[PATCH] user32/tests: Avoid a couple of unneeded strlen() calls.

[Francois Gouget]

Note that DeviceString is an array field and thus cannot be NULL.

Signed-off-by: Francois Gouget 
---
 dlls/user32/tests/monitor.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/dlls/user32/tests/monitor.c b/dlls/user32/tests/monitor.c
index 567870e3d1f..47e01b1a6a6 100644
--- a/dlls/user32/tests/monitor.c
+++ b/dlls/user32/tests/monitor.c
@@ -71,7 +71,6 @@ static void test_enumdisplaydevices_adapter(int index, const DISPLAY_DEVICEA *de
     int device_id;
     int subsys_id;
     int revision_id;
-    size_t length;
     HDC hdc;

adapter_count++;
@@ -87,9 +86,8 @@ static void test_enumdisplaydevices_adapter(int index, const DISPLAY_DEVICEA *de
        "#%d: wrong DeviceKey %s\n", index, device->DeviceKey);

/* DeviceString */
-    length = strlen(device->DeviceString);
-    ok(broken(length == 0) || /* XP on Testbot will return an empty string, whereas XP on real machine doesn't. Probably a bug in virtual adapter driver */
-       length > 0, "#%d: expect DeviceString not empty\n", index);
+    ok(broken(!*device->DeviceString) || /* XP on Testbot will return an empty string, whereas XP on real machine doesn't. Probably a bug in virtual adapter driver */
+       *device->DeviceString, "#%d: expect DeviceString not empty\n", index);

/* StateFlags */
     if (index == 0)
@@ -139,7 +137,7 @@ static void test_enumdisplaydevices_monitor(int adapter_index, int monitor_index
     ok(!strcmp(monitor_name, device->DeviceName), "#%d: expect %s, got %s\n", monitor_index, monitor_name, device->DeviceName);

/* DeviceString */
-    ok(strlen(device->DeviceString) > 0, "#%d: expect DeviceString not empty\n", monitor_index);
+    ok(*device->DeviceString, "#%d: expect DeviceString not empty\n", monitor_index);

/* StateFlags */
     if (adapter_index == 0 && monitor_index == 0)