feat: Add setting to omit selected Typogrify filters. Fixes #3436 (#3439)

docs/settings.rst

@@ -288,7 +288,14 @@ Basic settings
 
    A list of tags for Typogrify to ignore. By default Typogrify will ignore
    ``pre`` and ``code`` tags. This requires that Typogrify version 2.0.4 or
-   later is installed
+   later is installed.
+
+.. data:: TYPOGRIFY_OMIT_FILTERS = []
+
+   A list of Typogrify filters to skip. Allowed values are: ``'amp'``,
+   ``'smartypants'``, ``'caps'``, ``'initial_quotes'``, ``'widont'``. By
+   default, no filter is omitted (in other words, all filters get applied). This
+   setting requires that Typogrify version 2.1.0 or later is installed.
 
 .. data:: TYPOGRIFY_DASHES = 'default'
 

pelican/readers.py

@@ -648,9 +648,20 @@ class Readers(FileStampDataCacher):
             smartypants.Attr.default |= smartypants.Attr.w
 
             def typogrify_wrapper(text):
-                """Ensures ignore_tags feature is backward compatible"""
+                """Ensure compatibility with older versions of Typogrify.
+
+                The 'TYPOGRIFY_IGNORE_TAGS' and/or 'TYPOGRIFY_OMIT_FILTERS'
+                settings will be ignored if the installed version of Typogrify
+                doesn't have the corresponding features."""
                 try:
-                    return typogrify(text, self.settings["TYPOGRIFY_IGNORE_TAGS"])
+                    return typogrify(
+                        text,
+                        self.settings["TYPOGRIFY_IGNORE_TAGS"],
+                        **{f: False for f in self.settings["TYPOGRIFY_OMIT_FILTERS"]},
+                    )
+                except TypeError:
+                    try:
+                        typogrify(text, self.settings["TYPOGRIFY_IGNORE_TAGS"])
                     except TypeError:
                         return typogrify(text)
 

pelican/settings.py

@@ -149,6 +149,7 @@ DEFAULT_CONFIG = {
     "ARTICLE_PERMALINK_STRUCTURE": "",
     "TYPOGRIFY": False,
     "TYPOGRIFY_IGNORE_TAGS": [],
+    "TYPOGRIFY_OMIT_FILTERS": [],
     "TYPOGRIFY_DASHES": "default",
     "SUMMARY_END_SUFFIX": "…",
     "SUMMARY_MAX_LENGTH": 50,

pelican/tests/test_readers.py

@@ -409,6 +409,90 @@ class RstReaderTest(ReaderTest):
         except ImportError:
             return unittest.skip("need the typogrify distribution")
 
+    def test_typogrify_ignore_filters(self):
+        try:
+            # typogrify should be able to ignore user specified filters.
+            page = self.read_file(
+                path="article_with_code_block.rst",
+                TYPOGRIFY=True,
+                TYPOGRIFY_OMIT_FILTERS=["amp"],
+            )
+            expected = (
+                "<p>An article with some&nbsp;code</p>\n"
+                '<div class="highlight"><pre><span></span>'
+                '<span class="n">x</span>'
+                ' <span class="o">&amp;</span>'
+                ' <span class="n">y</span>\n</pre></div>\n'
+                "<p>A block&nbsp;quote:</p>\n<blockquote>\nx "
+                "&amp; y</blockquote>\n"
+                "<p>Normal:\nx &amp;&nbsp;y</p>\n"
+            )
+            self.assertEqual(page.content, expected)
+
+            page = self.read_file(
+                path="article.rst",
+                TYPOGRIFY=True,
+                TYPOGRIFY_OMIT_FILTERS=["smartypants"],
+            )
+            expected = (
+                '<p><span class="caps">THIS</span> is some content. '
+                "With some stuff to&nbsp;&quot;typogrify&quot;...</p>\n"
+                '<p>Now with added support for <abbr title="three letter '
+                'acronym"><span class="caps">TLA</span></abbr>.</p>\n'
+            )
+            self.assertEqual(page.content, expected)
+
+            page = self.read_file(
+                path="article.rst",
+                TYPOGRIFY=True,
+                TYPOGRIFY_OMIT_FILTERS=["caps"],
+            )
+            expected = (
+                "<p>THIS is some content. "
+                "With some stuff to&nbsp;&#8220;typogrify&#8221;&#8230;</p>\n"
+                '<p>Now with added support for <abbr title="three letter '
+                'acronym">TLA</abbr>.</p>\n'
+            )
+            self.assertEqual(page.content, expected)
+
+            page = self.read_file(
+                path="article.rst",
+                TYPOGRIFY=True,
+                TYPOGRIFY_OMIT_FILTERS=["initial_quotes"],
+            )
+            expected = (
+                '<p><span class="caps">THIS</span> is some content. '
+                "With some stuff to&nbsp;&#8220;typogrify&#8221;&#8230;</p>\n"
+                '<p>Now with added support for <abbr title="three letter '
+                'acronym"><span class="caps">TLA</span></abbr>.</p>\n'
+            )
+            self.assertEqual(page.content, expected)
+
+            page = self.read_file(
+                path="article.rst",
+                TYPOGRIFY=True,
+                TYPOGRIFY_OMIT_FILTERS=["widont"],
+            )
+            expected = (
+                '<p><span class="caps">THIS</span> is some content. '
+                "With some stuff to "
+                "&#8220;typogrify&#8221;&#8230;</p>\n<p>Now with added "
+                'support for <abbr title="three letter acronym">'
+                '<span class="caps">TLA</span></abbr>.</p>\n'
+            )
+            self.assertEqual(page.content, expected)
+
+            page = self.read_file(
+                path="article.rst",
+                TYPOGRIFY=True,
+                TYPOGRIFY_OMIT_FILTERS=["this-filter-does-not-exists"],
+            )
+            self.assertRaises(TypeError)
+        except ImportError:
+            return unittest.skip("need the typogrify distribution")
+        except TypeError:
+            return unittest.skip("need typogrify version 2.1.0 or later")
+
     def test_typogrify_ignore_tags(self):
         try:
             # typogrify should be able to ignore user specified tags.