Harendra Kumar
Harendra Kumar
We need parser tests for the following: * Functor (fmap) * Applicative (), (\*>), (>=) cc: @pranaysashank
From #159. Here is an algo to derive exposed APIs with no coverage: Run bin/test.sh --coverage Run cabal haddock --haddock-hoogle Use streamly to read the hoogle data file e.g. dist-newstyle/build/x86_64-linux/ghc-8.10.4/streamly-0.8.1/doc/html/streamly/streamly.txt...
For the 0.9.0 release. Make a list of those that do not have tests.
We can run something like this on all .hs files: ``` awk '{if (length($0) > 79) {print "Line length exceeds 79: " length($0); print $0; exit 1;}}' ``` We can...
Implement "split" operation: ``` split :: Monad m => Fold m x a -> Fold m x b -> Fold m x (a, b) ```
For the WordCountParalleUTF8 example: ``` import qualified Streamly.Internal.Unicode.Stream as Unicode (DecodeState, DecodeError(..), CodePoint, decodeUtf8Either , resumeDecodeUtf8Either) ```
Being used in the CmdClient example: ``` import qualified Streamly.Internal.Network.Inet.TCP as TCP (pipeBytes) ```
The capacity is available in the ByteArray# itself, we do not need to duplicate it.
See https://www.haskell.org/ghcup/guide/#nightlies . We are currently using this command to install GHC head version: ./ghcup install ghc -u https://gitlab.haskell.org/ghc/ghc/-/jobs/artifacts/master/raw/ghc-x86_64-linux-deb10-int_native-validate.tar.xz?job=x86_64-linux-deb10-int_native-validate head See https://github.com/composewell/streamly/blob/4e346939a57bc76f3b4a328aecaf3dd17b82035c/.github/workflows/haskell.yml#L211 Is it better to use the ghcup nightlies...