =
+    pre-commit
+commands = pre-commit run --all-files
+
+[testenv:typing]
+description = Run mypy.
+deps =
+    mypy
+commands =
+    mypy src tests setup.py
+
+[testenv:packaging]
+description = Check packaging for PyPI with twine
+skip_install = true
+allowlist_externals =
+    rm
+deps =
+    twine
+commands =
+    rm -rf dist
+    python setup.py sdist bdist_wheel
+    twine check dist/*