adapt to a custom connection pool

[severous]

In the official guidance given, the configuration of the data source is fixed. Therefore, we do not explicitly provide getter setter methods to set the properties of the data source. When I inject such a data source into sharding-jdbc, the following error occurs：

Caused by: java.lang.NullPointerException: Cannot invoke "Object.toString()" because the return value of "java.util.Map.get(Object)" is null
	at org.apache.shardingsphere.infra.metadata.database.resource.unit.StorageUnit.<init>(StorageUnit.java:56)
	at org.apache.shardingsphere.infra.config.database.impl.DataSourceProvidedDatabaseConfiguration.getStorageUnits(DataSourceProvidedDatabaseConfiguration.java:73)
	at org.apache.shardingsphere.infra.config.database.impl.DataSourceProvidedDatabaseConfiguration.<init>(DataSourceProvidedDatabaseConfiguration.java:54)
	at org.apache.shardingsphere.driver.jdbc.core.datasource.ShardingSphereDataSource.createContextManager(ShardingSphereDataSource.java:77)
	at org.apache.shardingsphere.driver.jdbc.core.datasource.ShardingSphereDataSource.<init>(ShardingSphereDataSource.java:67)
	at org.apache.shardingsphere.driver.api.ShardingSphereDataSourceFactory.createDataSource(ShardingSphereDataSourceFactory.java:95)
	at org.apache.shardingsphere.driver.api.ShardingSphereDataSourceFactory.createDataSource(ShardingSphereDataSourceFactory.java:78)
	at org.apache.shardingsphere.driver.api.ShardingSphereDataSourceFactory.createDataSource(ShardingSphereDataSourceFactory.java:139)

The error occurs in the following line of code because of standardProps is null

        Map<String, Object> standardProps = dataSourcePoolProperties.getConnectionPropertySynonyms().getStandardProperties();
        String url = standardProps.get("url").toString();

These properties are obtained from the DataSourcePoolReflection, which obtains them in a reflective manner. However, my custom datasource does not have these getter methods! I would like to know what conditions must be fulfilled for a custom data source

[terrymanu]

The title of this issue is not well enough, could you please update to a more understandable title by summarizing the content?

[severous]

I've updated the issue I'm asking. Please help.

[terrymanu]

What I mean is that the title isn't good enough for people with relevant expertise to get involved, as the title doesn't make it clear what the issue is about.
If the title cannot describe the relevant issue, then committers who will enter to view the content of this issue will correspondingly decrease.

[severous]

Okay. The title is updated

[terrymanu]

What is custom data source?
Did you mean you develop an new data source connection pool, not use HikariCP or DBCP?

[severous]

Yes, our custom data sources provide the ability to dynamically switch between data sources.

[terrymanu]

Based on your description, the title should be modified.
This is not related to dynamic switching; it is an issue of adapting to a custom connection pool, right?

So, is your question about how to dynamically switch data sources, or how to adapt to a custom connection pool?

[severous]

I will update the title to accommodate custom connection pooling

[terrymanu]

I apologize for the inconvenience, but apart from modifying the source code, ShardingSphere does not currently support custom data source connection pools.
Due to the wide variety and miss of standardization of data source connection pools, ShardingSphere can only support a few common ones and there are no plans at present to update this functionality.

However, ShardingSphere does support dynamic switching of data sources through the registry center.

[severous]

Thanks for your reply. Could you give me some tutorial about this？
Another thing that needs to be verified is whether sharding-jdbc gets the data source configuration in a reflective way.

[terrymanu]

If not an issue, let's move it to discussion.