[
https://issues.apache.org/jira/browse/AVRO-1704?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15353737#comment-15353737
]
Ryan Blue commented on AVRO-1704:
---------------------------------
I agree that the current interface is wide. I think we should have the datum
reuse methods, which doubles the API. I think we definitely want the ByteBuffer
methods. Do you think we don't need the InputStream methods? In the pull
request there are also byte array methods, but it's easy for callers to use
ByteBuffer instead.
I like having the interface so that alternative implementations can be
independent. There's no guarantee that Avro's base class is useful to
implementers and I don't see a need to force people to inherit from an Avro
class when it may not make sense. There's an optional base class for
convenience, so I think the benefits outweigh the cost.
+1 for getting rid of the performance pitfalls. I think we just need to find a
reusable ByteArrayInputStream and make sure we can change the buffer list in
ByteBufferInputStream. I'll look into it.
For thread safety we can just make the reused state thread-local like you
suggest. Right now the Specific methods use a thread-local
DatumEncoder/DatumDecoder. Do you think the DatumEncoder implementations should
be thread-safe?
I think we do need the raw format. Right now there are a lot of systems already
serializing Avro records in the equivalent of the raw format so I would like to
have an Avro class that helps move to the new spec. Also, if the schema is
fixed then there's no need for 10 extra bytes per payload so it is
independently useful. For example, I use the raw format to store JSON payloads.
The schema won't change and Avro is much smaller and faster.
> Standardized format for encoding messages with Avro
> ---------------------------------------------------
>
> Key: AVRO-1704
> URL: https://issues.apache.org/jira/browse/AVRO-1704
> Project: Avro
> Issue Type: Improvement
> Reporter: Daniel Schierbeck
> Assignee: Niels Basjes
> Attachments: AVRO-1704-2016-05-03-Unfinished.patch,
> AVRO-1704-20160410.patch
>
>
> I'm currently using the Datafile format for encoding messages that are
> written to Kafka and Cassandra. This seems rather wasteful:
> 1. I only encode a single record at a time, so there's no need for sync
> markers and other metadata related to multi-record files.
> 2. The entire schema is inlined every time.
> However, the Datafile format is the only one that has been standardized,
> meaning that I can read and write data with minimal effort across the various
> languages in use in my organization. If there was a standardized format for
> encoding single values that was optimized for out-of-band schema transfer, I
> would much rather use that.
> I think the necessary pieces of the format would be:
> 1. A format version number.
> 2. A schema fingerprint type identifier, i.e. Rabin, MD5, SHA256, etc.
> 3. The actual schema fingerprint (according to the type.)
> 4. Optional metadata map.
> 5. The encoded datum.
> The language libraries would implement a MessageWriter that would encode
> datums in this format, as well as a MessageReader that, given a SchemaStore,
> would be able to decode datums. The reader would decode the fingerprint and
> ask its SchemaStore to return the corresponding writer's schema.
> The idea is that SchemaStore would be an abstract interface that allowed
> library users to inject custom backends. A simple, file system based one
> could be provided out of the box.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
