I know what you're writing, please check the code which will answer your
questions: https://github.com/apache/spark/pull/29024
The API provides quite a freedom when a custom provider is implemented.

G


On Wed, Sep 30, 2020 at 8:29 PM Artemis User <[email protected]> wrote:

> I'm just curious in regard to what this JDBC connection provider does.  If
> just read data from a database to Spark, wouldn't it be just using the
> existing JDBC data source?
> http://spark.apache.org/docs/latest/sql-data-sources-jdbc.html
>
> BTW, OAuth is a web authentication protocol, not part of the JDBC.  If I
> understand this correctly, you use OAuth to gain user access at the web
> portal level, but use DBMS authentication at the JDBC level.
>
> -- ND
> On 9/30/20 2:11 PM, Gabor Somogyi wrote:
>
> Not sure there is already a way. I'm just implementing JDBC connection
> provider which will make it available in 3.1. Just to be clear when the API
> is available custom connection provider must be implemented.
>
> With actual Spark one can try to write a driver wrapper which does the
> authentication.
>
> G
>
>
>
> On Wed, 30 Sep 2020, 19:55 KhajaAsmath Mohammed, <[email protected]>
> wrote:
>
>> Hi,
>>
>> I am looking for some information on how to read database which has oauth
>> authentication with spark -jdbc. any links that point to this approach
>> would be really helpful
>>
>> Thanks,
>> Asmath
>>
>

Reply via email to