[PATCH v4 0/1] MR4665: winegstreamer: In error case don't dereference null pointer (Coverity)

[Fabian Maurer (＠DarkShadow44)]

--
  v4: winegstreamer: Don't check event for NULL, gstreamer already does that

https://gitlab.winehq.org/wine/wine/-/merge_requests/4665

[PATCH v4 1/1] winegstreamer: Don't check event for NULL, gstreamer already does that

[Fabian Maurer]

From: Fabian Maurer

---
 dlls/winegstreamer/wg_parser.c    | 4 +---
 dlls/winegstreamer/wg_transform.c | 4 +---
 2 files changed, 2 insertions(+), 6 deletions(-)

diff --git a/dlls/winegstreamer/wg_parser.c b/dlls/winegstreamer/wg_parser.c
index 7193062e89b..b0a344d5186 100644
--- a/dlls/winegstreamer/wg_parser.c
+++ b/dlls/winegstreamer/wg_parser.c
@@ -484,9 +484,7 @@ static NTSTATUS wg_parser_stream_notify_qos(void *args)
         GST_LOG("Ignoring QoS event.");
         return S_OK;
     }
-    if (!(event = gst_event_new_qos(params->underflow ? GST_QOS_TYPE_UNDERFLOW : GST_QOS_TYPE_OVERFLOW,
-            params->proportion, diff, stream_time)))
-        GST_ERROR("Failed to create QOS event.");
+    event = gst_event_new_qos(params->underflow ? GST_QOS_TYPE_UNDERFLOW : GST_QOS_TYPE_OVERFLOW, params->proportion, diff, stream_time);
     push_event(stream->my_sink, event);

return S_OK;
diff --git a/dlls/winegstreamer/wg_transform.c b/dlls/winegstreamer/wg_transform.c
index ab5ce381fc5..db2a5037476 100644
--- a/dlls/winegstreamer/wg_transform.c
+++ b/dlls/winegstreamer/wg_transform.c
@@ -953,9 +953,7 @@ NTSTATUS wg_transform_notify_qos(void *args)
         GST_LOG("Ignoring QoS event.");
         return S_OK;
     }
-    if (!(event = gst_event_new_qos(params->underflow ? GST_QOS_TYPE_UNDERFLOW : GST_QOS_TYPE_OVERFLOW,
-            params->proportion, diff, stream_time)))
-        GST_ERROR("Failed to create QOS event.");
+    event = gst_event_new_qos(params->underflow ? GST_QOS_TYPE_UNDERFLOW : GST_QOS_TYPE_OVERFLOW, params->proportion, diff, stream_time);
     push_event(transform->my_sink, event);

return S_OK;

[Fabian Maurer (＠DarkShadow44)]

On Sat Jan 20 21:57:43 2024 +0000, Rémi Bernon wrote:
> Fwiw gst_pad_push_event already does NULL check and prints a Glib
> critical error message if that's the case. That's semantically
> equivalent to an assert according to the Glib documentation, so IMO we
> could just get rid of the checks on our side entirely, and do `push_event(gst_event_new...)`.

Alright, updated