tajo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jihoon Son (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (TAJO-456) Separate tajo-jdbc and tajo-client from tajo-core-backend
Date Mon, 30 Dec 2013 07:08:51 GMT

    [ https://issues.apache.org/jira/browse/TAJO-456?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13858621#comment-13858621
] 

Jihoon Son commented on TAJO-456:
---------------------------------

I met a compile error as follows.
{noformat}
src/main/proto/tajo_protos.proto: No such file or directory
{noformat}

> Separate tajo-jdbc and tajo-client from tajo-core-backend
> ---------------------------------------------------------
>
>                 Key: TAJO-456
>                 URL: https://issues.apache.org/jira/browse/TAJO-456
>             Project: Tajo
>          Issue Type: Improvement
>          Components: build
>            Reporter: Hyunsik Choi
>            Assignee: Hyunsik Choi
>             Fix For: 0.8-incubating
>
>
> Currently, tajo-client and tajo-jdbc are included in tajo-core-backend, which depends
on lots of third-party libraries. So, even just client programs should include unnecessary
third-party libraries. This patch separates tajo-jdbc and tajo-client from tajo-core-backend
to individual maven modules. As a result, the client and jdbc's dependencies are more simplified
than before.
> After this patch, 'mvn package \-Pdist' commands generates tajo jdbc drivers in $\{TAJO_HOME}/tajo-dist/target/tajo-$\{tajo.version}/share/jdbc-dist.
There will be the following files in the directory:
> {noformat}
> joda-time-2.3.jar
> tajo-catalog-common-0.8.0-SNAPSHOT.jar
> tajo-client-0.8.0-SNAPSHOT.jar
> tajo-common-0.8.0-SNAPSHOT.jar
> tajo-jdbc-0.8.0-SNAPSHOT.jar
> tajo-rpc-0.8.0-SNAPSHOT.jar
> tajo-storage-0.8.0-SNAPSHOT.jar
> {noformat}
> In order to load the Tajo JDBC driver, client programs must be able to locate the all
above *JAR* files* and *hadoop's JAR files*.
> For that, users should set classpath for them. If the jar files are located in the directory
'/usr/local/share/tajo-jdbc' and hadoop binary is located in '/opt/hadoop', you should set
classpath as follows:
> {code}
> export CLASSPATH=`/opt/hadoop/bin/hadoop classpath`:/usr/local/share/tajo-jdbc/*:$CLASSPATH
> {code}
> Note that the command '$\{HADOOP_HOME}/bin/hadoop classpath' prints out hadoop's classpaths
via stdout.



--
This message was sent by Atlassian JIRA
(v6.1.5#6160)

Mime
View raw message