A C library for handling dynamic, binary chunks of data.
DataBlock is a C-header library that provides a unique data structure called a DataBlock, intended for easy handling of chunks of bytes, or binary data. DataBlock is packaged as a header to make it easy to embed and include in your project.
To use DataBlock, simply include the src/Datablock.h file in the source directory.
To build the test suite, in the directory, run make all, then run the test executable, ./bin/DataBlockTest.
DataBlock is based around a model of a two-way linked list, except every node also functions as a buffer itself. The blocks are singly allocated sections of data which had a header portion, that includes the previous block, next block, the capacity, and the length, and the body, which contains the actual binary data.
The DataBlock structure is implemented like this.
struct DataBlock
{
struct DataBlock* next;
struct DataBlock* prev;
size_t cap;
size_t len;
unsigned char data[0];
};The entire block is allocated into a single block of memory. This allows it to be variably sized.
Additionally, blocks can be written into one another, such as with the function DataBlock_put_block().
DataBlock's can be linked together in a chain, to form a stream-like data structure. DataBlock's can be used to implement a dynamic stream of data, that can be easily cut or expanded when needed.
For example, the DataBlock_print_debug() method prints out something like this.
___Block_1_______
Header::
@self = 0x7ff3aac02660
@next = 0x7ff3aac02690
@prev = 0x0
@length = 5
@capacity = 10
Body::
(0.) -> 5
(1.) -> 5
(2.) -> 5
(3.) -> 5
(4.) -> 5
___Block_1_end___
___Block_2_______
Header::
@self = 0x7ff3aac02690
@next = 0x7ff3aac026c0
@prev = 0x7ff3aac02660
@length = 10
@capacity = 15
Body::
(0.) -> 66
(1.) -> 66
(2.) -> 66
(3.) -> 66
(4.) -> 66
(5.) -> 66
(6.) -> 66
(7.) -> 66
(8.) -> 66
(9.) -> 66
___Block_2_end___
___Block_3_______
Header::
@self = 0x7ff3aac026c0
@next = 0x0
@prev = 0x7ff3aac02690
@length = 10
@capacity = 15
Body::
(0.) -> 66
(1.) -> 66
(2.) -> 66
(3.) -> 66
(4.) -> 66
(5.) -> 66
(6.) -> 66
(7.) -> 66
(8.) -> 66
(9.) -> 66
___Block_3_end___
The header of each DataBlock contains info about the block, its next and previous sister blocks, its length, and it's capacity.
The len property of each block can be reset to zero, allowing the existing data to be overwritten, which can conserve space.
DataBlock is licensed under the MIT license.