Optimize #hashCode() and #equals() for Protocol and Schema
----------------------------------------------------------
Key: AVRO-63
URL: https://issues.apache.org/jira/browse/AVRO-63
Project: Avro
Issue Type: Improvement
Components: java
Reporter: Doug Cutting
Fix For: 1.0.0
It's often useful to cache things derived from a Schema or Protocol in a
WeakHashMap. Each cache lookup involves at least one call to #equals() and
#hashCode() on the Schema or Protocol. But the Schema#hashCode() is expensive,
since it must walk the entire schema, allocating objects to detect circular
schemas. It would be better if Schema cached its hash code in a field. This
can also be used optimize Schema#equals(): if two Schema's do not have the same
hashCode then they cannot be equal. With these optimizations, the only slow
case would be a cache lookup of an equal but not identical Schema. Perhaps
even this can be optimized by caching its MD5 and assuming that, if the MD5
does not match, schemas are not equal.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
