Merge pull request #3321 from boxydog/warn_about_markdown

pelican/__init__.py

@@ -30,7 +30,6 @@ from pelican.generators import (
 )
 from pelican.plugins import signals
 from pelican.plugins._utils import get_plugin_name, load_plugins
-from pelican.readers import Readers
 from pelican.server import ComplexHTTPRequestHandler, RootedHTTPServer
 from pelican.settings import read_settings
 from pelican.utils import clean_output_dir, maybe_pluralize, wait_for_changes
@@ -126,6 +125,8 @@ class Pelican:
         for p in generators:
             if hasattr(p, "generate_context"):
                 p.generate_context()
+            if hasattr(p, "check_disabled_readers"):
+                p.check_disabled_readers()
 
         # for plugins that create/edit the summary
         logger.debug("Signal all_generators_finalized.send(<generators>)")
@@ -573,7 +574,7 @@ def autoreload(args, excqueue=None):
         try:
             pelican.run()
 
-            changed_files = wait_for_changes(args.settings, Readers, settings)
+            changed_files = wait_for_changes(args.settings, settings)
             changed_files = {c[1] for c in changed_files}
 
             if settings_file in changed_files:

pelican/generators.py

@@ -7,6 +7,7 @@ from collections import defaultdict
 from functools import partial
 from itertools import chain, groupby
 from operator import attrgetter
+from typing import List, Optional, Set
 
 from jinja2 import (
     BaseLoader,
@@ -156,7 +157,9 @@ class Generator:
 
         return False
 
-    def get_files(self, paths, exclude=None, extensions=None):
+    def get_files(
+        self, paths, exclude: Optional[List[str]] = None, extensions=None
+    ) -> Set[str]:
         """Return a list of files to use, based on rules
 
         :param paths: the list pf paths to search (relative to self.path)
@@ -250,6 +253,13 @@ class Generator:
         # return the name of the class for logging purposes
         return self.__class__.__name__
 
+    def _check_disabled_readers(self, paths, exclude: Optional[List[str]]) -> None:
+        """Log warnings for files that would have been processed by disabled readers."""
+        for fil in self.get_files(
+            paths, exclude=exclude, extensions=self.readers.disabled_extensions
+        ):
+            self.readers.check_file(fil)
+
 
 class CachingGenerator(Generator, FileStampDataCacher):
     """Subclass of Generator and FileStampDataCacher classes
@@ -643,6 +653,11 @@ class ArticlesGenerator(CachingGenerator):
         self.generate_authors(write)
         self.generate_drafts(write)
 
+    def check_disabled_readers(self) -> None:
+        self._check_disabled_readers(
+            self.settings["ARTICLE_PATHS"], exclude=self.settings["ARTICLE_EXCLUDES"]
+        )
+
     def generate_context(self):
         """Add the articles into the shared context"""
 
@@ -849,6 +864,11 @@ class PagesGenerator(CachingGenerator):
         super().__init__(*args, **kwargs)
         signals.page_generator_init.send(self)
 
+    def check_disabled_readers(self) -> None:
+        self._check_disabled_readers(
+            self.settings["PAGE_PATHS"], exclude=self.settings["PAGE_EXCLUDES"]
+        )
+
     def generate_context(self):
         all_pages = []
         hidden_pages = []
@@ -953,6 +973,11 @@ class StaticGenerator(Generator):
         self.fallback_to_symlinks = False
         signals.static_generator_init.send(self)
 
+    def check_disabled_readers(self) -> None:
+        self._check_disabled_readers(
+            self.settings["STATIC_PATHS"], exclude=self.settings["STATIC_EXCLUDES"]
+        )
+
     def generate_context(self):
         self.staticfiles = []
         linked_files = set(self.context["static_links"])

pelican/readers.py

@@ -17,7 +17,7 @@ from pelican import rstdirectives  # NOQA
 from pelican.cache import FileStampDataCacher
 from pelican.contents import Author, Category, Page, Tag
 from pelican.plugins import signals
-from pelican.utils import get_date, pelican_open, posixize_path
+from pelican.utils import file_suffix, get_date, pelican_open, posixize_path
 
 try:
     from markdown import Markdown
@@ -125,6 +125,10 @@ class BaseReader:
         metadata = {}
         return content, metadata
 
+    def disabled_message(self) -> str:
+        """Message about why this plugin was disabled."""
+        return ""
+
 
 class _FieldBodyTranslator(HTMLTranslator):
     def __init__(self, document):
@@ -347,6 +351,12 @@ class MarkdownReader(BaseReader):
             metadata = {}
         return content, metadata
 
+    def disabled_message(self) -> str:
+        return (
+            "Could not import markdown.Markdown.  "
+            "Have you installed the markdown package?"
+        )
+
 
 class HTMLReader(BaseReader):
     """Parses HTML files as input, looking for meta, title, and body tags"""
@@ -508,17 +518,23 @@ class Readers(FileStampDataCacher):
     def __init__(self, settings=None, cache_name=""):
         self.settings = settings or {}
         self.readers = {}
+        self.disabled_readers = {}
+        # extension => reader for readers that are enabled
         self.reader_classes = {}
+        # extension => reader for readers that are not enabled
+        disabled_reader_classes = {}
 
         for cls in [BaseReader] + BaseReader.__subclasses__():
             if not cls.enabled:
                 logger.debug(
                     "Missing dependencies for %s", ", ".join(cls.file_extensions)
                 )
-                continue
 
             for ext in cls.file_extensions:
-                self.reader_classes[ext] = cls
+                if cls.enabled:
+                    self.reader_classes[ext] = cls
+                else:
+                    disabled_reader_classes[ext] = cls
 
         if self.settings["READERS"]:
             self.reader_classes.update(self.settings["READERS"])
@@ -531,6 +547,9 @@ class Readers(FileStampDataCacher):
 
             self.readers[fmt] = reader_class(self.settings)
 
+        for fmt, reader_class in disabled_reader_classes.items():
+            self.disabled_readers[fmt] = reader_class(self.settings)
+
         # set up caching
         cache_this_level = (
             cache_name != "" and self.settings["CONTENT_CACHING_LAYER"] == "reader"
@@ -541,8 +560,13 @@ class Readers(FileStampDataCacher):
 
     @property
     def extensions(self):
+        """File extensions that will be processed by a reader."""
         return self.readers.keys()
 
+    @property
+    def disabled_extensions(self):
+        return self.disabled_readers.keys()
+
     def read_file(
         self,
         base_path,
@@ -562,8 +586,7 @@ class Readers(FileStampDataCacher):
         logger.debug("Read file %s -> %s", source_path, content_class.__name__)
 
         if not fmt:
-            _, ext = os.path.splitext(os.path.basename(path))
+            fmt = file_suffix(path)
-            fmt = ext[1:]
 
         if fmt not in self.readers:
             raise TypeError("Pelican does not know how to parse %s", path)
@@ -654,6 +677,12 @@ class Readers(FileStampDataCacher):
             context=context,
         )
 
+    def check_file(self, source_path: str) -> None:
+        """Log a warning if a file is processed by a disabled reader."""
+        reader = self.disabled_readers.get(file_suffix(source_path), None)
+        if reader:
+            logger.warning(f"{source_path}: {reader.disabled_message()}")
+
 
 def find_empty_alt(content, path):
     """Find images with empty alt

pelican/tests/test_pelican.py

@@ -9,10 +9,11 @@ import unittest
 from collections.abc import Sequence
 from shutil import rmtree
 from tempfile import TemporaryDirectory, mkdtemp
-from unittest.mock import patch
+from unittest.mock import PropertyMock, patch
 
 from rich.console import Console
 
+import pelican.readers
 from pelican import Pelican, __version__, main
 from pelican.generators import StaticGenerator
 from pelican.settings import read_settings
@@ -303,3 +304,24 @@ class TestPelican(LoggedTestCase):
                     main(["-o", temp_dir, "pelican/tests/simple_content"])
             self.assertIn("Processed 1 article", out.getvalue())
             self.assertEqual("", err.getvalue())
+
+    def test_main_on_content_markdown_disabled(self):
+        """Invoke main on simple_content directory."""
+        with patch.object(
+            pelican.readers.MarkdownReader, "enabled", new_callable=PropertyMock
+        ) as attr_mock:
+            attr_mock.return_value = False
+            out, err = io.StringIO(), io.StringIO()
+            with contextlib.redirect_stdout(out), contextlib.redirect_stderr(err):
+                with TemporaryDirectory() as temp_dir:
+                    # Don't highlight anything.
+                    # See https://rich.readthedocs.io/en/stable/highlighting.html
+                    with patch("pelican.console", new=Console(highlight=False)):
+                        main(["-o", temp_dir, "pelican/tests/simple_content"])
+                self.assertIn("Processed 0 articles", out.getvalue())
+                self.assertLogCountEqual(
+                    1,
+                    ".*article_with_md_extension.md: "
+                    "Could not import markdown.Markdown.  "
+                    "Have you installed the markdown package?",
+                )

pelican/tests/test_readers.py

@@ -1,5 +1,5 @@
 import os
-from unittest.mock import patch
+from unittest.mock import PropertyMock, patch
 
 from pelican import readers
 from pelican.tests.support import get_settings, unittest
@@ -32,6 +32,19 @@ class ReaderTest(unittest.TestCase):
             else:
                 self.fail(f"Expected {key} to have value {value}, but was not in Dict")
 
+    def test_markdown_disabled(self):
+        with patch.object(
+            readers.MarkdownReader, "enabled", new_callable=PropertyMock
+        ) as attr_mock:
+            attr_mock.return_value = False
+            readrs = readers.Readers(settings=get_settings())
+            self.assertEqual(
+                set(readers.MarkdownReader.file_extensions),
+                readrs.disabled_readers.keys(),
+            )
+            for val in readrs.disabled_readers.values():
+                self.assertEqual(readers.MarkdownReader, val.__class__)
+
 
 class TestAssertDictHasSubset(ReaderTest):
     def setUp(self):

pelican/tests/test_utils.py

@@ -966,3 +966,10 @@ class TestMemoized(unittest.TestCase):
             container.get.cache.clear()
             self.assertEqual("bar", container.get("bar"))
             get_mock.assert_called_once_with("bar")
+
+
+class TestStringUtils(unittest.TestCase):
+    def test_file_suffix(self):
+        self.assertEqual("", utils.file_suffix(""))
+        self.assertEqual("", utils.file_suffix("foo"))
+        self.assertEqual("md", utils.file_suffix("foo.md"))

pelican/utils.py

@@ -29,6 +29,7 @@ from typing import (
 )
 
 import dateutil.parser
+from watchfiles import Change
 
 try:
     from zoneinfo import ZoneInfo
@@ -39,7 +40,6 @@ from markupsafe import Markup
 
 if TYPE_CHECKING:
     from pelican.contents import Content
-    from pelican.readers import Readers
     from pelican.settings import Settings
 
 logger = logging.getLogger(__name__)
@@ -797,9 +797,8 @@ def order_content(
 
 def wait_for_changes(
     settings_file: str,
-    reader_class: type[Readers],
     settings: Settings,
-):
+) -> set[tuple[Change, str]]:
     content_path = settings.get("PATH", "")
     theme_path = settings.get("THEME", "")
     ignore_files = {
@@ -924,3 +923,13 @@ def temporary_locale(
         locale.setlocale(lc_category, temp_locale)
     yield
     locale.setlocale(lc_category, orig_locale)
+
+
+def file_suffix(path: str) -> str:
+    """Return the suffix of a filename in a path."""
+    _, ext = os.path.splitext(os.path.basename(path))
+    ret = ""
+    if len(ext) > 1:
+        # drop the ".", e.g., "exe", not ".exe"
+        ret = ext[1:]
+    return ret