[PATCH] user32/tests: Add more GetKeyboardLayout(Name) tests.

[Rémi Bernon]

At least on recent Windows, the HKL values are a combination of user
selected UI language in the low word, and keyboard layout id in the high
word.

The keyboard layout id used in HKL is sometimes different from its name,
as returned by GetKeyboardLayoutName, in which case it's the "Layout Id"
value in the corresponding registry key under the layout name.

Signed-off-by: Rémi Bernon 
---
 dlls/user32/tests/input.c | 53 ++++++++++++++++++++++++++++++++++-----
 1 file changed, 47 insertions(+), 6 deletions(-)

diff --git a/dlls/user32/tests/input.c b/dlls/user32/tests/input.c
index 2397b4104e4..4fceaa7a159 100644
--- a/dlls/user32/tests/input.c
+++ b/dlls/user32/tests/input.c
@@ -53,6 +53,7 @@
 #include "wingdi.h"
 #include "winuser.h"
 #include "winnls.h"
+#include "winreg.h"
 #include "ddk/hidsdi.h"

#include "wine/test.h"
@@ -3034,8 +3035,11 @@ static void test_get_async_key_state(void)

static void test_keyboard_layout_name(void)
 {
+    WCHAR klid[KL_NAMELENGTH], tmpklid[KL_NAMELENGTH], layout_path[MAX_PATH], value[5];
+    HKL layout, tmplayout, layouts[64];
+    DWORD status, value_size;
+    int i, len;
     BOOL ret;
-    char klid[KL_NAMELENGTH];

if (0) /* crashes on native system */
         ret = GetKeyboardLayoutNameA(NULL);
@@ -3045,12 +3049,49 @@ static void test_keyboard_layout_name(void)
     ok(!ret, "got %d\n", ret);
     ok(GetLastError() == ERROR_NOACCESS, "got %d\n", GetLastError());

-    if (GetKeyboardLayout(0) != (HKL)(ULONG_PTR)0x04090409) return;
+    layout = GetKeyboardLayout(0);
+
+    len = GetKeyboardLayoutList(ARRAY_SIZE(layouts), layouts);
+    ok(len > 0, "GetKeyboardLayoutList returned %d\n", len);
+
+    for (i = len - 1; i >= 0; --i)
+    {
+        UINT id = (UINT_PTR)layouts[i];
+        ActivateKeyboardLayout(layouts[i], 0);
+        GetKeyboardLayoutNameW(klid);
+
+        if (id & 0x80000000)
+        {
+            todo_wine ok((id >> 28) == 0xf, "hkl high bits %#x, expected 0xf\n", id >> 28);
+
+            value_size = sizeof(value);
+            wcscpy(layout_path, L"System\CurrentControlSet\Control\Keyboard Layouts\");
+            wcscat(layout_path, klid);
+            status = RegGetValueW(HKEY_LOCAL_MACHINE, layout_path, L"Layout Id", RRF_RT_REG_SZ, NULL, (void *)&value, &value_size);
+            ok(!status, "RegGetValueW returned %x\n", status);
+            todo_wine ok(value_size == 5 * sizeof(WCHAR), "RegGetValueW returned size %d\n", value_size);
+
+            swprintf(tmpklid, KL_NAMELENGTH, L"%04X", (id >> 16) & 0x0fff);
+            todo_wine ok(!wcsicmp(value, tmpklid), "RegGetValueW returned %s, expected %s\n", debugstr_w(value), debugstr_w(tmpklid));
+        }
+        else
+        {
+            swprintf(tmpklid, KL_NAMELENGTH, L"%08X", id >> 16);
+            ok(!wcsicmp(klid, tmpklid), "GetKeyboardLayoutNameW returned %s, expected %s\n", debugstr_w(klid), debugstr_w(tmpklid));
+        }
+
+        ActivateKeyboardLayout(layout, 0);
+        tmplayout = LoadKeyboardLayoutW(klid, KLF_ACTIVATE);
+
+        /* The low word of HKL is the selected user lang and may be different as LoadKeyboardLayoutW also selects the default lang from the layout */
+        ok(((UINT_PTR)tmplayout & ~0xffff) == ((UINT_PTR)layouts[i] & ~0xffff), "GetKeyboardLayout returned %p, expected %p\n", tmplayout, layouts[i]);
+
+        /* The layout name only depends on the keyboard layout: the high word of HKL. */
+        GetKeyboardLayoutNameW(tmpklid);
+        ok(!wcsicmp(klid, tmpklid), "GetKeyboardLayoutNameW returned %s, expected %s\n", debugstr_w(tmpklid), debugstr_w(klid));
+    }

-    klid[0] = 0;
-    ret = GetKeyboardLayoutNameA(klid);
-    ok(ret, "GetKeyboardLayoutNameA failed %u\n", GetLastError());
-    ok(!strcmp(klid, "00000409"), "expected 00000409, got %s\n", klid);
+    ActivateKeyboardLayout(layout, 0);
 }

static void test_key_names(void)

[Marvin]

Hi,

While running your changed tests, I think I found new failures.
Being a bot and all I'm not very good at pattern recognition, so I might be
wrong, but could you please double-check?

Full results can be found at:
https://testbot.winehq.org/JobDetails.pl?Key=89412

Your paranoid android.

=== w7u_2qxl (32 bit report) ===

user32:
input.c:757: Test failed:  0 (a4/0): 00 from 00 -> 80 unexpected
input.c:757: Test failed:  0 (a4/0): 41 from 01 -> 00 unexpected

=== w1064v1809 (32 bit report) ===

user32:
input.c:2748: Test failed: 0: expected WM_MOUSEMOVE message
input.c:2748: Test failed: 1: expected WM_MOUSEMOVE message
input.c:2748: Test failed: 2: expected WM_MOUSEMOVE message

=== w10pro64 (32 bit report) ===

user32:
input.c:3357: Test failed: expected WM_NCHITTEST message
input.c:3358: Test failed: expected WM_LBUTTONDOWN message
input.c:3359: Test failed: expected WM_LBUTTONUP message
input.c:3412: Test failed: expected loop with WM_NCHITTEST messages
input.c:3465: Test failed: expected WM_LBUTTONDOWN message
input.c:3466: Test failed: expected WM_LBUTTONUP message

=== wvistau64 (64 bit report) ===

user32:
input.c:757: Test failed:  0 (a4/0): 01 from 01 -> 00 unexpected
input.c:757: Test failed:  0 (a4/0): 11 from 01 -> 00 unexpected
input.c:757: Test failed:  0 (a4/0): a2 from 01 -> 00 unexpected

=== debiant2 (32 bit Japanese:Japan report) ===

user32:
input.c:3071: Test failed: RegGetValueW returned 2
input.c:3072: Test succeeded inside todo block: RegGetValueW returned size 10

=== debiant2 (32 bit Chinese:China report) ===

user32:
input.c:3071: Test failed: RegGetValueW returned 2
input.c:3072: Test succeeded inside todo block: RegGetValueW returned size 10