Hi liang,

I am very glad to know the direction of future development for shardingsphere. 

the interface about database scaling should be define or not?

I am working for sharding-scaling project,  I think the interface definition 
about database scaling is good for this project.

> 在 2019年9月28日,下午9:34,Juan Pan <[email protected]> 写道:
> 
> Making feature independent and use them in mixed mood?
> It sounds great, but challenging. It will take a lot of time to do, i think...
> 
> 
> 
> 
> Juan Pan
> 
> 
> [email protected]
> Juan Pan(Trista), Apache ShardingSphere
> On 09/28/2019 18:44, Sheng Wu wrote:
> Does making this change in 5 make more sense?
> And we provide v4 unchanged for user convenient.
> 
> Sheng
> 
> [email protected] <[email protected]>于2019年9月28日 周六下午6:15写道:
> 
>> The scope of ShardingSphere keep expanding. Sharding is not the unique and
>> core feature for ShardingSphere anymore.
>> 
>> So we plan to create pluggable infrastructure for database proxy and JDBC
>> driver which to let users run it without any additional function, just
>> transparent transmission. ShardingSphere can add more features into `EMPTY`
>> infrastructure, such as sharding, master-slave, encrypt,
>> distributed-transaction, orchestration and so on.
>> 
>> The API may look like: ShardingDataSource, MasterSlaveDataSource,
>> EncryDataSource, DistributedTransactionDataSource and
>> OrchestrationDataSource, the yaml and other configuration methods should
>> change to this way too.
>> 
>> Every features are independent and can be work together which using
>> composited and append-able way. We can provide SSDataSourceFacade to manage
>> how to use them together.
>> 
>> This is one of the reason to why we do not release for 4.0.0 stable version
>> soon, we want more discuss to make decision to change API before 4.0.0
>> stable version release.
>> 
>> Any suggestions?
>> 
>> ------------------
>> 
>> Liang Zhang (John)
>> Apache ShardingSphere & Dubbo
>> 
> --
> Sheng Wu
> SkyWalking, Shardingsphere and Zipkin

Reply via email to