Streamly.External.Archive

Read

readArchive :: MonadIO m => Unfold m (ReadOptions m Header -> ReadOptions m a, FilePath) (Either a ByteString)

Creates an unfold with which we can stream data out of the given archive.

By default (with id as the read options modifier), we get for each entry in the archive a Header followed by zero or more ByteStrings containing chunks of file data.

To modify the read options, one can use function composition.

Read options

data ReadOptions m a

mapHeaderMaybe :: (Header -> m (Maybe a)) -> ReadOptions m Header -> ReadOptions m a

If this returns Just for a header, that header (mapped to a different value if desired) and any following ByteString chunks are included in the readArchive unfold. If this returns Nothing for a header, that header and any following ByteString chunks are excluded from the readArchive unfold.

By default, all entries are included with unaltered headers.

Utility functions

Various utility functions that some might find useful.

groupByLeft :: Monad m => Fold m (Either a b) c -> Stream m (Either a b) -> Stream m c

Groups a stream of Eithers by the Lefts. The provided Fold processes a single Left followed by any subsequent (zero or more) Rights.

eitherByLeft :: Monad m => Stream m (Either a b) -> Stream m (a, b)

Associates each Right in a stream with the latest Left that came before it.

>>> l = [Right 10, Left "a", Right 1, Right 2, Left "b", Left "c", Right 20]
>>> S.fold F.toList . eitherByLeft . S.fromList $ l
[("a",1),("a",2),("c",20)]

chunkOn :: Monad m => Word8 -> Stream m (Either a ByteString) -> Stream m (Either a ByteString)

Chunks up the bytestrings following each Left by the given word, discarding the given word. (For instance, the word could be 10 (newline), which gives us lines as the chunks.) The bytestrings in the resulting stream are the desired chunks.

chunkOnFold :: Monad m => Word8 -> Fold m (Either a ByteString) b -> Fold m (Either a ByteString) b

Chunks up the bytestrings following each Left by the given word, discarding the given word. (For instance, the word could be 10 (newline), which gives us lines as the chunks.) The bytestrings in the provided fold are the desired chunks.

Header

data Header

Header information for an entry in the archive.

data FileType

Constructors

FileTypeRegular

FileTypeSymlink

FileTypeSocket

FileTypeCharDevice

FileTypeBlockDevice

FileTypeDirectory

FileTypeNamedPipe

Instances

Show FileType

Defined in Streamly.External.Archive.Internal.Foreign

Methods

showsPrec :: Int -> FileType -> ShowS

show :: FileType -> String

showList :: [FileType] -> ShowS

Eq FileType

Defined in Streamly.External.Archive.Internal.Foreign

Methods

(==) :: FileType -> FileType -> Bool

(/=) :: FileType -> FileType -> Bool

headerFileType :: Header -> IO (Maybe FileType)

headerPathName :: Header -> IO (Maybe ByteString)

headerPathNameUtf8 :: Header -> IO (Maybe ByteString)

headerSize :: Header -> IO (Maybe Int)

Returns the file size of the entry, if it has been set; returns Nothing otherwise.