Hello everyone!
I have started work on HIVE-1555 feature.
There are few questions:
1) Current module structure looks like:
├── jdbc-handler.iml
├── pom.xml
└── src
├── java
│ └── org
│ └── apache
│ └── hive
│ └── storagehandler
│ ├── JDBCStorageHandler.java
│ ├── db
│ │ ├── HiveJDBCTypeBridge.java
│ │ ├── HiveJDBCVendorBridge.java
│ │ └── mysql
│ │ ├── HiveMySqlTypeBridge.java
│ │ └── HiveMySqlVendorBridge.java
│ └── serde
│ └── JDBCSerde.java
├── main
│ └── resources
└── test
└── java
pom.xml fragment:
<build>
<sourceDirectory>${basedir}/src/java</sourceDirectory>
<testSourceDirectory>${basedir}/src/test</testSourceDirectory>
<plugins>
<plugin>
<groupId>org.apache.maven.plugins</groupId>
<artifactId>maven-jar-plugin</artifactId>
<executions>
<execution>
<goals>
<goal>test-jar</goal>
</goals>
</execution>
</executions>
</plugin>
</plugins>
</build>
So this module builds fine, but i could not see jdbc-handler-XXX-.jar inside
lib directory.
What should i add in assembly build.xml?
2) What is the main difference between HiveStoragePredicateHandler and
HiveStorageHandler? I’ve found that we need to implement
HiveStoragePredicateHandler if storage handler needs support decomposition of
predicates being pushed down into table scans. Is it true that RDBMS does not
have this feature and i need to implement HiveStorageHandler?
3) I’ve found many useful classes like
DBInputFormat,DBOutputFormat,{Oracle|Mysql*.Reader} which exists in
org.apache.hadoop.mapreduce.lib.db but org.apache.hadoop packages are excluded
from build. Should i use this classes in my code? Should i need to exclude
filtering for org.apache.hadoop group?
Thanks,
Dmitry Zagorulkin
