Added another tests to check some worries in Bug #582564.

Out-of-band events might lead to us calling g_object_notify() from a
non-streaming thread, which can cause crashes if g_object_notify() is
being called from the streaming thread at the same time. See #554460.

GObject may crash if two threads do concurrent g_object_notify() on the same
object. This may happen if fakesink receives an out-of-band event such as
FLUSH_START while processing a buffer or serialised event in the streaming
thread. Since this may happen with the default settings during a common
operation like a seek, and there seems to be little chance of a timely fix
in GObject (see #166020), we should hack around this issue by protecting all
of fakesink's direct g_object_notify() calls with a lock.

Also add unit test for the above.

Fixes #554460.

Make gst_tag_list_get_string() return FALSE for NULL strings and
empty strings, and gst_tag_list_get_pointer() return FALSE for
NULL pointers, like we do with dates and buffers.

Fixes #560345.

Also warn if an element or application tries to add a field with an
empty string to a structure (NULL strings are still needed and
allowed though) and do all those checks in the right function.

Fixes #559643.

Add convenience wrapper for gst_structure_id_empty_new() plus
gst_structure_id_set() and use it in a few places.

API: gst_structure_id_new()

Use gst_structure_id_empty_new() in combination with GST_QUARK
rather than gst_structure_id_new() when creating message, event,
query and taglist structures. Mostly just because we can.

Deprecate the old _set_stream_time and _get_last_stream_time methods because
they are now equivalent to the better named _set/_get_start_time.

Use the element START_TIME to keep track of the running time when the pipeline
paused so that it can be used to restore the base_time.
Take the start_time before setting the children to PAUSED so that we can
distribute the start_time to the children.

Simply set the start_time and base_time on the element instead of calling the
setters.

Set the start_time of the bin on the elements when they are added to the
pipeline and when a state change happens.

Add a start_time field and some methods. The start_time will contain the
running_time of when the element last went to paused. This time can be user to
report the position in PAUSED but also to do more correct clipping and
stepping later.

gst_adapter_masked_scan_uint32 could return values smaller than offset
if the first byte(s) of the mask are 0 and the pattern matches the
beginning of the adapter.
Added examples to documentation of gst_adapter_masked_scan_uint32().
Also added some more masked boundary tests.
Fixes #584118

Add pad private structure and move the new chainlistfunc into the private
struct. This avoids ABI breakage and allows us to expand in the future.

This was accidentially removed by my last commit.

Avoid leaking the caps of the dest buffer and avoid doing needless caps
refs.
When the source and target buffers are the same, return immediatly.

Fixes bug #582588.

This reverts commit bebfde75.

This change shouldn't be done in a stable release series as
applications are actually expecting the sender to be an
GstElement. One example is totem.

The check-enum-gettypes rule didn't work for 'make distcheck' since
it makes assumptions about the location of the source files from the
current working directory which isn't true during distchecking.

Attempt to fix the 'cannot create regular file build/image.entitites:
file exists' error I got.

Remove an assertion, this is not really an error in all cases.
Fixes #582010

Enable the monotonic clock by default when we can.
Fixes #583554

Keep a ref to the task on the pad so that a concurrent stop can stop and join
the task.

Unset the running flag after we released the lock for posting the stream-status
message. If we set the running flag to FALSE too early, the join method will
just continue without waiting for the message to be posted, leading to potential
crashes.

Only update the preset from system, if we had a preset before and system
version is newer.

When writing a cache chunk fails, we were freeing the node and jump to a final
cleanup which dereferenced a null pointer. Leve freeing the node to the cleanup
code in fail_free_list. (sorry for committing wrong fix before).