zhang-arvin commented on PR #7001: URL: https://github.com/apache/shenyu/pull/7001#issuecomment-6051673663
Thanks @Aias00 @sunnysabor @dengliming — the compile break and the regression-test gap are both addressed on the current head (`6440a7fa9`). **1. Mapper signature propagated everywhere (compile break)** Rebased onto current `master` and fixed every call site so the two-argument `SelectorMapper#selectByIdSet(Set, String)` / `#deleteByIds(List, String)` signature compiles across production and test code. The rebase had re-introduced new master-side single-argument call sites in `SelectorServiceTest`; those (and the `deleteByIds(any())` verify) now pass `SYS_DEFAULT_NAMESPACE_ID`. Verified locally: `mvn -pl shenyu-admin -am -DskipTests test-compile` → BUILD SUCCESS. **2. Namespace-boundary regression test (sunnysabor's ask)** Added `SelectorMapperTest#testSelectByIdSetAndDeleteByIdsAreNamespaceScoped` — exactly the probe you described: it inserts a selector in a second namespace, asserts `selectByIdSet(ids, SYS_DEFAULT_NAMESPACE_ID)` returns only the default-namespace row (not the other namespace's), then asserts `deleteByIds(ids, SYS_DEFAULT_NAMESPACE_ID)` deletes 1 row, leaves the other namespace's selector present, and then cleans it up. Verified it is not vacuous: removing the `namespace_id` predicate from either `selectByIdSet` or `deleteByIds` in `selector-sqlmap.xml` makes the test fail (`selectByIdSet` then returns 2 rows instead of 1). Local run: `SelectorMapperTest` + `SelectorServiceTest` → 30 tests, 0 failures, BUILD SUCCESS. Would you mind taking another look? -- 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]
