Robin Moffatt

@MichaelDrogalis any thoughts on this? thanks!

Here's another workaround, this one will execute all the SQL files in a mounted folder: ``` mssql: image: mcr.microsoft.com/mssql/server:2017-latest environment: - SA_PASSWORD=Admin123 - ACCEPT_EULA=Y volumes: - ./data/mssql:/scripts/ command: - /bin/bash...

thanks @twright-msft. I enhanced it a bit further with trying to detect if the engine has started or not: ``` command: - /bin/bash - -c - | # Launch MSSQL...

yes that would be 💯 what I need 👍

One workaround for now is to use `-J` in combination with `jq`: ``` kafkacat -b kafka:29092 -L -J|jq 'del(.topics)' { "originating_broker": { "id": 1, "name": "kafka:29092/1" }, "query": { "topic":...

TBH I honestly know very little about the current tooling and techniques, so I am very happy to be guided by people's advice here 👍 I also wonder if this...

When I created a JDBC connector the StringConverter was set automagically, but for the Debezium connector it's not doing. So, if it doesn't set it automagically, then it should :)

Hmm this is maybe not as straightforward as I thought. For Debezium with StringConverter the key ends up as `Struct{id=43}` ``` 10/11/19 11:58:50 AM UTC, Struct{id=43}, {"id": 43, "first_name": "Rick",...

So Debezium, unlike the JDBC source connector, sets the message key itself based on the PK of the source table. But letting it do that with `'key.converter'= 'org.apache.kafka.connect.storage.StringConverter',` results in...

does `numeric.mapping` only apply to `NUMERIC` types, not `DECIMAL`? https://github.com/confluentinc/kafka-connect-jdbc/blob/b23e31295bb91bad3720b378792aee345a09c876/src/main/java/io/confluent/connect/jdbc/dialect/GenericDatabaseDialect.java#L1146-L1189