[
https://issues.apache.org/jira/browse/PHOENIX-3696?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Marcin Januszkiewicz updated PHOENIX-3696:
--
Description:
If a phoenix table contains a dot, then the table cannot be used with the
phoenix-spark plugin. Given a simple table:
{code}
CREATE TABLE "dot" ("key" VARCHAR PRIMARY KEY, "first.name" VARCHAR)
{code}
The following code will fail:
{code}
val df = sqlContext.load("org.apache.phoenix.spark", Map("table" ->
SchemaUtil.getEscapedArgument("dot"), "zkUrl" -> quorumAddress))
//This will crash
df.show()
{code}
I ran into this in the phoenix-for-cloudera fork for version 4.9, but tested
that this is also a problem on the master branch in the main project.
was:
If a phoenix table contains a dot, then the table cannot be used with the
phoenix-spark plugin. Given a simple table:
{code}
CREATE TABLE "dot" ("key" VARCHAR PRIMARY KEY, "first.name" VARCHAR)
{code}
The following code will fail:
{code}
val df = sqlContext.load("org.apache.phoenix.spark", Map("table" ->
SchemaUtil.getEscapedArgument("dot"), "zkUrl" -> quorumAddress))
//This will crash
df.show()
{code}
> phoenix-spark plugin doesn't handle column names with dots properly
> ---
>
> Key: PHOENIX-3696
> URL: https://issues.apache.org/jira/browse/PHOENIX-3696
> Project: Phoenix
> Issue Type: Bug
>Reporter: Marcin Januszkiewicz
>
> If a phoenix table contains a dot, then the table cannot be used with the
> phoenix-spark plugin. Given a simple table:
> {code}
> CREATE TABLE "dot" ("key" VARCHAR PRIMARY KEY, "first.name" VARCHAR)
> {code}
> The following code will fail:
> {code}
> val df = sqlContext.load("org.apache.phoenix.spark", Map("table" ->
> SchemaUtil.getEscapedArgument("dot"), "zkUrl" -> quorumAddress))
> //This will crash
> df.show()
> {code}
> I ran into this in the phoenix-for-cloudera fork for version 4.9, but tested
> that this is also a problem on the master branch in the main project.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
