[jira] [Commented] (PHOENIX-6426) CsvBulkLoadTool should be able to parse the output of sqlline generated csv

["Sameer Pandit (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/PHOENIX-6426?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17540340#comment-17540340 ] 

Sameer Pandit commented on PHOENIX-6426:
----------------------------------------

[~shahrs87] I tried reproducing

generated data from sqlline via record as below, using examples WEB state table.
{code:java}
Saving all output to "/mnt/c/Users/sameer.pandit/work/openSourceCohort/phoenix/bin/data.csv". Enter "record" with no arguments to stop it.
'DOMAIN','AVERAGE_CPU_USAGE','AVERAGE_DB_USAGE'
'Salesforce.com','260.7272','257.6363'
'Google.com','212.875','213.75'
'Apple.com','114.1111','119.5555'
3 rows selected (0.01 seconds) {code}
schema to insert the above data.
{code:java}
CREATE TABLE IF NOT EXISTS WEB_USAGE (
              DOMAIN VARCHAR NOT NULL PRIMARY KEY,
              AVERAGE_CPU_USAGE FLOAT,
              AVERAGE_DB_USAGE FLOAT);{code}
Error while loading using psql 
{code:java}
22/05/21 08:43:05 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
no rows upserted
Time: 1.854 sec(s)csv columns from database.
22/05/21 08:43:12 WARN impl.MetricsSystemImpl: Phoenix metrics system already initialized!
java.lang.IllegalArgumentException: Table DATA not found
        at org.apache.phoenix.util.SchemaUtil.generateColumnInfo(SchemaUtil.java:996)
        at org.apache.phoenix.util.CSVCommonsLoader.buildColumnInfoList(CSVCommonsLoader.java:253)
        at org.apache.phoenix.util.CSVCommonsLoader.upsert(CSVCommonsLoader.java:202)
        at org.apache.phoenix.util.CSVCommonsLoader.upsert(CSVCommonsLoader.java:178)
        at org.apache.phoenix.util.PhoenixRuntime.main(PhoenixRuntime.java:322) 

{code}
Can you confirm is this the bug or you seeing some different error while loading?

Is the csvbulkloadtool you mention psql.py or using https://phoenix.apache.org/bulk_dataload.html?

> CsvBulkLoadTool should be able to parse the output of sqlline generated csv
> ---------------------------------------------------------------------------
>
>                 Key: PHOENIX-6426
>                 URL: https://issues.apache.org/jira/browse/PHOENIX-6426
>             Project: Phoenix
>          Issue Type: Improvement
>            Reporter: Gokcen Iskender
>            Assignee: Sameer Pandit
>            Priority: Minor
>
> Sqlline generates the csv file in a format that csvbulkloadtool cannot parse.
> [~rushabh.shah] recently hit this problem



--
This message was sent by Atlassian Jira
(v8.20.7#820007)