chore: pragma and sdist packaging cleanup

benchmark/Main.hs

@@ -7,11 +7,12 @@ import qualified DataFrame as D
 import qualified DataFrame.Functions as F
 
 import Control.DeepSeq (NFData (..))
-import Control.Monad (void)
+import Control.Monad (void, when)
 import Criterion.Main
 import DataFrame.Internal.DataFrame (forceDataFrame)
 import DataFrame.Operations.Join
 import DataFrame.Operators
+import System.Directory (doesFileExist)
 import System.Process hiding (env)
 import System.Random.Stateful
 
@@ -95,6 +96,15 @@ groupByExplorer = do
 parseFile :: String -> IO ()
 parseFile = void . D.readCsv
 
+covidCsv :: String
+covidCsv = "./data/effects-of-covid-19-on-trade-at-15-december-2021-provisional.csv"
+
+-- The 9.5 MB covid input is not shipped in the sdist; no-op when absent.
+parseFileIfPresent :: String -> IO ()
+parseFileIfPresent path = do
+    exists <- doesFileExist path
+    when exists (parseFile path)
+
 parseHousingCSV :: IO ()
 parseHousingCSV = parseFile "./data/housing.csv"
 
@@ -223,10 +233,7 @@ main = do
             ]
         , bgroup
             "effects-of-covid-19-on-trade-at-15-december-2021-provisional.csv (9.1 MB)"
-            [ bench "Attoparsec" $
-                nfIO $
-                    parseFile
-                        "./data/effects-of-covid-19-on-trade-at-15-december-2021-provisional.csv"
+            [ bench "Attoparsec" $ nfIO $ parseFileIfPresent covidCsv
             ]
         , bgroup
             "join/inner/1:1"

dataframe-operations/src/DataFrame/Functions.hs

@@ -2,7 +2,6 @@
 {-# LANGUAGE FlexibleContexts #-}
 {-# LANGUAGE FlexibleInstances #-}
 {-# LANGUAGE GADTs #-}
-{-# LANGUAGE IncoherentInstances #-}
 {-# LANGUAGE MultiParamTypeClasses #-}
 {-# LANGUAGE OverloadedStrings #-}
 {-# LANGUAGE RankNTypes #-}

dataframe.cabal

@@ -28,6 +28,13 @@ extra-source-files: cbits/arrow_abi.h
                     tests/data/*.parquet
                     tests/data/unstable_csv/*.csv
                     tests/data/unstable_csv/*.tsv
+                    data/housing.csv
+                    data/starwars.csv
+                    data/chipotle.tsv
+                    data/measurements.txt
+                    benchmark/pandas/*.py
+                    benchmark/polars/*.py
+                    benchmark/explorer/*.exs
                     -- tests/data/*.md
                     -- tests/data/*.bin
                     -- tests/data/*.json
@@ -290,6 +297,7 @@ benchmark dataframe-benchmark
     build-depends: base >= 4 && < 5,
                    criterion >= 1 && < 2,
                    deepseq >= 1.4 && < 2,
+                   directory >= 1.3 && < 2,
                    process >= 1.6 && < 2,
                    dataframe >= 3.5 && < 3.6,
                    dataframe-core >= 2.4 && < 2.5,