[
https://issues.apache.org/jira/browse/PIG-1229?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12852190#action_12852190
]
Ashutosh Chauhan commented on PIG-1229:
---------------------------------------
Few suggestions:
Reading from test case, currently store statements look like:
{code}
b = store a into 'dummy' using
org.apache.pig.piggybank.storage.DBStorage('org.hsqldb.jdbcDriver','jdbc:hsqldb:file:/tmp/batchtest;hsqldb.default_table_type=cached;hsqldb.cache_rows=100','insert
into a...');
{code}
here 'dummy' is totally ignored. while this works, from a user experience
following might be better:
{code}
b = store a into 'jdbc:hsqldb:file:/tmp/batchtest' using
org.apache.pig.piggybank.storage.DBStorage('org.hsqldb.jdbcDriver','hsqldb.default_table_type=cached;hsqldb.cache_rows=100','insert
into a');
{code}
that is, have db url as store location and second param of store func as db
params. you can use setStoreLocation() to store url. Apart from more intuitive
store stmt, this will also allow you to check whether DB is reachable or not at
compile time itself, instead of at runtime. You can do that via
checkOutputSpecs().
Doing DataType.findType() on every element of every tuple will be expensive. I
am wondering if you can get hold of schema in your store func and use that to
map pig types to sql types.
All of these suggestions may come in as later patches. So, if you want to get
this committed and track these separately I think that also will work as this
patch is functionally complete.
> allow pig to write output into a JDBC db
> ----------------------------------------
>
> Key: PIG-1229
> URL: https://issues.apache.org/jira/browse/PIG-1229
> Project: Pig
> Issue Type: New Feature
> Components: impl
> Reporter: Ian Holsman
> Assignee: Ankur
> Priority: Minor
> Fix For: 0.8.0
>
> Attachments: jira-1229-v2.patch
>
>
> UDF to store data into a DB
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
