Mailing-List: contact dev-help@brooklyn.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@brooklyn.apache.org
From: ahgittin <git@git.apache.org>
To: dev@brooklyn.apache.org
Reply-To: dev@brooklyn.apache.org
References: <git-pr-816-brooklyn-server@git.apache.org>
In-Reply-To: <git-pr-816-brooklyn-server@git.apache.org>
Subject: [GitHub] brooklyn-server pull request #816: Tasks code improvements - prep for better...
Content-Type: text/plain
Message-Id: <20170928142303.DC4EFF3261@git1-us-west.apache.org>
Date: Thu, 28 Sep 2017 14:23:03 +0000 (UTC)
archived-at: Thu, 28 Sep 2017 14:23:06 -0000

Github user ahgittin commented on a diff in the pull request:

    https://github.com/apache/brooklyn-server/pull/816#discussion_r141632615
  
    --- Diff: core/src/test/java/org/apache/brooklyn/util/core/task/BasicTasksFutureTest.java ---
    @@ -174,29 +175,53 @@ public T call() {
         }
     
         @Test
    -    public void testCancelAfterStartTriggersListenableFuture() throws Exception {
    -        doTestCancelTriggersListenableFuture(Duration.millis(50));
    +    public void testCancelAfterStartTriggersListenableFutureDynamic() throws Exception {
    +        doTestCancelTriggersListenableFuture(Duration.millis(50), true);
         }
         @Test
    -    public void testCancelImmediateTriggersListenableFuture() throws Exception {
    +    public void testCancelImmediateTriggersListenableFutureDynamic() throws Exception {
             // if cancel fires after submit but before it passes to the executor,
    -        // that needs handling separately; this doesn't guarantee this code path,
    -        // but it happens sometimes (and it should be handled)
    -        doTestCancelTriggersListenableFuture(Duration.ZERO);
    +        // that needs handling separately as it falls into an edge where the future is set and cancelled
    +        // so the executor won't run it, and our wrapper logic doesn't apply; 
    +        // this test doesn't guarantee this code path, but makes it likely enough it happens once in a while.
    +        doTestCancelTriggersListenableFuture(Duration.ZERO, true);
         }
    -    public void doTestCancelTriggersListenableFuture(Duration delay) throws Exception {
    -        Task<String> t = waitForSemaphore(Duration.TEN_SECONDS, true, "x");
    +    @Test
    +    public void testCancelBeforeTriggersListenableFutureDynamic() throws Exception {
    +        doTestCancelTriggersListenableFuture(Duration.millis(-50), true);
    +    }
    +    @Test
    +    public void testCancelAfterStartTriggersListenableFutureSimple() throws Exception {
    +        doTestCancelTriggersListenableFuture(Duration.millis(50), true);
    +    }
    +    @Test
    +    public void testCancelImmediateTriggersListenableFutureSimple() throws Exception {
    +        doTestCancelTriggersListenableFuture(Duration.ZERO, false);
    +    }
    +    @Test
    +    public void testCancelBeforeTriggersListenableFutureSimple() throws Exception {
    +        doTestCancelTriggersListenableFuture(Duration.millis(-50), false);
    +    }
    +    public void doTestCancelTriggersListenableFuture(Duration delay, boolean dynamic) throws Exception {
    --- End diff --
    
    good idea


---