A efficient wrapper for short immutable null-terminated byte sequences which can be automatically freed by ghc garbage collector.

The main use case of this type is to ease the bridging of C FFI APIs, since most of the unix APIs use null-terminated string. On windows you're encouraged to use a compatibility layer like 'WideCharToMultiByte/MultiByteToWideChar' and keep the same interface, e.g. libuv do this when deal with file paths.

CBytes don't support O(1) slicing, it's not suitable to use it to store large byte chunk, If you need advance editing, convert CBytes to/from Bytes with CB pattern or toBytes / fromBytes, then use vector combinators.

When textual represatation is needed e.g. converting to String, Text, Show instance, etc., we assume CBytes using UTF-8 encodings, CBytes can be used with OverloadedString, literal encoding is UTF-8 with some modifications: \NUL is encoded to 'C0 80', and \xD800 ~ \xDFFF is encoded as a three bytes normal utf-8 codepoint.

Note most of the unix API is not unicode awared though, you may find a scandir call return a filename which is not proper encoded in any unicode encoding at all. But still, UTF-8 is recommanded to be used when text represatation is needed.

Convert to a \NUL terminated PrimArray,

There's an invariance that this array never contains extra \NUL except terminator.

Construct a CBytes from arbitrary array, result will be trimmed down to first \NUL byte if there's any.

Construct a CBytes from a MutablePrimArray.

Result will be shrinked to first \NUL byte without copy. If there is no \NUL found in the array, We will resize the origin MutablePrimArray, so, to avoid undefined behaviour, the original MutablePrimArray shall not be accessed anymore. Moreover, no reference to the old one should be kept in order to allow garbage collection of the original MutablePrimArray in case a new MutablePrimArray had to be allocated.

O(1), convert to Bytes, which can be processed by vector combinators.

O(1), convert to Bytes with its NULL terminator.

O(n), convert from Bytes

Result will be trimmed down to first \NUL byte if there's any.

O(n), convert to Text using UTF8 encoding assumption.

Throw InvalidUTF8Exception in case of invalid codepoint.

O(n), convert to Text using UTF8 encoding assumption.

Return Nothing in case of invalid codepoint.

O(n), convert from Text,

Result will be trimmed down to first \NUL byte if there's any.

Write CBytes 's byte sequence to buffer.

This function is different from Print instance in that it directly write byte sequence without checking if it's UTF8 encoded.

Write CBytes 's byte sequence to buffer, with its NULL terminator.

Build a CBytes with builder, will automatically be trimmed down to first \NUL byte if there's any, or append with one if there's none.

Pack a String into CBytes.

\NUL is encoded as two bytes C0 80 , \xD800 ~ \xDFFF is encoded as a three bytes normal UTF-8 codepoint.

O(n) Convert cbytes to a char list using UTF8 encoding assumption.

This function is much tolerant than toText, it simply decoding codepoints using UTF8 decodeChar without checking errors such as overlong or invalid range.

Unpacking is done lazily. i.e. we will retain reference to the array until all element are consumed.

This function is a good producer in the sense of build/foldr fusion.

Return True if CBytes is empty.

O(1), Return the BTYE length of CBytes without NULL terminator.

Empty CBytes

Singleton CBytes.

Concatenate two CBytes.

O(n) Concatenate a list of CBytes.

O(n) The intercalate function takes a CBytes and a list of CBytes s and concatenates the list after interspersing the first argument between each element of the list.

Note: intercalate will force the entire CBytes list.

O(n) An efficient way to join CByte s with a byte.

Intercalate bytes list with \NUL will effectively leave the first bytes in the list.

Copy a CString type into a CBytes, return empty if the pointer is NULL.

After copying you're free to free the CString 's memory.

Same with fromCString, but only take at most N bytes.

Result will be trimmed down to first \NUL byte if there's any.

Run FFI in bracket and marshall std::string* result into CBytes, memory pointed by std::string* will be delete ed.

Pass CBytes to foreign function as a const char*.

USE THIS FUNCTION WITH UNSAFE FFI CALL ONLY.

Pass CBytes to foreign function as a const char*.

Don't pass a forever loop to this function, see #14346.

Create a CBytes with IO action.

If (<=0) capacity is provided, a pointer pointing to \NUL is passed to initialize function and empty will be returned. This behavior is different from allocCBytes, which may cause trouble for some FFI functions.

USE THIS FUNCTION WITH UNSAFE FFI CALL ONLY.

Create a CBytes with IO action.

If (<=0) capacity is provided, a nullPtr is passed to initialize function and empty will be returned. Other than that, User have to make sure a \NUL ternimated string will be written.

Pass CBytes list to foreign function as a StgArrBytes**.

Enable UnliftedFFITypes extension in your haskell code, use StgArrBytes**(>=8.10) or StgMutArrPtrs*(<8.10) pointer type and HsInt to marshall BAArray# and Int arguments on C side, check the example with BAArray#.

USE THIS FUNCTION WITH UNSAFE FFI CALL ONLY.

Pass CBytes list to foreign function as a const char**.

Check Z.Foreign module for more detail on how to marshall params in C side.

Poke CBytes with \NUL terminator.

Poke CBytes until a \NUL terminator(or to the end of the array if there's none).

A C string is a reference to an array of C characters terminated by NUL.