Uhm, this patch introduces test failures. I had assumed the Marvin email is enough to reject this with "Test failures".
I doubt this patch is the real culprit but I didn't have time to properly debug the issue.
On 10/22/19 11:57 PM, Alexandre Julliard wrote:
Module: wine Branch: master Commit: f8979fee1fa2e6aeed5a5026465dffa680c0848f URL: https://source.winehq.org/git/wine.git/?a=commit;h=f8979fee1fa2e6aeed5a50264...
Author: Michael Stefaniuc mstefani@winehq.org Date: Tue Oct 22 01:24:32 2019 +0200
comctl32/tests: Get rid of variables holding empty strings.
Signed-off-by: Michael Stefaniuc mstefani@winehq.org Signed-off-by: Alexandre Julliard julliard@winehq.org
diff --git a/dlls/comctl32/tests/pager.c b/dlls/comctl32/tests/pager.c index 0e2c25b08a..30005c88fa 100644 --- a/dlls/comctl32/tests/pager.c +++ b/dlls/comctl32/tests/pager.c @@ -132,17 +130,17 @@ static const struct notify_test_send test_dont_convert_send_data[] =
static const struct notify_test_receive test_convert_receive_data[] = {
- {empty_w, sizeof(empty_w), ARRAY_SIZE(buffer), NULL, test_a, sizeof(test_a), -1, test_w, ARRAY_SIZE(buffer)},
- {empty_w, sizeof(empty_w), ARRAY_SIZE(buffer), test_a, NULL, 0, -1, test_w, ARRAY_SIZE(buffer)},
- {NULL, sizeof(empty_w), ARRAY_SIZE(buffer), test_a, NULL, 0, -1, NULL, ARRAY_SIZE(buffer)},
- {empty_w, sizeof(empty_w), ARRAY_SIZE(buffer), large_a, NULL, 0, -1, large_truncated_65_w, ARRAY_SIZE(buffer)},
- {empty_w, sizeof(empty_w), ARRAY_SIZE(buffer), empty_a, 0, 0, 1, empty_w, 1},
- {L"", sizeof(L""), ARRAY_SIZE(buffer), NULL, test_a, sizeof(test_a), -1, test_w, ARRAY_SIZE(buffer)},
- {L"", sizeof(L""), ARRAY_SIZE(buffer), test_a, NULL, 0, -1, test_w, ARRAY_SIZE(buffer)},
- {NULL, sizeof(L""), ARRAY_SIZE(buffer), test_a, NULL, 0, -1, NULL, ARRAY_SIZE(buffer)},
- {L"", sizeof(L""), ARRAY_SIZE(buffer), large_a, NULL, 0, -1, large_truncated_65_w, ARRAY_SIZE(buffer)}, ^^^ changing that back to empty_w makes the test failures go away...
bye michael
On 10/23/19 12:43 AM, Michael Stefaniuc wrote:
Uhm, this patch introduces test failures. I had assumed the Marvin email is enough to reject this with "Test failures".
I doubt this patch is the real culprit but I didn't have time to properly debug the issue.
diff --git a/dlls/comctl32/tests/pager.c b/dlls/comctl32/tests/pager.c index 0e2c25b08a..30005c88fa 100644 --- a/dlls/comctl32/tests/pager.c +++ b/dlls/comctl32/tests/pager.c @@ -132,17 +130,17 @@ static const struct notify_test_send test_dont_convert_send_data[] =
static const struct notify_test_receive test_convert_receive_data[] = {
- {empty_w, sizeof(empty_w), ARRAY_SIZE(buffer), NULL, test_a, sizeof(test_a), -1, test_w, ARRAY_SIZE(buffer)},
- {empty_w, sizeof(empty_w), ARRAY_SIZE(buffer), test_a, NULL, 0, -1, test_w, ARRAY_SIZE(buffer)},
- {NULL, sizeof(empty_w), ARRAY_SIZE(buffer), test_a, NULL, 0, -1, NULL, ARRAY_SIZE(buffer)},
- {empty_w, sizeof(empty_w), ARRAY_SIZE(buffer), large_a, NULL, 0, -1, large_truncated_65_w, ARRAY_SIZE(buffer)},
- {empty_w, sizeof(empty_w), ARRAY_SIZE(buffer), empty_a, 0, 0, 1, empty_w, 1},
- {L"", sizeof(L""), ARRAY_SIZE(buffer), NULL, test_a, sizeof(test_a), -1, test_w, ARRAY_SIZE(buffer)},
- {L"", sizeof(L""), ARRAY_SIZE(buffer), test_a, NULL, 0, -1, test_w, ARRAY_SIZE(buffer)},
- {NULL, sizeof(L""), ARRAY_SIZE(buffer), test_a, NULL, 0, -1, NULL, ARRAY_SIZE(buffer)},
- {L"", sizeof(L""), ARRAY_SIZE(buffer), large_a, NULL, 0, -1, large_truncated_65_w, ARRAY_SIZE(buffer)},
^^^ changing that back to empty_w makes the test failures go away...
$^@$#(@!^ text-flawed in thunderbird.
bye michael
Michael Stefaniuc mstefani@winehq.org writes:
Uhm, this patch introduces test failures. I had assumed the Marvin email is enough to reject this with "Test failures".
Sadly, there are so many spurious failure reports from Marvin these days that I'm basically ignoring them as false positives. Yes, that's a bad thing...
-
Alexandre Julliard -
Michael Stefaniuc