Rémi Bernon (@rbernon) commented about dlls/bluetoothapis/tests/sdp.c:
+ ok( ret == error, "Expected BluetoothSdpGetElementData to return %ld, got %ld.\n", error, ret ); + if (error == ERROR_SUCCESS) + { + ok( result.type == sdp_data->type, "Expected SDP_TYPE %s, got %s.\n", + debugstr_SDP_TYPE( sdp_data->type ), debugstr_SDP_TYPE( result.type ) ); + ok( result.specificType == sdp_data->specificType, + "Expected SDP_SPECIFIC_TYPE %s, got %s.\n", + debugstr_SDP_SPECIFIC_TYPE( sdp_data->specificType ), + debugstr_SDP_SPECIFIC_TYPE( result.specificType ) ); + ok( !memcmp( &sdp_data->data, &result.data, sizeof( result.data ) ), "Expected %s, got %s.\n", + debugstr_SDP_ELEMENT_DATA( sdp_data ), debugstr_SDP_ELEMENT_DATA( &result ) ); + + } +} + +#define TEST_CASE_NAME( n ) ("%s %d"), __func__, (int)(n) Similarly, it is often a good idea to keep the test context as short as possible. It's being added to every todo line, and quickly adds up to the output size failure threshold.
-- https://gitlab.winehq.org/wine/wine/-/merge_requests/6402#note_82790