Introduce row for partially parsed in columnInfo

Author: mchav

src/Data/DataFrame/Operations/Core.hs

@@ -23,6 +23,7 @@ import Data.DataFrame.Internal.Column ( Column(..), toColumn', toColumn, columnL
 import Data.DataFrame.Internal.DataFrame (DataFrame(..), getColumn, null, empty)
 import Data.DataFrame.Internal.Parsing (isNullish)
 import Data.DataFrame.Internal.Types (Columnable)
+import Data.Either
 import Data.Function (on, (&))
 import Data.Maybe
 import Data.Type.Equality (type (:~:)(Refl), TestEquality(..))
@@ -152,27 +153,30 @@ columnInfo :: DataFrame -> DataFrame
 columnInfo df = empty & insertColumn' "Column Name" (Just $ toColumn (map fst' triples))
                       & insertColumn' "# Non-null Values" (Just $ toColumn (map snd' triples))
                       & insertColumn' "# Null Values" (Just $ toColumn (map thd' triples))
-                      & insertColumn' "Type" (Just $ toColumn (map fth' triples))
+                      & insertColumn' "# Partially parsed" (Just $ toColumn (map fth' triples))
+                      & insertColumn' "Type" (Just $ toColumn (map ffth' triples))
   where
-    triples = L.sortBy (compare `on` snd') (V.ifoldl' go [] (columns df)) :: [(T.Text, Int,  Int, T.Text)]
+    triples = L.sortBy (compare `on` snd') (V.ifoldl' go [] (columns df)) :: [(T.Text, Int,  Int, Int, T.Text)]
     indexMap = M.fromList (map (\(a, b) -> (b, a)) $ M.toList (columnIndices df))
     columnName i = indexMap M.! i
     go acc i Nothing = acc
     go acc i (Just col@(BoxedColumn (c :: V.Vector a))) = let
         cname = columnName i
         countNulls = nulls col
+        countPartial = partiallyParsed col
         columnType = T.pack $ show $ typeRep @a
-      in (cname, VG.length c - countNulls, countNulls, columnType) : acc
+      in (cname, columnLength col - countNulls, countNulls, countPartial, columnType) : acc
     go acc i (Just col@(UnboxedColumn c)) = let
         cname = columnName i
         columnType = T.pack $ columnTypeString col
         -- Unboxed columns cannot have nulls since Maybe
         -- is not an instance of Unbox a
-      in (cname, VG.length c, 0, columnType) : acc
-    fst' (!x, _, _, _) = x
-    snd' (_, !x, _, _) = x
-    thd' (_, _, !x, _) = x
-    fth' (_, _, _, !x) = x
+      in (cname, columnLength col, 0, 0, columnType) : acc
+    fst'  (!x, _, _, _, _) = x
+    snd'  (_, !x, _, _, _) = x
+    thd'  (_, _, !x, _, _) = x
+    fth'  (_, _, _, !x, _) = x
+    ffth' (_, _, _, _, !x) = x
 
 
 nulls :: Column -> Int
@@ -187,6 +191,15 @@ nulls (BoxedColumn (xs :: V.Vector a)) = case testEquality (typeRep @a) (typeRep
       _ -> 0
 nulls _ = 0
 
+partiallyParsed :: Column -> Int
+partiallyParsed (BoxedColumn (xs :: V.Vector a)) =
+  case typeRep @a of
+    App (App tycon t1) t2 -> case eqTypeRep tycon (typeRep @Either) of
+      Just HRefl -> VG.length $ VG.filter isLeft xs
+      Nothing -> 0
+    _ -> 0
+partiallyParsed _ = 0
+
 fromList :: [(T.Text, Column)] -> DataFrame
 fromList = L.foldl' (\df (name, column) -> insertColumn' name (Just column) df) empty