The fact that it broke the unit tests is probably a signal that it
should stay at 'Centre'.

1) MD event workspace have changed from DND axis basis to SQW axis basis
2) DND axis are fixed to read different number of dimensions objects properly but are not used at the moment.

Each submodule has its own directory along with the PythonAlgorithms.

Refs #6235

When setting the initial value for the run_start property, use the time
from the most recently received packet instead of the current system
time.  (Doesn't matter too much when using actual live data, but when
replaying old data it makes a lot more sense.)

Refs #6235

Each file will require the __main__ check at the bottom.

Python 2.6 doesn't have the discover module.
This reverts commit 54b3849c.

This reverts commit 956a6a92.

This reverts commit 2830dd04.

Other properties such as WriteXError are still not tested.

Signed-off-by: Karl Palmen <karl.palmen@stfc.ac.uk>

Still to test content of output file

Signed-off-by: Karl Palmen <karl.palmen@stfc.ac.uk>

Just tests that SaveAscii executes.

Signed-off-by: Karl Palmen <karl.palmen@stfc.ac.uk>

It will only show the centre detector for lean events & histo workspaces.

This is for both interior peaks and Edge peaks. Interior peaks now
give GSAS results matching the best results from other methods. The
Edge pixels are now not terrible, but still could need improvements.

We use the timestamp from the most recently received packet to set the
value of the run_start property when we initialize the workspace.  (The
property will be overwritten with a proper value once we get a RunStatus
packet saying a new run is starting.)

Refs #6259

...filter events from the live stream when the run has been paused.  (By
default, paused events ARE filtered out.)

Also fix the timestamp values for everything in the sample logs.

Refs #6235