What geo-database features are/will be supported?

[ConstantB]

Hi,
nice work!

I have the following two questions:

1. Are there plans to support also the SQL simple features specification (http://www.opengeospatial.org/standards/sfs) that is supported geospatial databases? For example, it would be useful if we could pose a spatial join with an ST_Overlaps condition, e.g., to retrieve overlapping geometries, etc.
2. Can we use Simba via jdbc? If not, are there plans for that in the future?

Thanks!
Konstantina

[dongx-psu]

Hi Konstantina,
Thanks for your attention to our project. These are the answers to your question:

1. We can support sfs in Simba since it can be reduced to a parser problem or a query rewriting problem. I can see the reason why people want this feature since sfs is a wide accepted specification for spatial operations in SQL. We will consider integrating this feature in following up versions.
2. Currently, no. In Spark 1.x, JDBC service is bounded with hive context which uses a different parser from SQLContext. However, they merge SQLContext and HiveContext together as SparkSession in Spark 2.x. I think it will be natually supported when we migrate our project to Spark 2.x.

After answering these questions above, I would like to mention that I am a first Ph.D. student by now and I am pretty heavy loaded by my course works and other research. I will try my best to make some time in adding new features and migrating Simba to Spark 2.x. But please understand that we cannot do this in a fast and efficient industry working cycle.

Thanks again!

Dong

[ricosfeifei]

Hi Konstantina,

Thanks for your interest to the project!

As Dong mentioned, we are in the process of migrating to Spark 2.0.
Hopefully we will get that done in the next few months, after which we
should be be able to support JDBC as well as additional query feature
support. Stay tuned...

Thanks!
feifei

On Tue, Nov 1, 2016 at 9:38 AM, Dong Xie [email protected] wrote:

Hi Konstantina,
Thanks for your attention to our project. These are the answers to your
question:

We can support sfs in Simba since it can be reduced to a parser
problem or a query rewriting problem. I can see the reason why people want
this feature since sfs is a wide accepted specification for spatial
operations in SQL. We will consider integrating this feature in following
up versions.
2.

Currently, no. In Spark 1.x, JDBC service is bounded with hive context
which uses a different parser from SQLContext. However, they merge
SQLContext and HiveContext together as SparkSession in Spark 2.x. I think
it will be natually supported when we migrate our project to Spark 2.x.

After answering these questions above, I would like to mention that I am a
first Ph.D. student by now and I am pretty heavy loaded by my course works
and other research. I will try my best to make some time in adding new
features and migrating Simba to Spark 2.x. But please understand that we
cannot do this in a fast and efficient industry working cycle.

Thanks again!

Dong

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
#64 (comment),
or mute the thread
https://github.com/notifications/unsubscribe-auth/AEX3pzl8eJQtb2AxfTyzRfpQZc6GnPUHks5q51z7gaJpZM4Kl7PG
.

[ConstantB]

Hi Dong,

thanks a lot for the reply! Of course I understand that it takes a good amount of effort to support these things, my intention was more to understand what Simba currently offers as I am about to use it than to request for the new features. So no worries at all, thanks again and keep up the good work!

Best,
Konstantina