[
https://issues.apache.org/jira/browse/HCATALOG-275?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sushanth Sowmyan updated HCATALOG-275:
--------------------------------------
Description:
1) JsonSerDe uses the column names specified in the data instead of the schema
for serialization : The JSON SerDe uses column names as specified in the data
for serialization instead of relying on the schema provided during initialize.
This is good for position-independent serialization if the object were to not
be so, but it breaks when the objectinspector being provided to the serializer
does not use the same names. A case in point is with the RCFile Object
Inspector, which uses generic "col0", "col1" names for the top level object.
2) Does not handle null complex fields. nulls for simple types are handled, but
not for complex types. SerDe expects empty array( [] ) or empty map( {} ) or
empty struct( {} ) instead of null.
was:
1) JsonSerDe uses the column names specified in the data instead of the schema
for serialization : The JSON SerDe uses column names as specified in the data
for serialization instead of relying on the schema provided during initialize.
This is good for position-independent serialization if the object were to not
be so, but it breaks when the objectinspector being provided to the serializer
does not use the same names. A case in point is with the RCFile Object
Inspector, which uses generic "col0", "col1" names for the top level object.
2) Does not handle null complex fields. nulls for simple types are handled, but
not for complex types. SerDe expects empty array or empty map or empty struct (
{} , [] , {} ) instead of null.
> JSON SerDe issues
> -----------------
>
> Key: HCATALOG-275
> URL: https://issues.apache.org/jira/browse/HCATALOG-275
> Project: HCatalog
> Issue Type: Bug
> Affects Versions: 0.4
> Reporter: Sushanth Sowmyan
> Assignee: Sushanth Sowmyan
> Fix For: 0.4
>
> Attachments: HCATALOG-275.patch, HCATALOG-275.patch.2
>
>
> 1) JsonSerDe uses the column names specified in the data instead of the
> schema for serialization : The JSON SerDe uses column names as specified in
> the data for serialization instead of relying on the schema provided during
> initialize. This is good for position-independent serialization if the object
> were to not be so, but it breaks when the objectinspector being provided to
> the serializer does not use the same names. A case in point is with the
> RCFile Object Inspector, which uses generic "col0", "col1" names for the top
> level object.
> 2) Does not handle null complex fields. nulls for simple types are handled,
> but not for complex types. SerDe expects empty array( [] ) or empty map( {} )
> or empty struct( {} ) instead of null.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
