Test case in GrantRevokeDDLTest looks to be accidentally commented out...
-------------------------------------------------------------------------

                 Key: DERBY-3376
                 URL: https://issues.apache.org/jira/browse/DERBY-3376
             Project: Derby
          Issue Type: Test
            Reporter: A B
            Priority: Minor
             Fix For: 10.4.0.0


In lang/GrantRevokeDDLTest.java, beginning at line 8057 (at time of writing), 
there looks to be a test case that was inadvertently commented out:

        // set connection user1
        //ij(USER3)> -- test multiple FKs DERBY-1589?set
        // connection user1drop table user3.rt3drop table
        // user2.rt2drop table user1.rt1create table rt1 (c1 int
        // primary key not null, c2 int)insert into rt1 values
        // (1,1), (2,2)grant references on rt1 to PUBLIC, user2,
        // user3set connection user2 XJ001 occurred at create table
        // rt2...create table rt2 (c1 int primary key not null,
        // constraint rt2fk foreign key(c1) references
        // user1.rt1)insert into rt2 values (1), (2)grant
        // references on rt2 to PUBLIC, user3set connection
        // user3create table rt3 (c1 int primary key not null,
        // constraint rt3fk1 foreign key(c1) references
        // user1.rt1,   constraint rt3fk2 foreign key(c1) references
        // user1.rt2)insert into rt3 values (1), (2)set connection
        // user1 rt3fk1 should get dropped.revoke references on rt1
        // from PUBLICrevoke references on rt1 from user3set
        // connection user2revoke references on rt2 from PUBLIC
        // expect errorinsert into rt2 values (3)set connection
        // user3 expect error, use user3 references privilege,
        // rt3fk2 still in effectinsert into rt3 values (3)set
        // connection user2revoke references on rt2 from user3set
        // connection user3 ok, rt3fk2 should be dropped.insert
        // into rt3 values (3)

Based on the format of the comment, my guess is that this JUnit test was 
created using the conversion script attached to DERBY-2151, and that the script 
somehow messed up the conversion.

The first line of the comment mentions DERBY-1589, but a) there is a question 
mark after it, and b) the commit for that issue does not appear to include the 
aforementioned lines.

If the test case is _supposed_ to be commented out, then some explanatory 
comments explaining why might be nice...

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply via email to