fix : fix compute hash #9
Conversation
README.md
Outdated
|
|
||
| 1. erasure package support RSEncoder which contain basic Encode and Decode reedSolomon APIs | ||
| (1) erasure package support RSEncoder which contain basic Encode and Decode reedSolomon APIs | ||
| ``` | ||
| RSEncoderStorage, err := NewRSEncoder(dataShards, parityShards, int64(blockSize)) |
There was a problem hiding this comment.
1.int64(blocksize) -> blocksize
2.Explain the meaning of blocksize?
README.md
Outdated
| // decodes the input erasure encoded data shards data. | ||
| func (r *RSEncoder) DecodeDataShards(content [][]byte) error { | ||
| func (r *RSEncoder) DecodeDataShards(content [][]byte) error { |
There was a problem hiding this comment.
ReconstructData or RecoverData seems clearer
There was a problem hiding this comment.
DecodeDataShards and DecodeShards is corresponding to EncodeShards.
I added some clarifications to explain the function
| @@ -22,7 +22,7 @@ func EncodeRawSegment(content []byte, dataShards, parityShards int) ([][]byte, e | |||
| func DecodeRawSegment(pieceData [][]byte, segmentSize int64, dataShards, parityShards int) ([]byte, error) | |||
There was a problem hiding this comment.
1.What's the difference between segmentSize and blockSize?
2.piceceData->shardDataList maybe a good choice.
There was a problem hiding this comment.
No difference, These two interfaces are applied on the SP side as the segments is called in gnfd side, so the naming is closer to the application layer
README.md
Outdated
| @@ -22,7 +22,7 @@ func EncodeRawSegment(content []byte, dataShards, parityShards int) ([][]byte, e | |||
| func DecodeRawSegment(pieceData [][]byte, segmentSize int64, dataShards, parityShards int) ([]byte, error) | |||
| ``` | |||
|
|
|||
| ### Compute sha256 hash of file content | |||
| ### 2. Compute sha256 hash of file content | |||
|
|
|||
| hash package support methods to compute hash roots of greenfield objects , the computed methods is based on | |||
| redundancy Strategy of greenfield | |||
go/hash/hash.go
Outdated
| @@ -32,51 +36,44 @@ func ComputerHash(reader io.Reader, segmentSize int64, dataShards, parityShards | |||
| } | |||
| if n > 0 { | |||
There was a problem hiding this comment.
Maybe n != segmentSize and err == nil, need retry read from io.reader.
There was a problem hiding this comment.
as the API description below , the n may be less than segmentSize (end-of-file condition) , so if the err happen, just return an error. If end-of-file condition, the n just need to be less than segmentSize .
// When Read encounters an error or end-of-file condition after successfully reading n > 0 bytes, it returns the number of bytes read
changed to condition to " if n > 0 && n <= int(segmentSize)"
go/hash/hash_test.go
Outdated
| for _, hash := range hashResult { | ||
| if len(hash) != 64 { | ||
| if len(hash) != 32 { |
| @@ -86,7 +83,7 @@ func ComputerHash(reader io.Reader, segmentSize int64, dataShards, parityShards | |||
| } | |||
There was a problem hiding this comment.
why result = append(result, hashList[i]) ?directly return hashList?
There was a problem hiding this comment.
Yes, some adjustments have been made, the result variable is no longer needed
go/hash/hash.go
Outdated
| ) | ||
|
|
||
| // ComputerHash split the reader into segment, ec encode the data, compute the hash roots of pieces | ||
| // return the hash result array list and data size | ||
| func ComputerHash(reader io.Reader, segmentSize int64, dataShards, parityShards int) ([]string, int64, error) { | ||
| func ComputerHash(reader io.Reader, segmentSize int64, dataShards, parityShards int) ([][]byte, int64, error) { |
There was a problem hiding this comment.
CompluterHash's Computer is a noun, which seems inappropriate function name.
There was a problem hiding this comment.
changed to ComputeIntegrityHash
Description
fix bugs of computing hash root