[
https://issues.apache.org/jira/browse/AVRO-1142?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13455673#comment-13455673
]
Martin Nagy commented on AVRO-1142:
-----------------------------------
That sounds good, though I would maybe use "uncompressed data" or "uncompressed
data block" instead of "uncompressed block", to make sure there's no ambiguity
between serialized objects and file data block. Sections for null and deflate
codecs more or less use the same terminology for this.
In any case, thanks for fixing this :)
> Specification is misleading in regards to CRC32 checksum of Snappy blocks
> -------------------------------------------------------------------------
>
> Key: AVRO-1142
> URL: https://issues.apache.org/jira/browse/AVRO-1142
> Project: Avro
> Issue Type: Bug
> Components: spec
> Reporter: Martin Nagy
>
> The specification says:
> ??Each compressed block is followed by its 4-byte, big-endian CRC32
> checksum.??
> This, in my opinion, suggests that the CRC32 checksum is calculated from the
> compressed block, when in fact it is computed from the block before
> compression (as it is implemented in java). I've implemented a snappy codec
> for C++ and had trouble with this :(
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
