olavloite opened a new issue, #29985:
URL: https://github.com/apache/beam/issues/29985
### What happened?
Note: Nothing has broken (yet), but there is a potential bug in the Spanner
IO connector.
The Cloud Spanner IO connector reads the tables that are currently in the
database. As Cloud Spanner supports two different dialects (GoogleSQL and
PostgreSQL), there are also different system queries for this. These behave
slightly differently:
1. GoogleSQL: Selects all tables that are in the unnamed schema. All other
tables are ignored.
2. PostgreSQL: Selects all tables that are __not__ in a fixed list of three
system schemas: ` ('information_schema', 'spanner_sys', 'pg_catalog')`
The latter can cause unwanted behavior if any of the following happens:
1. Cloud Spanner introduces a new system schema.
2. Cloud Spanner introduces support for named user schemas.
The above could break if a new schema contains a table with the same name as
a table in the existing default schema.
The second filter should therefore be modified to do the same as the
GoogleSQL version: It should only select the tables in the default user schema.
For PostgreSQL, that is `public`.
### Issue Priority
Priority: 3 (minor)
### Issue Components
- [ ] Component: Python SDK
- [ ] Component: Java SDK
- [ ] Component: Go SDK
- [ ] Component: Typescript SDK
- [X] Component: IO connector
- [ ] Component: Beam YAML
- [ ] Component: Beam examples
- [ ] Component: Beam playground
- [ ] Component: Beam katas
- [ ] Component: Website
- [ ] Component: Spark Runner
- [ ] Component: Flink Runner
- [ ] Component: Samza Runner
- [ ] Component: Twister2 Runner
- [ ] Component: Hazelcast Jet Runner
- [ ] Component: Google Cloud Dataflow Runner
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
