Minor variation in write_test.out which appears to be PG running the ROLLBACK handler after the error is reported. I can't see any reason that's a major concern, but I noted it in the changelog.
diff -U3 /build/test-3.9/expected/write_test.out /build/results/write_test.out
--- /build/test-3.9/expected/write_test.out 2024-05-14 13:15:17.955798144 +0000
+++ /build/results/write_test.out 2024-05-14 13:15:20.226788667 +0000
@@ -15,23 +15,23 @@
NOTICE: [('option1', 'option1'), ('test_type', 'nowrite'), ('tx_hook', 'true'), ('usermapping', 'test')]
NOTICE: [('test1', 'character varying'), ('test2', 'character varying')]
NOTICE: BEGIN
-NOTICE: ROLLBACK
ERROR: Error in python: NotImplementedError
DETAIL: This FDW does not support the writable API
+NOTICE: ROLLBACK
update testmulticorn set test1 = 'test';
NOTICE: BEGIN
NOTICE: []
NOTICE: ['test1', 'test2']
-NOTICE: ROLLBACK
ERROR: Error in python: NotImplementedError
DETAIL: This FDW does not support the writable API
+NOTICE: ROLLBACK
delete from testmulticorn where test2 = 'test2 2 0';
NOTICE: BEGIN
NOTICE: [test2 = test2 2 0]
NOTICE: ['test1', 'test2']
-NOTICE: ROLLBACK
ERROR: Error in python: NotImplementedError
DETAIL: This FDW does not support the writable API
+NOTICE: ROLLBACK
CREATE foreign table testmulticorn_write (
test1 character varying,
test2 character varying
Minor variation in write_test.out which appears to be PG running the ROLLBACK handler after the error is reported. I can't see any reason that's a major concern, but I noted it in the changelog.