the size of the hash output

Sum appends the current hash to b and returns the resulting slice. It does not change the underlying hash state. Note that this is to allow Sum() to r

Uses all default hashes (MD5 & rollsum16)

cannot be called concurrently

Builds an index in which chunks can be found, with their corresponding offsets We use this for the

get the data that was evicted by the last write

Sorted list of blocks, based on StartBlock

Creates a list of spans that are missing. note that maxBlock is blockCount-1

Can be used on multiple streams of results simultaneously starts working asyncronously, call from the initiating goroutine

Splits successive strings into blocks of size n 2, ABCD -> AB, CD 2, ABCD, E, FG -> AB, CD, E, FG

Add a single byte into the rollsum

get the current buffer contents of block

Gets the Hash function for the strong hash used on each block defaults to md5, but can be overriden by the generator

Loads chunks from a reader, assuming alternating weak then strong hashes

GenerateChecksums reads each block of the input file, and outputs first the weak, then the strong checksum to the output writer. It will return a chec

If the block source encounters an unsurmountable problem

Rollsum32Base decouples the rollsum algorithm from the implementation of hash.Hash and the storage the rolling checksum window this allows us to write

Generates an index from a reader This is mostly a utility function to avoid being overly verbose in tests that need an index to work, but don't want t

Puts the sum into b. Avoids allocation. b must have length >= 4

Remove a byte from the end of the rollsum Use the previous length (before removal)

* This simple example currently doesn't do any pipelining of needed blocks, nor does it deal with blocks being delivered out of order. */

the total written, up to the blocksize

if merged, the block span remaining is the one with the lower start block

compares a checksum to another based on the checksums, not the offset

Patch the files

Split blocks into chunks of the desired size, or less. This implementation assumes a fixed block size at the source.

Shortens the content of the circular buffer and returns the content removed

pairs up bytes to do remove/add in the right order

GetBlockSize gets the size of each block

the number of bytes

reads the file headers and checks the magic string, then the semantic versioning

Close - close open files, copy to the final location from a temporary one if neede

This method is called on multiple goroutines, and must support simultaneous requests

If the block source encounters an unsurmountable problem

Gets the Hash function for the overall file used on each block defaults to md5

GetFileSize gets the file size of the file