Restore gist extraction

src/Xrefcheck/Scanners/Markdown.hs

@@ -22,6 +22,7 @@ import Control.Lens hiding ((^?))
 import Control.Monad.Except (MonadError, throwError)
 import Data.Aeson.TH (deriveFromJSON)
 import Data.ByteString.Lazy qualified as BSL
+import Data.Char (isAlpha)
 import Data.DList qualified as DList
 import Data.Default (def)
 import Data.List (isSubsequenceOf)
@@ -206,7 +207,9 @@ nodeExtractInfo input@(Node _ _ nSubs) = do
 
 urlIsASubsequence :: CopyPaste -> Bool
 urlIsASubsequence paste =
-  T.unpack (cpAnchorText paste) `isSubsequenceOf` T.unpack (cpPlainText paste)
+  gist (cpAnchorText paste) `isSubsequenceOf` gist (cpPlainText paste)
+  where
+    gist = T.unpack  . T.toLower . T.filter isAlpha
 
 merge :: (Monad m, Monoid b) => [a -> m b] -> a -> m b
 merge fs a = mconcat <$> traverse ($ a) fs

tests/Test/Xrefcheck/CopyPasteInListsSpec.hs

@@ -18,7 +18,7 @@ spec = do
     for_ allFlavors $ \fl -> do
       it ("is detected (" <> show fl <> ")") $ do
         fi <- getFI fl "tests/markdowns/without-annotations/copy-paste_in_lists.md"
-        getPasta fi `shouldBe`[("a", "c")]
+        getPasta fi `shouldBe`[("foo-bar","Foo Kek")]
   where
     getPasta :: FileInfo -> [(Text, Text)]
     getPasta fi = map (cpAnchorText &&& cpPlainText) $ fi ^. fiCopyPastes

tests/markdowns/without-annotations/copy-paste_in_lists.md

@@ -6,12 +6,12 @@
 
 A list with bad copy-paste:
 
-- [a](a) e
-- [b](b) e
-- [c](a) e
+- [Foo Bar](foo-bar) e
+- [Foo Qux](foo-qux) e
+- [Foo Kek](foo-bar) e
 
 A list that is completely fine:
 
-- [a](a) d
-- [b](b) d
-- [c](c) d
+- [Foo Bar](foo-bar) e
+- [Foo Qux](foo-qux) e
+- [Foo Kek](foo-kek) e