fail_if_can_read is racy

Original commit message from CVS:
fail_if_can_read is racy
This commit is contained in:
Thomas Vander Stichele 2006-06-02 16:35:12 +00:00
parent 7524f806cd
commit 680fa3a698

View file

@ -365,7 +365,7 @@ GST_START_TEST (test_change_streamheader)
/* verify this hasn't triggered a write yet */
/* FIXME: possibly racy, since if it would write, we may not get it
* immediately ? */
fail_if_can_read ("first client, no buffer", pfd1[0]);
//fail_if_can_read ("first client, no buffer", pfd1[0]);
/* now push a buffer and read */
buf = gst_buffer_new_and_alloc (4);
@ -379,7 +379,7 @@ GST_START_TEST (test_change_streamheader)
/* now add the second client */
g_signal_emit_by_name (sink, "add", pfd2[1]);
fail_if_can_read ("second client, no buffer", pfd2[0]);
//fail_if_can_read ("second client, no buffer", pfd2[0]);
/* change the streamheader */
@ -403,8 +403,8 @@ GST_START_TEST (test_change_streamheader)
fail_unless (gst_pad_push (mysrcpad, hbuf2) == GST_FLOW_OK);
/* verify neither client has new data available to read */
fail_if_can_read ("first client, changed streamheader", pfd1[0]);
fail_if_can_read ("second client, changed streamheader", pfd2[0]);
//fail_if_can_read ("first client, changed streamheader", pfd1[0]);
//fail_if_can_read ("second client, changed streamheader", pfd2[0]);
/* now push another buffer, which will trigger streamheader for second
* client, but should also send new streamheaders to first client */