binary-io 0.5.0 → 0.6.0

raw patch · 3 files changed

+11/−4 lines, 3 filesPVP ok

version bump matches the API change (PVP)

API changes (from Hackage documentation)

- Data.Binary.IO.Lifted: newWriterWith :: Functor m => (ByteString -> m ()) -> Writer m

+ Data.Binary.IO.Lifted: newWriterWith :: Applicative m => (ByteString -> m ()) -> Writer m

Files

ChangeLog.md +4/−0
binary-io.cabal +1/−1
lib/Data/Binary/IO/Lifted.hs +6/−3

ChangeLog.md view

@@ -1,5 +1,9 @@ # binary-io +## 0.6.0++* Respect chunking from the underlying PutM when writing resulting ByteStrings+ ## 0.5.0  * Let runPut and company return the result of the PutM operation

binary-io.cabal view

@@ -5,7 +5,7 @@   binary-io  version:-  0.5.0+  0.6.0  category:   Data, Parsing, IO

lib/Data/Binary/IO/Lifted.hs view

@@ -58,7 +58,8 @@ import qualified Data.Binary.Put as Put import           Data.ByteString (ByteString) import qualified Data.ByteString as ByteString-import           Data.ByteString.Lazy (toStrict)+import           Data.ByteString.Lazy (toChunks, toStrict)+import           Data.Foldable (traverse_) import           Data.IORef (atomicModifyIORef', mkWeakIORef, newIORef) import qualified Deque.Strict as Deque import           System.IO (Handle, hSetBinaryMode)@@ -241,12 +242,14 @@ -- -- @since 0.4.0 newWriterWith-  :: Functor m+  :: Applicative m   => (ByteString -> m ()) -- ^ Chunk writer   -> Writer m newWriterWith write = Writer $ \put -> do   let (result, body) = Put.runPutM put-  result <$ write (toStrict body)+  result <$+    -- Instead of allocating a single big+    traverse_ write (toChunks body)  -- | Create a writer. --