Convert mobile Wikipedia links to standard ones

tildes/tests/test_url_transform.py

@@ -42,3 +42,20 @@ def test_reddit_tracking_removed():
     cleaned_url = apply_url_transformations(url)
 
     assert cleaned_url == "https://www.reddit.com/r/tildes/comments/8k14is/_/?sort=new"
+
+
+def test_wikipedia_mobile_conversion():
+    """Ensure that links to a Wikipedia page's mobile version are converted."""
+    url = "https://en.m.wikipedia.org/wiki/Tilde"
+    transformed_url = apply_url_transformations(url)
+
+    assert transformed_url == "https://en.wikipedia.org/wiki/Tilde"
+
+
+def test_wikipedia_mobile_homepage_not_converted():
+    """Ensure that a link to the homepage of mobile Wikipedia doesn't get converted."""
+    url = "https://en.m.wikipedia.org"
+
+    # check both with and without a trailing slash
+    for test_url in (url, url + "/"):
+        assert apply_url_transformations(test_url) == test_url

tildes/tildes/lib/url_transform.py

@@ -69,6 +69,11 @@ def _apply_all_transformations(parsed_url: ParseResult) -> ParseResult:
     return parsed_url
 
 
+def has_path(parsed_url: ParseResult) -> bool:
+    """Whether a parsed url has a path component (and not just a trailing slash)."""
+    return parsed_url.path not in ("", "/")
+
+
 class UrlTransformer(ABC):
     """Abstract base class for url transformers.
 
@@ -148,3 +153,18 @@ class RedditTrackingRemover(UrlTransformer):
         query_params.pop("sh", None)
 
         return parsed_url._replace(query=urlencode(query_params, doseq=True))
+
+
+class WikipediaMobileConverter(UrlTransformer):
+    """Convert links to Wikipedia mobile version to the standard version."""
+
+    @classmethod
+    def is_applicable(cls, parsed_url: ParseResult) -> bool:
+        """Return whether this transformation should be applied to the url."""
+        return parsed_url.hostname.endswith(".m.wikipedia.org") and has_path(parsed_url)
+
+    @classmethod
+    def apply_transformation(cls, parsed_url: ParseResult) -> ParseResult:
+        """Apply the actual transformation process to the url."""
+        new_domain = parsed_url.hostname.replace(".m.wikipedia.org", ".wikipedia.org")
+        return parsed_url._replace(netloc=new_domain)