[PATCH] wmvcore/tests: add tests for WMReader methods

[Jefferson Carpenter]

Hello,

On Tue, Jul 25, 2017 at 5:48 PM, Alistair Leslie-Hughes <
leslie_alistair at hotmail.com> wrote:

> Hi Jefferson,
> >
> > +typedef struct {
> > +    IWMReaderCallback IWMReaderCallback_iface;
> > +    LONG ref;
> > +    HRESULT (*onStatusCallback)(WMT_STATUS, HRESULT,
> WMT_ATTR_DATATYPE, BYTE*, void*, void*);
> > +    HRESULT (*onSampleCallback)(DWORD, QWORD, QWORD, DWORD,
> INSSBuffer*, void*, void*);
> > +    void *pvCallbackContext;
> > +} TestWMReaderCallback;
>
> I'm still a little confused as to why you need this structure when the
> IWMReaderCallbackVtbl with a few globals would work just as well.
>

Well, I it could be useful to only implement QueryInterface, AddRef, and
Release once, and use dependency injection to supply the other methods to
the vtable.  I wasn't sure whether I was over-engineering that, however.


>
> > +static HRESULT onStatus_test_wmreader_play_wma(
> > +    WMT_STATUS status,
> > +    HRESULT hr,
> > +    WMT_ATTR_DATATYPE dwType,
> > +    BYTE *pValue,
> > +    void *pvContext,
> > +    void *pvCallbackContext)
> > +{
> > +    struct callback_context_test_wmreader_play_wma *pTestContext =
> > +        (struct callback_context_test_wmreader_play_wma*)
> pvCallbackContext;
> > +    DWORD *dwordValue;
> > +
> > +    CHECK_EXPECT(onStatus);
> > +    ok(status == onStatus_expected_statuses_test_wmreader_play_wma[
> pTestContext->onStatus_call_count],
> > +       "onStatus called with unexpected status: %d\n", status);
> > +    if (status == WMT_OPENED) {
> > +        ok(pvContext == pTestContext->pvContextOpen,
> > +           "onStatus was not passed pvContext: expected %p, actual
> %p\n", pTestContext->pvContextOpen, pvContext);
> > +    }
> > +    else /* status == WMT_STARTED, or later */ {
> > +        ok(pvContext == pTestContext->pvContextStart,
> > +           "onStatus was not passed pvContext: expected %p, actual
> %p\n", pTestContext->pvContextStart, pvContext);
> > +    }
> > +    switch (status) {
> > +    case WMT_STARTED:
> > +        // MSDN says dwType is a QWORD, but it is actually a DWORD
> > +        ok(dwType == WMT_TYPE_DWORD, "onStatus pValue not
> WMT_TYPE_QWORD (%d), but %d\n", (int)WMT_TYPE_DWORD, dwType);
> WMT_TYPE_QWORD typo?
>
> > +        dwordValue = (DWORD*)pValue;
> > +        ok(*dwordValue == 0, "onStatus call for WMT_STARTED not passed
> correct starting timestamp: %u\n", *dwordValue);
> > +        SET_EXPECT(onSample);
> > +        break;
> > +    default:
> > +        break;
> > +    }
> The switch and if could be merged
>

Very well.


>
> > +    /* Get format count. */
> > +    formatCount = 0;
> > +    hr = IWMReader_GetOutputFormatCount(reader, 0, &formatCount);
> > +    todo_wine ok(hr == S_OK, "Failed to get output format count:
> 0x%08x\n", hr);
> formatCount should be checked
>
>
Okay - I'll have the test check that formatCount is equal to 37.  That's
the format count that I get on my windows machine.  The number 37 seems
oddly specific; I'm not sure how much wiggle room wmvcore implementations
have to provide varying sets of formats.  However it would be best if
wine's were identical to the Windows behavior, obviously.


> Best Regards
>  Alistair Leslie-Hughes
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.winehq.org/pipermail/wine-devel/attachments/20170725/6058deda/attachment-0001.html>