Fix tag parsing bugs, and bring the project setup up to date

.codespellrc

@@ -0,0 +1,4 @@
+[codespell]
+# 'sav' is the SPSS file extension, used by read_spss()
+ignore-words-list = sav
+skip = ./.git,./.venv,./site,./uv.lock,./.ruff_cache,./.pytest_cache,./*.egg-info,./src/*.egg-info,./tests/fixtures/encoding,*.xlsx,*.feather,*.parquet,*.orc,*.dta,*.xpt,*.h5

.github/workflows/documentation.yml

@@ -0,0 +1,27 @@
+name: Deploy documentation
+
+on:
+  push:
+    branches: [master]
+  workflow_dispatch:
+
+permissions:
+  contents: write
+
+jobs:
+  deploy:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+      with:
+        # git-authors and git-revision-date-localized need the full history
+        fetch-depth: 0
+
+    - name: Install uv
+      uses: astral-sh/setup-uv@v5
+
+    - name: Deploy to GitHub Pages
+      run: |
+        git config user.name "github-actions[bot]"
+        git config user.email "41898282+github-actions[bot]@users.noreply.github.com"
+        uv run mkdocs gh-deploy --force

.github/workflows/workflow.yml

@@ -4,11 +4,11 @@ jobs:
   run:
     runs-on: ${{ matrix.os }}
     strategy:
-      matrix: 
+      matrix:
         os: [ubuntu-latest, macos-latest, windows-latest]
         python-version: ["3.10", "3.11", "3.12", "3.13", "3.14"]
     steps:
-    - uses: actions/checkout@master
+    - uses: actions/checkout@v4
 
     - name: Install uv and set the python version
       uses: astral-sh/setup-uv@v5
@@ -19,11 +19,12 @@ jobs:
       run: |
         uv run pytest --cov=mkdocs_table_reader_plugin --cov-report=xml
 
+    # Upload once per run, from a single matrix job
     - name: Upload coverage to Codecov
-      if: contains(env.USING_COVERAGE, matrix.python-version) && github.ref == 'refs/heads/master'
-      uses: codecov/codecov-action@v4
+      if: matrix.os == 'ubuntu-latest' && matrix.python-version == '3.13' && github.ref == 'refs/heads/master'
+      uses: codecov/codecov-action@v5
       with:
         token: ${{ secrets.CODECOV_TOKEN }}
-        file: ./coverage.xml
+        files: ./coverage.xml
         flags: unittests
         fail_ci_if_error: true

CHANGELOG.md

@@ -0,0 +1,18 @@
+# Changelog
+
+Releases before 4.0.1 are described in the [GitHub releases](https://github.com/timvink/mkdocs-table-reader-plugin/releases).
+
+## 4.0.1
+
+Bug fixes:
+
+- Two reader tags on the same line are now two tables. The tag pattern matched greedily, so it swallowed everything between the first and the last tag on a line, and then failed to parse the result.
+- An `=` inside an argument value no longer breaks parsing. `{{ read_csv('a=b.csv') }}`, `sep='='` and `na_values=['a=1']` all raised a `SyntaxError` before.
+- A comma inside a dict argument no longer breaks parsing, so `dtype={'a': 'str', 'b': 'int'}` works. Lists and tuples already worked.
+- `convert_to_md_table()` no longer escapes pipe characters in the DataFrame you pass in. When a `mkdocs-macros-plugin` user rendered the same DataFrame twice, the second table showed doubly escaped pipes.
+
+Project maintenance:
+
+- The documentation site now deploys on every push to `master`. It was deployed by hand, and had fallen behind the readers added in 4.0.0.
+- Coverage is uploaded to Codecov again. The upload step tested an environment variable that was never set, so it never ran.
+- `enabled` is included in `schema.json`, so editors stop flagging it as an unknown option.

CONTRIBUTING.md

@@ -7,46 +7,55 @@ Thanks for considering to contribute to this project! Some guidelines:
 - This package tries to be as simple as possible for the user (hide any complexity from the user). Options are only added when there is clear value to the majority of users.
 - When issues or pull requests are not going to be resolved or merged, they should be closed as soon as possible. This is kinder than deciding this after a long period. Our issue tracker should reflect work to be done.
 
-## Testing
+## Development setup
 
-Make sure to install an editable version before running tests:
+This project uses [uv](https://docs.astral.sh/uv/). Install the project and its development dependencies with:
 
-```python
-pip install -r tests/test_requirements.txt
-pip install -e .
-pytest --cov=mkdocs_table_reader_plugin --cov-report term-missing tests
+```bash
+uv sync
 ```
 
-If it makes sense, writing tests for your PRs is always appreciated and will help get them merged.
+## Testing
+
+Run the unit tests and the linter with:
+
+```bash
+make test
+```
 
-In addition, this project uses pyflakes for static code checking:
+Or separately:
 
-```python
-pip install pyflakes
-pyflakes tests/ mkdocs_table_reader_plugin/
+```bash
+uv run pytest --cov=mkdocs_table_reader_plugin --cov-report term-missing tests
+uv run ruff check src/ tests/
 ```
 
-#### Code Style
+If it makes sense, writing tests for your PRs is always appreciated and will help get them merged.
+
+### Code Style
 
-Make sure your code *roughly* follows [PEP-8](https://www.python.org/dev/peps/pep-0008/) and keeps things consistent with the rest of the code.
+Make sure your code *roughly* follows [PEP-8](https://www.python.org/dev/peps/pep-0008/) and keeps things consistent with the rest of the code. `ruff` is configured in `pyproject.toml` and fixes what it can automatically.
 
 We use google-style docstrings.
 
 ## Documentation
 
-They need to be deployed manually:
+Preview the documentation site locally with:
 
 ```bash
-mkdocs gh-deploy --force
+make serve_docs
 ```
 
+Every push to `master` deploys the site to GitHub Pages through the `documentation.yml` workflow. You can also deploy by hand with `make deploy_docs`.
+
 ## Release
 
-Update `setup.py`.
+1. Update `__version__` in `src/mkdocs_table_reader_plugin/__init__.py` and add an entry to `CHANGELOG.md`.
+2. Commit, then tag and push:
 
-```bash
-git tag <version>
-git push origin <version>
-```
+    ```bash
+    git tag v<version>
+    git push origin master --tags
+    ```
 
-Then manually create a github release to trigger publishing to pypi.
+3. Create a GitHub release for the tag. That triggers the `pythonpublish.yml` workflow, which runs the tests and publishes to PyPI.

Makefile

@@ -1,11 +1,13 @@
 
 setup:
-	pip install -r tests/test_requirements.txt
-	pip install -e .
+	uv sync
 
 test:
-	pyflakes tests/ mkdocs_table_reader_plugin/
-	pytest --cov=mkdocs_table_reader_plugin --cov-report term-missing tests
+	uv run ruff check src/ tests/
+	uv run pytest --cov=mkdocs_table_reader_plugin --cov-report term-missing tests
+
+serve_docs:
+	uv run mkdocs serve
 
 deploy_docs:
-	mkdocs gh-deploy --force
+	uv run mkdocs gh-deploy --force

README.md

@@ -41,7 +41,7 @@ In your markdown files you can now use:
 Where the path is relative to the location of your project's `mkdocs.yml` file, _or_ your project's `docs/` directory, _or_ the location of your markdown source file (all 3 possible locations will be searched, in that order).
 
 - There are [readers](https://timvink.github.io/mkdocs-table-reader-plugin/readers/) available for many common table formats, like `.csv`, `.fwf`, `.json`, `.xls`, `.xlsx`, `.yaml`, `.feather`, `.tsv`, `.parquet`, `.orc`, `.html` and `.xml`, as well as HDF5, SPSS, SAS and Stata files. There is also the `read_raw()` reader that will allow you to insert tables (or other content) already in markdown format.
-- `table-reader` is compatible with [`mkdocs-macros-plugin`](https://mkdocs-macros-plugin.readthedocs.io/en/latest/). This enables further automation like filtering tables or inserting directories of tables. See the documentation on [compatibility with macros plugin](howto/use_jinja2.md) for more examples.
+- `table-reader` is compatible with [`mkdocs-macros-plugin`](https://mkdocs-macros-plugin.readthedocs.io/en/latest/). This enables further automation like filtering tables or inserting directories of tables. See the documentation on [compatibility with macros plugin](https://timvink.github.io/mkdocs-table-reader-plugin/howto/use_jinja2/) for more examples.
 
 ## Documentation and how-to guides
 

docs/schema.json

@@ -15,6 +15,12 @@
           "markdownDescription": "https://timvink.github.io/mkdocs-table-reader-plugin/options/",
           "type": "object",
           "properties": {
+            "enabled": {
+              "title": "Enables you to deactivate this plugin.",
+              "markdownDescription": "https://timvink.github.io/mkdocs-table-reader-plugin/options/#enabled",
+              "type": "boolean",
+              "default": true
+            },
             "data_path": {
               "title": "Additional path to search",
               "markdownDescription": "https://timvink.github.io/mkdocs-table-reader-plugin/options/#data_path",

pyproject.toml

@@ -67,10 +67,8 @@ lint.ignore = ['D104'
         ,'E722'
         ,'D104'
         ,'E402'
-        ,"UP038" # UP038 Use `X | Y` in `isinstance` call instead of `(X, Y)`
 ]
 
-# Exclude files in tests dir
 lint.exclude = [
     ".bzr",
     ".direnv",
@@ -112,10 +110,9 @@ target-version = "py310"
 # Always autofix
 fix = true
 
-[tool.uv]
-dev-dependencies = [
+[dependency-groups]
+dev = [
     "click>=8.1.8",
-    "codecov>=2.1.13",
     # used by pd.read_html() and pd.read_xml()
     "lxml>=5",
     "mkdocs-git-authors-plugin>=0.9.4",

src/mkdocs_table_reader_plugin/__init__.py

@@ -1 +1 @@
-__version__ = "4.0.0"
+__version__ = "4.0.1"

src/mkdocs_table_reader_plugin/markdown.py

@@ -62,8 +62,9 @@ def escape(value):
             value = replace_newlines(value)
         return value
 
-    df.columns = [escape(c) for c in df.columns]
+    # Escape a copy, so that a DataFrame passed in by a macros user is left alone
     df = df.map(escape)
+    df.columns = [escape(c) for c in df.columns]
 
     return df.to_markdown(**markdown_kwargs)
 

src/mkdocs_table_reader_plugin/plugin.py

@@ -48,17 +48,17 @@ def on_config(self, config, **kwargs):
                 mkdocs_config=config, plugin_config=self.config
             )
             for reader in self.config.get("select_readers")
-            if reader in self.config.get("select_readers", [])
         }
 
         # Regex pattern for tags like {{ read_csv(..) }}, for all selected readers at once,
         # so that every page is scanned only once, no matter how many readers are selected.
         # match group 1: to extract any leading whitespace
         # match group 2: to extract the reader
-        # match group 3: to extract the arguments (positional and keywords)
+        # match group 3: to extract the arguments (positional and keywords). Matched
+        #   lazily, so that two tags on the same line are two matches instead of one.
         # Note that a reader never matches when none are selected
         self.tag_pattern = re.compile(
-            r"( *)\{\{\s+(%s)\((.+)\)\s+\}\}" % "|".join(self.readers or ["(?!)"]), # noqa: UP031
+            r"( *)\{\{\s+(%s)\((.+?)\)\s+\}\}" % "|".join(self.readers or ["(?!)"]), # noqa: UP031
             flags=re.IGNORECASE,
         )