phoenix-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Thomas D'Silva (JIRA)" <>
Subject [jira] [Updated] (PHOENIX-2088) Prevent splitting and recombining select expressions for MR integration
Date Tue, 21 Jul 2015 20:40:04 GMT


Thomas D'Silva updated PHOENIX-2088:
    Attachment: PHOENIX-2088-4.4-HBase-0.98-v2.patch

After talking offline with [~jamestaylor], we decided to bring back ColumnInfoToStringEncoderDecoder,
and change it to use one config value per columnInfo for serialization. I have uploaded a
patch with my latest changes.

[~jmahonin] I refactored and now there is a new method setPhysicalTableName that is used to
create local indexes by IndexTool. For spark setPhysicalTableName can be called with the same
input as setOutputTableName. 
The spark tests are failing because of the way ColumnInfoToStringEncoderDecoder serializes
the column info
Any pointers on how to fix this?

 Cause: org.apache.hadoop.conf.Configuration
Serialization stack:
	- object not serializable (class: org.apache.hadoop.conf.Configuration, value: Configuration:
core-default.xml, core-site.xml, mapred-default.xml, mapred-site.xml, yarn-default.xml, yarn-site.xml,
hdfs-default.xml, hdfs-site.xml, hbase-default.xml, hbase-site.xml)
	- field (class: org.apache.phoenix.spark.DataFrameFunctions$$anonfun$1, name: config$1, type:
class org.apache.hadoop.conf.Configuration)
	- object (class org.apache.phoenix.spark.DataFrameFunctions$$anonfun$1, <function1>)

[] I have fixed the pig test failure.

> Prevent splitting and recombining select expressions for MR integration
> -----------------------------------------------------------------------
>                 Key: PHOENIX-2088
>                 URL:
>             Project: Phoenix
>          Issue Type: Bug
>            Reporter: James Taylor
>            Assignee: Thomas D'Silva
>         Attachments: PHOENIX-2088-4.4-HBase-0.98-v2.patch, PHOENIX-2088-4.4-HBase-0.98.patch,
PHOENIX-2088-pig.patch, PHOENIX-2088-wip-v2.patch, PHOENIX-2088-wip-v3.patch, PHOENIX-2088-wip.patch
> We currently send in the select expressions for the MR integration with a delimiter separated
string, split based on the delimiter, and then recombine again using a comma separator. This
is problematic because the delimiter character may appear in a select expression, thus breaking
this logic. Instead, we should use a comma as the delimiter and avoid splitting and recombining
as it's not necessary in that case. Instead, the entire string can be used as-is in that case
to form the select expressions.

This message was sent by Atlassian JIRA

View raw message