You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@avro.apache.org by "ASF subversion and git services (Jira)" <ji...@apache.org> on 2022/03/08 18:35:00 UTC
[jira] [Commented] (AVRO-3370) [Spec] Inconsistent behaviour on types as invalid names.
[ https://issues.apache.org/jira/browse/AVRO-3370?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17503111#comment-17503111 ]
ASF subversion and git services commented on AVRO-3370:
-------------------------------------------------------
Commit 4b8dac4d0f210dce0b37d9f5c990d265f4ffc20b in avro's branch refs/heads/master from Ryan Skraba
[ https://gitbox.apache.org/repos/asf?p=avro.git;h=4b8dac4 ]
AVRO-3370: Clarify spec with regards to permitted names (#1573)
* AVRO-3370: Clarify spec with regards to permitted names
* Suggestion when to use a namespace
> [Spec] Inconsistent behaviour on types as invalid names.
> --------------------------------------------------------
>
> Key: AVRO-3370
> URL: https://issues.apache.org/jira/browse/AVRO-3370
> Project: Apache Avro
> Issue Type: Bug
> Reporter: Ryan Skraba
> Assignee: Ryan Skraba
> Priority: Major
> Labels: pull-request-available
> Fix For: 1.11.1
>
> Time Spent: 2h 20m
> Remaining Estimate: 0h
>
> We've run across this in some code that interoperates between Java and Python.
> The spec [currently forbids|https://avro.apache.org/docs/current/spec.html#names] using a primitive type name as a keyword: _*Primitive type names have no namespace and their names may not be defined in any namespace.*_
> {code:java}
> {"type":"record","name":"long","fields":[{"name":"a1","type":"long"}]} {code}
> That fails in Java with {{"org.apache.avro.AvroTypeException: Schemas may not be named after primitives: long"}}
> What do we expect to happen when a named schema uses a complex type?
> {code:java}
> {"type":"record","name":"record","fields":[{"name":"a1","type":"long"}]} {code}
> This currently *succeeds* in Java and the schema can be used to serialize and deserialize data.
> This currently *fails* in Python with: {{avro.schema.SchemaParseException: record is a reserved type name}}
> Which one is the correct behaviour?
> This gets a bit more complicated when we consider using the name as a reference.
> The following two schemas both work in Java:
> {code:java}
> {"type":"record","name":"LinkedList",
> "fields":[
> {"name":"value","type":"int},
> {"name":"next","type":["null","LinkedList"]}]}" {code}
> {code:java}
> {"type":"record","name":"LinkedList",
> "fields":[
> {"name":"value","type":"int},
> {"name":"next","type":["null",{"type":"LinkedList"}]}]}"
> {code}
> If we rename {{LinkedList}} to {{record}} the former succeeds in Java and the latter fails with {{{}org.apache.avro.SchemaParseException: No name in schema: {"type":"record"{}}}}
> {*}Edit{*}: The consensus on the [mailing list|https://lists.apache.org/thread/0wmgyx6z69gy07lvj9ndko75752b8cn2] is the "permissive" behaviour of Java should be adopted, in order to align the SDKs. The specification doesn't currently forbid these, and this should be clarified explicitly. We should probably say that it's a best practice to avoid doing this, especially in the null namespace, since it can be confusing to a reader and potentially cause ambiguities when JSON encoding data.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)