Fix Windows test portability: line-ending translation, locale encoding, and /tmp assumptions

.gitattributes

@@ -0,0 +1,3 @@
+# git should not translate line endings on checkout.
+*.csv -text
+*.tsv -text

dataframe-fastcsv/tests/Operations/Projection.hs

@@ -10,7 +10,7 @@ module Operations.Projection (tests) where
 
 import qualified Data.Map as M
 import qualified Data.Text as T
-import qualified Data.Text.IO as TIO
+import qualified Data.Text.IO.Utf8 as TIO
 
 import Control.Exception (SomeException, evaluate, try)
 import Data.List (isInfixOf)

dataframe-fastcsv/tests/Operations/ReadCsv.hs

@@ -37,7 +37,14 @@ import DataFrame.Internal.DataFrame (
  )
 import DataFrame.Schema (Schema (..), SchemaType (..))
 import System.Directory (removeFile)
-import System.IO (IOMode (..), withFile)
+import System.IO (
+    IOMode (..),
+    hSetEncoding,
+    hSetNewlineMode,
+    noNewlineTranslation,
+    utf8,
+    withFile,
+ )
 import Test.HUnit
 import Type.Reflection (typeRep)
 
@@ -59,6 +66,8 @@ prettyPrintTsv = prettyPrintSeparated '\t'
 
 prettyPrintSeparated :: Char -> FilePath -> DataFrame -> IO ()
 prettyPrintSeparated sep filepath df = withFile filepath WriteMode $ \handle -> do
+    hSetEncoding handle utf8
+    hSetNewlineMode handle noNewlineTranslation
     let (rows, _) = dataframeDimensions df
     let headers = map fst (L.sortBy (compare `on` snd) (M.toList (columnIndices df)))
     TIO.hPutStrLn

dataframe-fastcsv/tests/Operations/TypedExtraction.hs

@@ -12,7 +12,11 @@ import qualified Data.Map as M
 import qualified Data.Proxy as P
 import qualified Data.Text as T
 import qualified Data.Text.Encoding as TE
-import qualified Data.Text.IO as TIO
+
+-- UTF-8 byte-mode IO: the plain Data.Text.IO writer honours the
+-- handle's text mode, which on Windows turns \n into \r\n and
+-- corrupts inputs meant for byte-level parsers.
+import qualified Data.Text.IO.Utf8 as TIO
 
 import Control.Exception (ErrorCall, evaluate, try)
 import Data.Time (Day)

dataframe-fastcsv/tests/Properties/Csv.hs

@@ -20,7 +20,7 @@ import qualified Data.List as L
 import qualified Data.Map as M
 import qualified Data.Text as T
 import qualified Data.Text.Encoding as TE
-import qualified Data.Text.IO as TIO
+import qualified Data.Text.IO.Utf8 as TIO
 import qualified Data.Vector as V
 
 import DataFrame.IO.CSV (defaultReadOptions)
@@ -130,7 +130,7 @@ the temp file afterwards no matter what.
 -}
 withCsvFile :: String -> T.Text -> (FilePath -> IO a) -> IO a
 withCsvFile label body action = do
-    let path = "/tmp/fastcsv_prop_" <> label <> ".csv"
+    let path = "./tests/data/unstable_csv/fastcsv_prop_" <> label <> ".csv"
     TIO.writeFile path body
     r <- action path
     removeFile path
@@ -209,7 +209,7 @@ prop_unclosed_quote_throws = forAll (listOf1 arbitrary) $ \(cells :: [Cell]) ->
                 T.intercalate "," (map (encodeCell ',' . unCell) cells)
             csv = "v\n" <> plainRow <> ",\"dangling\n"
         result <- run $ do
-            let path = "/tmp/fastcsv_prop_unclosed.csv"
+            let path = "./tests/data/unstable_csv/fastcsv_prop_unclosed.csv"
             TIO.writeFile path csv
             r <- try @CsvParseError (D.fastReadCsv path)
             removeFile path

tests/Operations/Record.hs

@@ -17,7 +17,7 @@ module Operations.Record where
 import Data.Int (Int64)
 import qualified Data.Map.Strict as M
 import qualified Data.Text as T
-import qualified Data.Text.IO as TIO
+import qualified Data.Text.IO.Utf8 as TIO
 import GHC.Generics (Generic)
 
 import qualified DataFrame as D
@@ -27,6 +27,7 @@ import DataFrame.Operators
 import qualified DataFrame.Schema as IS
 import DataFrame.Typed (Schema)
 import qualified DataFrame.Typed as DT
+import System.Directory (getTemporaryDirectory)
 
 import Test.HUnit
 
@@ -289,7 +290,8 @@ deriveSchemaReadsCsv = TestCase $ do
                 , "2,eu,20.5"
                 , "3,ap,30.0"
                 ]
-        tmp = "/tmp/dataframe_test_deriveSchema.csv"
+    tmpDir <- getTemporaryDirectory
+    let tmp = tmpDir <> "/dataframe_test_deriveSchema.csv"
     TIO.writeFile tmp csv
     df <- D.readCsvWithSchema orderSchema tmp
     assertEqual

tests/Operations/WriteCsv.hs

@@ -4,10 +4,11 @@
 module Operations.WriteCsv where
 
 import qualified Data.Text as T
-import qualified Data.Text.IO as TIO
+import qualified Data.Text.IO.Utf8 as TIO
 import qualified DataFrame as D
 import qualified DataFrame.Internal.Column as DI
 import DataFrame.Internal.DataFrame (DataFrame (..), toCsv, toSeparated)
+import System.Directory (getTemporaryDirectory)
 import Test.HUnit
 
 -- Basic test: Int and Text columns produce correct CSV
@@ -81,7 +82,8 @@ toCsvRoundTrip = TestLabel "toCsv_roundTrip" $ TestCase $ do
                 , ("b", DI.fromList @T.Text ["hello", "world", "test"])
                 ]
     let csvText = toCsv df
-    let tmpPath = "/tmp/dataframe_test_toCsv_roundtrip.csv"
+    tmpDir <- getTemporaryDirectory
+    let tmpPath = tmpDir <> "/dataframe_test_toCsv_roundtrip.csv"
     TIO.writeFile tmpPath csvText
     df' <- D.readCsv tmpPath
     assertEqual