nim-codex/codex/manifest/coders.nim

180 lines
5.1 KiB
Nim
Raw Normal View History

2022-05-19 19:56:03 +00:00
## Nim-Codex
## Copyright (c) 2022 Status Research & Development GmbH
## Licensed under either of
## * Apache License, version 2.0, ([LICENSE-APACHE](LICENSE-APACHE))
## * MIT license ([LICENSE-MIT](LICENSE-MIT))
## at your option.
## This file may not be copied, modified, or distributed except according to
## those terms.
Download files without padding (#218) The initial goal of this patch was to allow to download of a file via REST API in exactly the same size as it was uploaded, which required adding fields Chunker.offset and Manifest.originalBytes to keep that size. On top of that, we added more integrity checks to operations on Manifest, and reorganized TestNode.nim to test the actual interaction between node.store and node.retrieve operations. Note that the wire format of Manifest was changed, so we need to recreate all BlockStores. * Download without padding * Fixed chunker tests * Chunker: get rid of RabinChunker * Verify offset in the chunker tests * Use manifest.originalBytesPadded in StoreStream.size * StoreStream: replace emptyBlock with zeroMem * Manifest.bytes: compute how many bytes corresponding StoreStream(Manifest, pad) will return * Manifest: verify originalBytes and originalLen on new/encode/decode Also set originalBytes in each Manifest creation/update scenario * Manifest: comments, split code into sections * Reordered parameters to deal with int64 size in 32-bit builds * TestNode.nim: combine Store and Retrieve tests 1. Instead of copy-pasting code from node.nim, new test calls node.store() and node.retrieve() in order to check that they can correctly store and then retrieve data 2. New test compares only file contents, manifest contents considered an implementation detail 3. New test chunks at odd chunkSize=BlockSize/1.618 in order to ensure that data retrieved correctly even when buffer sizes mismatch * TestNode.nim: code refactoring * Manifest.add: one more test * Manifest.verify: return Result instead of raising Defect * Node.store: added blockSize parameter
2022-08-24 12:15:59 +00:00
# This module implements serialization and deserialization of Manifest
import pkg/upraises
push: {.upraises: [].}
import std/tables
import pkg/libp2p
import pkg/questionable
import pkg/questionable/results
import pkg/chronicles
import pkg/chronos
import ./manifest
import ../errors
import ../blocktype
import ./types
proc encode*(coder: DagPBCoder, manifest: Manifest): ?!seq[byte] =
## Encode the manifest into a ``ManifestCodec``
## multicodec container (Dag-pb) for now
##
Download files without padding (#218) The initial goal of this patch was to allow to download of a file via REST API in exactly the same size as it was uploaded, which required adding fields Chunker.offset and Manifest.originalBytes to keep that size. On top of that, we added more integrity checks to operations on Manifest, and reorganized TestNode.nim to test the actual interaction between node.store and node.retrieve operations. Note that the wire format of Manifest was changed, so we need to recreate all BlockStores. * Download without padding * Fixed chunker tests * Chunker: get rid of RabinChunker * Verify offset in the chunker tests * Use manifest.originalBytesPadded in StoreStream.size * StoreStream: replace emptyBlock with zeroMem * Manifest.bytes: compute how many bytes corresponding StoreStream(Manifest, pad) will return * Manifest: verify originalBytes and originalLen on new/encode/decode Also set originalBytes in each Manifest creation/update scenario * Manifest: comments, split code into sections * Reordered parameters to deal with int64 size in 32-bit builds * TestNode.nim: combine Store and Retrieve tests 1. Instead of copy-pasting code from node.nim, new test calls node.store() and node.retrieve() in order to check that they can correctly store and then retrieve data 2. New test compares only file contents, manifest contents considered an implementation detail 3. New test chunks at odd chunkSize=BlockSize/1.618 in order to ensure that data retrieved correctly even when buffer sizes mismatch * TestNode.nim: code refactoring * Manifest.add: one more test * Manifest.verify: return Result instead of raising Defect * Node.store: added blockSize parameter
2022-08-24 12:15:59 +00:00
? manifest.verify()
var pbNode = initProtoBuffer()
# NOTE: The `Data` field in the the `dag-pb`
# contains the following protobuf `Message`
#
# ```protobuf
# Message ErasureInfo {
# optional uint32 ecK = 1; # number of encoded blocks
# optional uint32 ecM = 2; # number of parity blocks
# optional bytes originalManifest = 3; # manifest of the original dataset
# }
# Message Header {
# optional bytes treeCid = 1; # cid (root) of the tree
# optional uint32 blockSize = 2; # size of a single block
# optional uint64 datasetSize = 3; # size of the dataset
# optional ErasureInfo erasure = 4; # erasure coding info
# }
# ```
#
# var treeRootVBuf = initVBuffer()
var header = initProtoBuffer()
header.write(1, manifest.treeCid.data.buffer)
header.write(2, manifest.blockSize.uint32)
header.write(3, manifest.datasetSize.uint32)
if manifest.protected:
var erasureInfo = initProtoBuffer()
erasureInfo.write(1, manifest.ecK.uint32)
erasureInfo.write(2, manifest.ecM.uint32)
erasureInfo.write(3, manifest.interleave.uint32)
erasureInfo.write(4, ? coder.encode(manifest.originalManifest)) # TODO: fix check
erasureInfo.finish()
header.write(4, erasureInfo)
pbNode.write(1, header) # set the treeCid as the data field
pbNode.finish()
return pbNode.buffer.success
proc decode*(coder: DagPBCoder, data: openArray[byte]): ?!Manifest =
## Decode a manifest from a data blob
##
var
pbNode = initProtoBuffer(data)
pbHeader: ProtoBuffer
pbErasureInfo: ProtoBuffer
treeCidBuf: seq[byte]
originalManifest: Manifest
datasetSize: uint32
blockSize: uint32
ecK, ecM, interleave: uint32
# Decode `Header` message
if pbNode.getField(1, pbHeader).isErr:
return failure("Unable to decode `Header` from dag-pb manifest!")
# Decode `Header` contents
if pbHeader.getField(1, treeCidBuf).isErr:
return failure("Unable to decode `treeCid` from manifest!")
if pbHeader.getField(2, blockSize).isErr:
return failure("Unable to decode `blockSize` from manifest!")
if pbHeader.getField(3, datasetSize).isErr:
return failure("Unable to decode `datasetSize` from manifest!")
Download files without padding (#218) The initial goal of this patch was to allow to download of a file via REST API in exactly the same size as it was uploaded, which required adding fields Chunker.offset and Manifest.originalBytes to keep that size. On top of that, we added more integrity checks to operations on Manifest, and reorganized TestNode.nim to test the actual interaction between node.store and node.retrieve operations. Note that the wire format of Manifest was changed, so we need to recreate all BlockStores. * Download without padding * Fixed chunker tests * Chunker: get rid of RabinChunker * Verify offset in the chunker tests * Use manifest.originalBytesPadded in StoreStream.size * StoreStream: replace emptyBlock with zeroMem * Manifest.bytes: compute how many bytes corresponding StoreStream(Manifest, pad) will return * Manifest: verify originalBytes and originalLen on new/encode/decode Also set originalBytes in each Manifest creation/update scenario * Manifest: comments, split code into sections * Reordered parameters to deal with int64 size in 32-bit builds * TestNode.nim: combine Store and Retrieve tests 1. Instead of copy-pasting code from node.nim, new test calls node.store() and node.retrieve() in order to check that they can correctly store and then retrieve data 2. New test compares only file contents, manifest contents considered an implementation detail 3. New test chunks at odd chunkSize=BlockSize/1.618 in order to ensure that data retrieved correctly even when buffer sizes mismatch * TestNode.nim: code refactoring * Manifest.add: one more test * Manifest.verify: return Result instead of raising Defect * Node.store: added blockSize parameter
2022-08-24 12:15:59 +00:00
if pbHeader.getField(4, pbErasureInfo).isErr:
return failure("Unable to decode `erasureInfo` from manifest!")
let protected = pbErasureInfo.buffer.len > 0
if protected:
if pbErasureInfo.getField(1, ecK).isErr:
return failure("Unable to decode `K` from manifest!")
if pbErasureInfo.getField(2, ecM).isErr:
return failure("Unable to decode `M` from manifest!")
if pbErasureInfo.getField(3, interleave).isErr:
return failure("Unable to decode `interleave` from manifest!")
var buffer = newSeq[byte]()
if pbErasureInfo.getField(4, buffer).isErr:
return failure("Unable to decode `originalManifest` from manifest!")
originalManifest = coder.decode(buffer).get # TODO: fix check
let
treeCid = ? Cid.init(treeCidBuf).mapFailure
let
self = if protected:
Manifest.new(
treeCid = treeCid,
datasetSize = datasetSize.NBytes,
blockSize = blockSize.NBytes,
version = treeCid.cidver,
hcodec = (? treeCid.mhash.mapFailure).mcodec,
codec = treeCid.mcodec,
ecK = ecK.int,
ecM = ecM.int,
interleave = interleave.int, #TODO
originalManifest = originalManifest
)
else:
Manifest.new(
treeCid = treeCid,
datasetSize = datasetSize.NBytes,
blockSize = blockSize.NBytes,
version = treeCid.cidver,
hcodec = (? treeCid.mhash.mapFailure).mcodec,
codec = treeCid.mcodec
)
Download files without padding (#218) The initial goal of this patch was to allow to download of a file via REST API in exactly the same size as it was uploaded, which required adding fields Chunker.offset and Manifest.originalBytes to keep that size. On top of that, we added more integrity checks to operations on Manifest, and reorganized TestNode.nim to test the actual interaction between node.store and node.retrieve operations. Note that the wire format of Manifest was changed, so we need to recreate all BlockStores. * Download without padding * Fixed chunker tests * Chunker: get rid of RabinChunker * Verify offset in the chunker tests * Use manifest.originalBytesPadded in StoreStream.size * StoreStream: replace emptyBlock with zeroMem * Manifest.bytes: compute how many bytes corresponding StoreStream(Manifest, pad) will return * Manifest: verify originalBytes and originalLen on new/encode/decode Also set originalBytes in each Manifest creation/update scenario * Manifest: comments, split code into sections * Reordered parameters to deal with int64 size in 32-bit builds * TestNode.nim: combine Store and Retrieve tests 1. Instead of copy-pasting code from node.nim, new test calls node.store() and node.retrieve() in order to check that they can correctly store and then retrieve data 2. New test compares only file contents, manifest contents considered an implementation detail 3. New test chunks at odd chunkSize=BlockSize/1.618 in order to ensure that data retrieved correctly even when buffer sizes mismatch * TestNode.nim: code refactoring * Manifest.add: one more test * Manifest.verify: return Result instead of raising Defect * Node.store: added blockSize parameter
2022-08-24 12:15:59 +00:00
? self.verify()
self.success
proc encode*(
self: Manifest,
encoder = ManifestContainers[$DagPBCodec]
): ?!seq[byte] =
## Encode a manifest using `encoder`
##
encoder.encode(self)
func decode*(
_: type Manifest,
data: openArray[byte],
decoder = ManifestContainers[$DagPBCodec]
): ?!Manifest =
## Decode a manifest using `decoder`
##
decoder.decode(data)
func decode*(_: type Manifest, blk: Block): ?!Manifest =
## Decode a manifest using `decoder`
##
if not ? blk.cid.isManifest:
return failure "Cid not a manifest codec"
Manifest.decode(
blk.data,
? ManifestContainers[$(?blk.cid.contentType().mapFailure)].catch)