Use pre-commit (#396)

* Add a basic pre-commit file

* Run `pre-commit run -a`

* Add pyupgrade as a pre-commit hook

* Run `pre-commit run -a`

* Add black as a pre-commit hook

* Run `pre-commit run -a`

* Add isort as a pre-commit hook

* Run `pre-commit run -a`

* Add flake8 as a pre-commit hook

* Manually fix all flake8 errors

* Add actionlint as a pre-commit hook

* Manually fix errors reported by actionlint

* Add Kurt McKee to the list of contributors

Author: kurtmckee

.flake8

@@ -0,0 +1,4 @@
+[flake8]
+max-line-length = 80
+extend-select = B950
+extend-ignore = E203,E501,E701

.github/workflows/ci.yaml

@@ -10,7 +10,7 @@ defaults:
 concurrency:
   group: ci-tests-${{ github.ref }}-1
   cancel-in-progress: true
-  
+
 jobs:
   test:
     strategy:
@@ -40,7 +40,7 @@ jobs:
         id: cached-poetry-dependencies
         with:
           path: .venv
-          key: venv-${{ runner.os }}-${{ runner.arch }}-${{ steps.full-python-version.outputs.version }}-${{ hashFiles('**/poetry.lock') }}
+          key: venv-${{ runner.os }}-${{ runner.arch }}-${{ steps.setup-python.outputs.python-version }}-${{ hashFiles('**/poetry.lock') }}
       - name: Install dependencies
         run: poetry install
         if: steps.cached-poetry-dependencies.outputs.cache-hit != 'true'
@@ -95,7 +95,7 @@ jobs:
           # - pandoc does not publish binaries for Linux 32bit
           CIBW_ARCHS_LINUX: "auto64 aarch64"
           CIBW_ARCHS_MACOS: "x86_64 arm64"
-          CIBW_ARCHS_WINDOWS: "AMD64" 
+          CIBW_ARCHS_WINDOWS: "AMD64"
       - name: Upload artifacts
         uses: actions/upload-artifact@v4
         with:

.pre-commit-config.yaml

@@ -0,0 +1,53 @@
+ci:
+  autoupdate_schedule: "quarterly"
+
+default_language_version:
+  python: "python3.13"
+
+repos:
+  - repo: "meta"
+    hooks:
+      - id: "check-hooks-apply"
+      - id: "check-useless-excludes"
+
+  - repo: "https://github.com/pre-commit/pre-commit-hooks"
+    rev: "v5.0.0"
+    hooks:
+      - id: "check-added-large-files"
+      - id: "check-merge-conflict"
+      - id: "check-yaml"
+      - id: "end-of-file-fixer"
+      - id: "mixed-line-ending"
+        args:
+          - "--fix=lf"
+      - id: "trailing-whitespace"
+
+  - repo: "https://github.com/asottile/pyupgrade"
+    rev: "v3.19.1"
+    hooks:
+      - id: "pyupgrade"
+        name: "Enforce Python 3.7+ idioms"
+        args:
+          - "--py37-plus"
+
+  - repo: "https://github.com/psf/black-pre-commit-mirror"
+    rev: "25.1.0"
+    hooks:
+      - id: "black"
+
+  - repo: "https://github.com/pycqa/isort"
+    rev: "6.0.0"
+    hooks:
+      - id: "isort"
+
+  - repo: "https://github.com/pycqa/flake8"
+    rev: "7.1.1"
+    hooks:
+      - id: "flake8"
+        additional_dependencies:
+          - "flake8-bugbear==24.12.12"
+
+  - repo: "https://github.com/rhysd/actionlint"
+    rev: "v1.7.7"
+    hooks:
+      - id: "actionlint"

Dockerfile

@@ -12,7 +12,7 @@ RUN apt update && apt upgrade -y && apt install pandoc -y
 # Update pip
 RUN pip install --upgrade pip
 
-# Copy the files to container   
+# Copy the files to container
 COPY . pypandoc
 WORKDIR pypandoc
 

README.md

@@ -237,10 +237,10 @@ it won't work. This gotcha has to do with the way
 ## Logging Messages
 
 Pypandoc logs messages using the [Python logging library](https://docs.python.org/3/library/logging.html).
-By default, it will send messages to the console, including any messages 
+By default, it will send messages to the console, including any messages
 generated by Pandoc. If desired, this behaviour can be changed by adding
-[handlers](https://docs.python.org/3/library/logging.html#handler-objects) to 
-the pypandoc logger **before calling any functions**. For example, to mute all 
+[handlers](https://docs.python.org/3/library/logging.html#handler-objects) to
+the pypandoc logger **before calling any functions**. For example, to mute all
 logging add a [null handler](https://docs.python.org/3/library/logging.handlers.html#nullhandler):
 
 ```python
@@ -313,6 +313,7 @@ Note that for citeproc tests to pass you'll need to have [pandoc-citeproc](https
 * [Juho Vepsäläinen](https://github.com/bebraw/) - Creator and former maintainer of pypandoc
 * [Connor](https://github.com/DisSupEng/) - Updated Dockerfile to Python 3.9 image and added docker compose file
 * [Colin Bull](https://github.com/colinbull) - Added ability to control whether files are sorted before being passed to pandoc process.
+* [Kurt McKee](https://github.com/kurtmckee) - Project infrastructure improvements
 
 ## License
 

examples/services.py

@@ -1,16 +1,17 @@
-# -*- coding: utf-8 -*-
 """
 Example Services for using pypandoc
 """
+
 from tempfile import NamedTemporaryFile
 
 import pypandoc
 
 
-class BasePandocService(object):
+class BasePandocService:
     """
     Base class for converting provided HTML to a doc or docx
     """
+
     file_object = None
 
     def __init__(self):
@@ -27,23 +28,22 @@ class PandocPDFService(BasePandocService):
     """
     Generate html to pdf format
     """
+
     def generate(self, html, **kwargs):
         """
         generate the pdf but needs to be set as tex so pandoc handles it
         correctly see docs: http://johnmacfarlane.net/pandoc/ #search pdf
         """
-        from_format = kwargs.get('from_format', 'html')
-        to_format = kwargs.get('to_format', 'tex')
+        from_format = kwargs.get("from_format", "html")
+        to_format = kwargs.get("to_format", "tex")
         # create temp file
-        self.file_object = NamedTemporaryFile(suffix='.pdf')
+        self.file_object = NamedTemporaryFile(suffix=".pdf")
 
-        extra_args = (
-            '--smart',
-            '--standalone',
-            '-o', self.file_object.name
-        )
+        extra_args = ("--smart", "--standalone", "-o", self.file_object.name)
         # generate it using pandoc
-        self.service.convert_text(html, to_format, format=from_format, extra_args=extra_args)
+        self.service.convert_text(
+            html, to_format, format=from_format, extra_args=extra_args
+        )
         # return the file which is now populated with the docx forms
         return self.file_object
 
@@ -52,18 +52,17 @@ class PandocDocxService(BasePandocService):
     """
     Generate html to docx format
     """
+
     def generate(self, html, **kwargs):
-        from_format = kwargs.get('from_format', 'html')
-        to_format = kwargs.get('to_format', 'docx')
+        from_format = kwargs.get("from_format", "html")
+        to_format = kwargs.get("to_format", "docx")
         # create temp file
-        self.file_object = NamedTemporaryFile(suffix='.docx')
+        self.file_object = NamedTemporaryFile(suffix=".docx")
 
-        extra_args = (
-            '--smart',
-            '--standalone',
-            '-o', self.file_object.name
-        )
+        extra_args = ("--smart", "--standalone", "-o", self.file_object.name)
         # generate it using pandoc
-        self.service.convert_text(html, to_format, format=from_format, extra_args=extra_args)
+        self.service.convert_text(
+            html, to_format, format=from_format, extra_args=extra_args
+        )
         # return the file which is now populated with the docx forms
         return self.file_object