[FLINK-33761][Connector/JDBC] Add Snowflake JDBC Dialect

[borislitvak]

Feature: New JBDC connector dialect - connect to Snowflake.

Testing: As Snowflake is a SAAS offering, I did not provide any tests for it. We don't want to incur any more expenses. Note that it's a new feature. Any potential breakages in this PR should not impact existing users.

Usage Example: This is how I use it in my sample code:

    public static Table getSnowflakeTable(StreamTableEnvironment tEnv, SnowflakeConfig config) {
        tEnv.createTemporaryTable(
                "XXX",
                TableDescriptor.forConnector("jdbc")
                        .schema(Schema.newBuilder()
                                .column("F1", DataTypes.BIGINT())
                                .column("F2", DataTypes.TIMESTAMP().bridgedTo(java.time.LocalDateTime.class))
                                .build()
                        )
                        .option("driver", config.getDriver())
                        // https://docs.snowflake.com/developer-guide/jdbc/jdbc-configure#connection-parameters
                        .option("url", config.getUrl() + "&CLIENT_TIMESTAMP_TYPE_MAPPING=TIMESTAMP_NTZ")
                        .option("table-name", "Snowflake-Sample-Table-Replace-This")
                        .build()
        );
        return tEnv.sqlQuery("SELECT * FROM XXX");
    }

[boring-cyborg]

Thanks for opening this pull request! Please check out our contributing guidelines. (https://flink.apache.org/contributing/how-to-contribute.html)

[borislitvak]

@davidradl WDYT?

[EcaterinaL]

Hello all,

Is there any progress on this? will it be merged, if yes, when do you think this would be?

Thank you!

[borislitvak]

Hello all,

Is there any progress on this? will it be merged, if yes, when do you think this would be?

Thank you!

Ecaterina, I am still waiting for a months for a year. Using this in prod with 1.19.1. @EcaterinaL @davidradl

[EcaterinaL]

Thank you for your prompt answer!,
Do you actively use it?
Have you merge it with your master?

[borislitvak]

I actively use it, yes. You can built from branch.

…

On Wed, Nov 20, 2024, 17:24 Ecaterina Lazarescu ***@***.***> wrote: Thank you for your prompt answer!, Do you actively use it? Have you merge it with your master? — Reply to this email directly, view it on GitHub <#118 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACKDRNMTPNR7WTWC4LVYVT32BSSRRAVCNFSM6AAAAABG77RBHWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIOBYHA3TINRZGE> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[davidradl]

Hi @borislitvak @EcaterinaL , thanks for the changes - I cannot merge this as I am not a committer.
I notice the branch has conflicts.
I have suggested this be included in the forthcoming JDBC connector for Flink 1.20.

@snuyanzin is this something you can help with reviewing/merging once the conflicts are resolved please?

[EcaterinaL]

@davidradl, I used this repo to try and create a connector myself, however I found that some of the methods(ex createInputSerializer()) are not available on flink 1.19 I have to refactor. Is it possible to resolve some of them in 1.20? And when do you think the new version will be available ? :)

Thank you!

[davidradl]

@davidradl, I used this repo to try and create a connector myself, however I found that some of the methods(ex createInputSerializer()) are not available on flink 1.19 I have to refactor. Is it possible to resolve some of them in 1.20? And when do you think the new version will be available ? :)

Thank you!

Hi @EcaterinaL , I am not sure what you mean by "Is it possible to resolve some of them in 1.20?" I suggest checking the git history to see when the methods were removed and how callers now get this capability. At the moment it is early days in getting a JDBC connector out that works with 1.20; I initiated a conversation on the dev list to get this started. The release manager has not been agreed yet, so I do not have a timescale for you.

[EcaterinaL]

Thank you @davidradl , I'll look at the history, good idea. I'm going to keep monitor this thread, for me is important to have this connector ready in coming months.

[davidradl]

Thank you @davidradl , I'll look at the history, good idea. I'm going to keep monitor this thread, for me is important to have this connector ready in coming months.

no worries - I also suggest that you respond with your need to my dev list post - I proposed that we can get a 1.20 release of the Flink JDBC connector cut off main - then if your PR is merged this dialect will be included.