Home | About | Sematext search-lucene.com search-hadoop.com
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB
 Search Hadoop and all its subprojects:

Switch to Threaded View
Avro >> mail # user >> imports and avpr files


Copy link to this message
-
Re: imports and avpr files
I'm not using a parser. I'm just using the avro compiler to generate source
files which I use in my code (I'm running mapreduce jobs whose
inputs/outputs are avro files).
J

On Thu, Feb 16, 2012 at 8:18 AM, Doug Cutting <[EMAIL PROTECTED]> wrote:

> On 02/16/2012 07:06 AM, Jeremy Lewi wrote:
> > I have an avpr file and I'd like to include other avpr files so that I
> > can split my schemas into files with one file per namespace. Is this
> > possible or do I need to use Avro IDL if I want to use imports.
>
> You could use the same API that the IDL parser uses.  It copies each of
> the types and messages from included protocols to a new protocol
> instance (http://s.apache.org/wj).
>
> Perhaps we should add a Protocol.Parser API like Schema.Parser.  One
> could then repeatedly call parser.parseProtocol(avprFile) then call
> parser.getProtocol() to return a protocol with all of the types and
> methods.  Types defined in earlier files would be available for
> reference by subsequent files.  This would make things somewhat simpler
> than the current API does.  If that sounds like it would be useful,
> please file an issue in Jira.
>
> Doug
>
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB