[
https://issues.apache.org/jira/browse/AVRO-1704?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15271360#comment-15271360
]
Doug Cutting commented on AVRO-1704:
------------------------------------
A few more comments:
- I think we can move all of the SpecificRecord#toBytes() and #fromBytes() code
to SpecificRecordBase instead of generating it for each class. I prefer to
minimize generated code. This might look like:{code}
public class SpecificRecordBase<T extends SpecificRecordBase> {
...
public T fromBytes(byte[]) { return (T)...; }
}
public class Player extends SpecificRecordBase<Player> {
...
}
{code}
- I suspect using DataInputStream and DataOutputStream in public APIs may be
problematic for performance long-term. Maybe the only public API in the first
version should be 'T fromMessage(byte[])' and 'byte[] toMessage(T)'? This can
then be optimized, and, if needed a higher-performance lower-level API can be
added.
- We should implement this API for more than just specific data. This should
work for generic data, Thrift, protobuf, etc., producing an identical format.
So the base implementation should be passed a GenericData, which all of these
inherit from, since it can create an appropriate DatumReader or DatumWriter.
So this might look something like:{code}
package org.apache.avro.data;
public class MessageCoder<T> {
private GenericData data;
public MessageCoder(GenericData data, MessageSchemaRepo repo) { this.data =
data; }
public byte[] toMessage(T object) { ... }
public T fromMessage(byte[] bytes) { ... }
}{code}
- Permitting alternate schema repos and alternate in-memory object
representations is important, but supporting alternate message formats is not.
The goal here is to standardize a message format, so I would not design things
for extensibility on that axis.
> 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)
