Conversation
|
We should also consider ScanSpec for building function privileges of perm function usages in queries, as in pending PR 3633 (#3633). The core logic to this is scanning logical plan by |
How can ScanSpec configure functions? |
If |
...uubi-spark-authz/src/main/scala/org/apache/kyuubi/plugin/spark/authz/serde/CommandSpec.scala
Outdated
Show resolved
Hide resolved
...c/main/resources/META-INF/services/org.apache.kyuubi.plugin.spark.authz.serde.TableExtractor
Show resolved
Hide resolved
|
LGTM. Just some minor comments. Let's merge it after passing the CI. Glad to see this PR reaching a milestone for Serde refactoring in Authz module, by completely removing hardcoded rules and extraction of Big applause to @yaooqinn for your elegant redesign and all the hard work. 👏🏻 🍻 |
|
thanks @bowenliang123 for the check, merged to master |
Why are the changes needed?
Add serde layer for queries, which extracts table information from a scan relation, such as HiveTableRelation, logical relation, etc.
How was this patch tested?
Add some test cases that check the changes thoroughly including negative and positive cases if possible
Add screenshots for manual tests if appropriate
Run test locally before make a pull request