Beacon chain block processing #1156
Conversation
|
PR is ready for merge now. Tests and docs have been added. |
| try { | ||
| MessageDigest digest = MessageDigest.getInstance("BLAKE2B-512"); | ||
| digest.update(data); | ||
| return Arrays.copyOf(digest.digest(), 32); |
There was a problem hiding this comment.
512 = 64 bytes, so you are getting only first half?
There was a problem hiding this comment.
yes, exactly. That was a Vitalik's suggestion, cause not every lib supports configurable lengths for BLAKE2b. This suggestion was agreed among implementors.
There was a problem hiding this comment.
Could you, please, add comment about it so it will not confuse
| return commonConfig.keyValueDataSource("beaconchain", settings); | ||
| } | ||
|
|
||
| private DbFlushManager beaconDbFlusher; |
There was a problem hiding this comment.
Maybe make it non-primary Bean instead?
There was a problem hiding this comment.
@Bean("beaconDbFlusher")
public DbFlushManager beaconDbFlusher() {
beaconDbFlusher = new DbFlushManager(systemProperties, Collections.emptySet(), beaconChainDbCache());
shardingWorldManager.setBeaconDbFlusher(beaconDbFlusher);
return beaconDbFlusher;
}
and mark original dbflushmanager with @Primary
There was a problem hiding this comment.
Trying to keep it distant from ethereumj-core module and do not add changes in the code of it. Actually, beaconDbFlusher is not a fully functioning flush manager, it doesn't close databases for example, this is still lays on original flush manager. That is why it's created in such a dirty way.
| } | ||
|
|
||
| public boolean isParentOf(Beacon other) { | ||
| return FastByteComparisons.equal(this.getHash(), other.getHash()); |
There was a problem hiding this comment.
Looks like equals, not parent of
|
|
||
| public static final long SLOT = 0; | ||
|
|
||
| private static final byte[] NULL = new byte[32]; |
|
|
||
| /** | ||
| * Returns max block number that is presented in the store. | ||
| * @return max number or -1 is store is empty |
| * Saves block to the storage. | ||
| * | ||
| * @param block the block | ||
| * @param chainScore total score of the chain that is calculated after given block has been improted |
| byte[] shortList = RLP.encodeList(shortItem, shortItem, shortItem); | ||
| byte[] longList = RLP.encodeList(longItem, longItem, longItem); | ||
|
|
||
| byte[] encoded = RLP.wrapList(nullArray, singleZero, singleByte, shortItem, longItem, shortList, longList); |
There was a problem hiding this comment.
I don't understand this
You have encodeElement for each data piece inside wrapList but prepared everything above with encodeElement too
I think you should put raw data instead
There was a problem hiding this comment.
Not each element of wrapping list is encoded with encodeElement. Basically, this test checks that applying wrapList and then unwrapList results in data items that equal to original data.
There was a problem hiding this comment.
I mean that you test it on safe already encoded data, while it could be used with unsafe notencoded data.
These changes are not ready for merge. PR is created in purpose of getting a feedback on general design of Beacon chain block processing part.
ToDo:
Resolves #1145