Data.Bytes · byteslice-0.2.13.2

Break a byte sequence into pieces separated by the byte argument, consuming the delimiter. This function is a good producer for list fusion. It is common to immidiately consume the results of split with foldl', traverse_, foldlM, and being a good producer helps in this situation.

Note: this function differs from its counterpart in bytestring. If the byte sequence is empty, this returns a singleton list with the empty byte sequence.

valuesplitU :: Word8 -> Bytes -> UnliftedArray ByteArray

#

Variant of split that returns an array of unsliced byte sequences. Unlike split, this is not a good producer for list fusion. (It does not return a list, so it could not be.) Prefer split if the result is going to be consumed exactly once by a good consumer. Prefer splitU if the result of the split is going to be around for a while and inspected multiple times.

valuesplitInit :: Word8 -> Bytes -> [Bytes]

#

Variant of split that drops the trailing element. This behaves correctly even if the byte sequence is empty. This is a good producer for list fusion. This is useful when splitting a text file into lines. POSIX mandates that text files end with a newline, so the list resulting from split always has an empty byte sequence as its last element. With splitInit, that unwanted element is discarded.

valuesplitInitU :: Word8 -> Bytes -> UnliftedArray ByteArray

#

Variant of splitU that drops the trailing element. See splitInit for an explanation of why this may be useful.

valuesplitNonEmpty :: Word8 -> Bytes -> NonEmpty Bytes

#

Variant of split that returns the result as a NonEmpty instead of []. This is also eligible for stream fusion.

valuesplitStream :: Applicative m => Word8 -> Bytes -> Stream m Bytes

#

Variant of split that intended for use with stream fusion rather than build-foldr fusion.

Fixed from Beginning

valuesplit1 :: Word8 -> Bytes -> Maybe (Bytes, Bytes)

#

Split a byte sequence on the first occurrence of the target byte. The target is removed from the result. For example:

Example1 expression

>>> split1 0xA [0x1,0x2,0xA,0xB]Just ([0x1,0x2],[0xB])

valuesplitTetragram1

:: Word8
-> Word8
-> Word8
-> Word8
-> Bytes
-> Maybe (Bytes, Bytes)

#

valuesplit2 :: Word8 -> Bytes -> Maybe (Bytes, Bytes, Bytes)

#

Split a byte sequence on the first and second occurrences of the target byte. The target is removed from the result. For example:

Example1 expression

>>> split2 0xA [0x1,0x2,0xA,0xB,0xA,0xA,0xA]Just ([0x1,0x2],[0xB],[0xA,0xA])

valuesplit3 :: Word8 -> Bytes -> Maybe (Bytes, Bytes, Bytes, Bytes)

#

Split a byte sequence on the first, second, and third occurrences of the target byte. The target is removed from the result. For example:

Example1 expression

>>> split3 0xA [0x1,0x2,0xA,0xB,0xA,0xA,0xA]Just ([0x1,0x2],[0xB],[],[0xA])

valuesplit4 :: Word8 -> Bytes -> Maybe (Bytes, Bytes, Bytes, Bytes, Bytes)

#

Split a byte sequence on the first, second, third, and fourth occurrences of the target byte. The target is removed from the result. For example:

Example1 expression

>>> split4 0xA [0x1,0x2,0xA,0xB,0xA,0xA,0xA]Just ([0x1,0x2],[0xB],[],[],[])

Fixed from End

valuesplitEnd1 :: Word8 -> Bytes -> Maybe (Bytes, Bytes)

#

Split a byte sequence on the last occurrence of the target byte. The target is removed from the result. For example:

Example1 expression

>>> split1 0xA [0x1,0x2,0xA,0xB,0xA,0xC]Just ([0x1,0x2,0xA,0xB],[0xC])

Combining

4 declarations

valueintercalate

:: Bytes
Separator (interspersed into the list)
-> [Bytes]
List
-> Bytes

#

O(n) The intercalate function takes a separator Bytes and a list of Bytes and concatenates the list elements by interspersing the separator between each element.

valueintercalateByte2

:: Word8
Separator
-> Bytes
First byte sequence
-> Bytes
Second byte sequence
-> Bytes

#

Specialization of intercalate where the separator is a single byte and there are exactly two byte sequences that are being concatenated.

valueconcatArray :: Array Bytes -> Bytes

#

valueconcatArrayU :: Array Bytes -> ByteArray

#

Searching

3 declarations

valuereplace

:: Bytes
needle, must not be empty
-> Bytes
replacement
-> Bytes
haystack
-> Bytes

#

Replace every non-overlapping occurrence of needle in haystack with replacement.

valuefindIndices

:: Bytes
needle
-> Bytes
haystack
-> PrimArray Int

#

Find locations of non-overlapping instances of needle within haystack.

valuefindTetragramIndex :: Word8 -> Word8 -> Word8 -> Word8 -> Bytes -> Maybe Int

#

Counting

1 declaration

valuecount :: Word8 -> Bytes -> Int

#

Count the number of times the byte appears in the sequence.

Prefix and Suffix

0 declarations

Byte Sequence

valueisPrefixOf :: Bytes -> Bytes -> Bool

#

Is the first argument a prefix of the second argument?

valueisSuffixOf :: Bytes -> Bytes -> Bool

#

Is the first argument a suffix of the second argument?

valueisInfixOf

:: Bytes
String to search for
-> Bytes
String to search in
-> Bool

#

Is the first argument an infix of the second argument?

Uses the Rabin-Karp algorithm: expected time O(n+m), worst-case O(nm).

valuestripPrefix :: Bytes -> Bytes -> Maybe Bytes

#

O(n) Return the suffix of the second string if its prefix matches the entire first string.

valuestripOptionalPrefix :: Bytes -> Bytes -> Bytes

#

O(n) Return the suffix of the second string if its prefix matches the entire first string. Otherwise, return the second string unchanged.

valuestripSuffix :: Bytes -> Bytes -> Maybe Bytes

#

O(n) Return the prefix of the second string if its suffix matches the entire first string.

valuestripOptionalSuffix :: Bytes -> Bytes -> Bytes

#

O(n) Return the prefix of the second string if its suffix matches the entire first string. Otherwise, return the second string unchanged.

valuelongestCommonPrefix :: Bytes -> Bytes -> Bytes

#

Find the longest string which is a prefix of both arguments.

C Strings

valuestripCStringPrefix :: CString -> Bytes -> Maybe Bytes

#

O(n) Variant of stripPrefix that takes a NUL-terminated C String as the prefix to test for.

Single Byte

valueisBytePrefixOf :: Word8 -> Bytes -> Bool

#

Does the byte sequence begin with the given byte? False if the byte sequence is empty.

valueisByteSuffixOf :: Word8 -> Bytes -> Bool

#

Does the byte sequence end with the given byte? False if the byte sequence is empty.

Equality

0 declarations

Fixed Characters

valueequalsLatin1 :: Char -> Bytes -> Bool

#

Deprecated. use Data.Bytes.Text.Latin1.equals1 instead

Is the byte sequence, when interpreted as ISO-8859-1-encoded text, a singleton whose element matches the character?

valueequalsLatin2 :: Char -> Char -> Bytes -> Bool

#

Deprecated. use Data.Bytes.Text.Latin1.equals2 instead

Is the byte sequence, when interpreted as ISO-8859-1-encoded text, a doubleton whose elements match the characters?

valueequalsLatin3 :: Char -> Char -> Char -> Bytes -> Bool

#

Deprecated. use Data.Bytes.Text.Latin1.equals3 instead

Is the byte sequence, when interpreted as ISO-8859-1-encoded text, a tripleton whose elements match the characters?

valueequalsLatin4 :: Char -> Char -> Char -> Char -> Bytes -> Bool

#

Deprecated. use Data.Bytes.Text.Latin1.equals4 instead

Is the byte sequence, when interpreted as ISO-8859-1-encoded text, a quadrupleton whose elements match the characters?

valueequalsLatin5 :: Char -> Char -> Char -> Char -> Char -> Bytes -> Bool

#

Deprecated. use Data.Bytes.Text.Latin1.equals5 instead

Is the byte sequence, when interpreted as ISO-8859-1-encoded text, a quintupleton whose elements match the characters?

valueequalsLatin6

:: Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Bytes
-> Bool

#

Deprecated. use Data.Bytes.Text.Latin1.equals6 instead

Is the byte sequence, when interpreted as ISO-8859-1-encoded text, a sextupleton whose elements match the characters?

valueequalsLatin7

:: Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Bytes
-> Bool

#

Deprecated. use Data.Bytes.Text.Latin1.equals7 instead

Is the byte sequence, when interpreted as ISO-8859-1-encoded text, a septupleton whose elements match the characters?

valueequalsLatin8

:: Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Bytes
-> Bool

#

Deprecated. use Data.Bytes.Text.Latin1.equals8 instead

Is the byte sequence, when interpreted as ISO-8859-1-encoded text, an octupleton whose elements match the characters?

valueequalsLatin9

:: Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Bytes
-> Bool

#

Deprecated. use Data.Bytes.Text.Latin1.equals9 instead

Is the byte sequence, when interpreted as ISO-8859-1-encoded text, a 9-tuple whose elements match the characters?

valueequalsLatin10

:: Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Bytes
-> Bool

#

Deprecated. use Data.Bytes.Text.Latin1.equals10 instead

Is the byte sequence, when interpreted as ISO-8859-1-encoded text, a 10-tuple whose elements match the characters?

valueequalsLatin11

:: Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Bytes
-> Bool

#

Deprecated. use Data.Bytes.Text.Latin1.equals11 instead

Is the byte sequence, when interpreted as ISO-8859-1-encoded text, a 11-tuple whose elements match the characters?

valueequalsLatin12

:: Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Char
-> Bytes
-> Bool

#

Deprecated. use Data.Bytes.Text.Latin1.equals12 instead

Is the byte sequence, when interpreted as ISO-8859-1-encoded text, a 12-tuple whose elements match the characters?

C Strings

valueequalsCString :: CString -> Bytes -> Bool

#

Is the byte sequence equal to the NUL-terminated C String? The C string must be a constant.

Hashing

2 declarations

valuefnv1a32 :: Bytes -> Word32

#

Hash byte sequence with 32-bit variant of FNV-1a.

valuefnv1a64 :: Bytes -> Word64

#

Hash byte sequence with 64-bit variant of FNV-1a.

Unsafe Slicing

4 declarations

valueunsafeTake :: Int -> Bytes -> Bytes

#

Take the first n bytes from the argument. Precondition: n ≤ len

valueunsafeDrop :: Int -> Bytes -> Bytes

#

Drop the first n bytes from the argument. Precondition: n ≤ len

valueunsafeIndex :: Bytes -> Int -> Word8

#

Index into the byte sequence at the given position. This index must be less than the length.

valueunsafeHead :: Bytes -> Word8

#

Access the first byte. The given Bytes must be non-empty.

Copying

1 declaration

valueunsafeCopy

:: PrimMonad m
=> MutableByteArray (PrimState m)
Destination
-> Int
Destination Offset
-> Bytes
Source
-> m ()

#

Copy the byte sequence into a mutable buffer. The buffer must have enough space to accomodate the byte sequence, but this this is not checked.

Pointers

3 declarations

valuepin :: Bytes -> Bytes

#

Yields a pinned byte sequence whose contents are identical to those of the original byte sequence. If the ByteArray backing the argument was already pinned, this simply aliases the argument and does not perform any copying.

valuecontents :: Bytes -> Ptr Word8

#

Yields a pointer to the beginning of the byte sequence. It is only safe to call this on a Bytes backed by a pinned ByteArray.

valuetouch :: PrimMonad m => Bytes -> m ()

#

Touch the byte array backing the byte sequence. This sometimes needed after calling contents so that the ByteArray does not get garbage collected.

Conversion

19 declarations

valuetoByteArray :: Bytes -> ByteArray

#

Convert the sliced Bytes to an unsliced ByteArray. This reuses the array backing the sliced Bytes if the slicing metadata implies that all of the bytes are used. Otherwise, it makes a copy.

valuetoByteArrayClone :: Bytes -> ByteArray

#

Variant of toByteArray that unconditionally makes a copy of the array backing the sliced Bytes even if the original array could be reused. Prefer toByteArray.

valuetoPinnedByteArray :: Bytes -> ByteArray

#

Convert the sliced Bytes to an unsliced ByteArray. This reuses the array backing the sliced Bytes if the slicing metadata implies that all of the bytes are used and they are already pinned. Otherwise, it makes a copy.

valuetoPinnedByteArrayClone :: Bytes -> ByteArray

#

Variant of toPinnedByteArray that unconditionally makes a copy of the array backing the sliced Bytes even if the original array could be reused. Prefer toPinnedByteArray.

valuefromAsciiString :: String -> Bytes

#

Deprecated. use Data.Bytes.Text.Ascii.fromString instead

Convert a String consisting of only characters in the ASCII block to a byte sequence. Any character with a codepoint above U+007F is replaced by U+0000.

valuefromLatinString :: String -> Bytes

#

Deprecated. use Data.Bytes.Text.Latin1.fromString instead

Convert a String consisting of only characters representable by ISO-8859-1. These are encoded with ISO-8859-1. Any character with a codepoint above U+00FF is replaced by an unspecified byte.

valuefromByteArray :: ByteArray -> Bytes

#

Create a slice of Bytes that spans the entire argument array.

valuefromPrimArray :: PrimArray Word8 -> Bytes

#

Create a slice of Bytes that spans the entire PrimArray of 8-bit words.

valuetoLatinString :: Bytes -> String

#

Deprecated. use Data.Bytes.Text.Latin1.toString instead

Interpret a byte sequence as text encoded by ISO-8859-1.

valuefromCString# :: Addr# -> Bytes

#

Copy a primitive string literal into managed memory.

valuetoByteString :: Bytes -> ByteString

#

O(n) when unpinned, O(1) when pinned. Create a ByteString from a byte sequence. This only copies the byte sequence if it is not pinned.

valuepinnedToByteString :: Bytes -> ByteString

#

Convert a pinned Bytes to a ByteString O(1) Precondition: bytes are pinned. Behavior is undefined otherwise.

valuefromByteString :: ByteString -> Bytes

#

O(n) Copy a ByteString to a byte sequence.

valuefromLazyByteString :: ByteString -> Bytes

#

O(n) Copy a lazy bytestring to a byte sequence.

valuefromShortByteString :: ShortByteString -> Bytes

#

O(1) Create Bytes from a ShortByteString.

valuefromShortText :: ShortText -> Bytes

#

O(1) Create Bytes from a ShortText. This encodes the text as UTF-8. It is a no-op.

valuetoShortByteString :: Bytes -> ShortByteString

#

Convert the sliced Bytes to an unsliced ShortByteString. This reuses the array backing the sliced Bytes if the slicing metadata implies that all of the bytes are used. Otherwise, it makes a copy.

valuetoShortByteStringClone :: Bytes -> ShortByteString

#

Variant of toShortByteString that unconditionally makes a copy of the array backing the sliced Bytes even if the original array could be reused. Prefer toShortByteString.

valuetoLowerAsciiByteArrayClone :: Bytes -> ByteArray

#

Deprecated. use Data.BytesTextAsciiExt.toLowerU

O(n) Interpreting the bytes an ASCII-encoded characters, convert the string to lowercase. This adds 0x20 to bytes in the range [0x41,0x5A] and leaves all other bytes alone. Unconditionally copies the bytes.