Sequencer: overview

Sequencer tools are provided that allow you to automate data taking, for example to take a series of runs with detectors set at different voltages.

There are two versions:

• msequencer which supports a custom "MSL" syntax
• pysequencer which supports python syntax

Both versions provide a web interface that allows you to edit/load scripts, start/stop them, and track their progress.

Feature comparison

	msequencer	pysequencer
Script language	MSL	Python
Language support	Midas commands only	Midas commands, plus any python functions/libraries
Full web interface
Webpage name	Sequencer	PySequencer
ODB directory	/Sequencer	/PySequencer
Script file extension	.msl	.py
Command to start program	msequencer -D	python $MIDASSYS/python/midas/sequencer.py

Web interface

The sequencer web pages allow you to:

• load, edit and save sequencer script
• start and stop scripts
• input parameters when the script starts
• track the state of the script, including progress through loops and any errors that were encountered

Starting the sequencer programs

If msequencer/sequencer.py is not running when you open the relevant webpage, the GUI will ask you whether it should start it automatically. If this fails, you should run the shown command in a terminal and debug any problems.

Loading and saving

The icon allows you to create a new file, allows you to open an existing file, and allows you to save the file you are currently editing. You can click the + icon to open a new tab and edit multiple files at once.

Starting a script

The icon will load and start a script. You will be prompted to confirm that you want to start the script, be prompted to provide any parameters that the script requires, and then the script will start.

Debugging a script

For debugging, click a line number in the editor gutter to add or remove a breakpoint; active breakpoints are shown as red circles and are synchronized through the ODB so they appear in every browser window. The Commands → Breakpoints... dialog lists all breakpoints and lets you enable, disable, delete, or jump to them. A normal start ignores the breakpoints and displays them in grey, while the play icon containing a starts in debug mode and runs until it reaches an enabled breakpoint, where it pauses before executing that line. Use the resume control to continue to the next breakpoint or to execute one line at a time; pause and wait icons beside the current line distinguish a debugger stop from a script that is still running inside a wait condition.

Monitoring a script

While a script is running, you are shown the state of any local variables, a progress bar for any loops or "wait" conditions, and error messages if there are any problems.

The icon allows you to pause a script on the current line that is executing, will stop execution immediately, and there are also options for stopping a sequence after the current run ends, and resuming execution that has been paused.

Running multiple scripts in succession

You may run a list of sequence scripts back-to-back (a sequence of sequences!). This is done using the "next files" dialog (or the "Next filename" array in the ODB settings). On the web interface, click "Set as next" and the current file being edited will be added to the queue.

The list is not displayed until you start a script. Once a script is running, you can continue to add more files to the list, reorder the list, or remove scripts from the list.

ODB interface

The sequencers are generally designed to be controlled via the web interface. However they can also be controlled via settings in the ODB, for example to integrate sequencer usage with other pages/tools that your experiment uses.

The interfaces for both sequencers are very similar, but use different top-level ODB directories (/Sequencer vs /PySequencer). The relevant controls are listed below. See /Sequencer and /PySequencer for the full details.

• In /Sequencer/Command or /PySequencer/Command:
  • Load new file - load the file specified in /Sequencer/State/Filename or /PySequencer/State/Filename
  • Start script - start the script that has been loaded
  • Pause script - pause the script (by preventing the next line of code from executing)
  • Resume script - resume the script after pausing it
  • Stop immediately - stop the sequence as soon as possible
  • Stop after run - stop the sequence when the current run ends. The Python sequencer stops however the run ends. msequencer only stops when the script itself stops the run (with TRANSITION STOP)
  • Cancel stop after run - cancel the "Stop after run" request
  • Debug script - start the script in debug mode and pause when an enabled breakpoint is reached
  • Step over - execute the current line and pause again when running in debug mode
• In /Sequencer/State or /PySequencer/State
  • Path - specify a subdirectory beneath <experiment_dir>/userfiles/sequencer where your files are located. May be empty.
  • Filename - filename to load (relative to <experiment_dir>/userfiles/sequencer or <experiment_dir>/userfiles/sequencer/<path>)
  • Next filename - if running multiple scripts in succession, then next file to load once the current script has finished.
  • Message - if "Message wait" is true, the sequence will not continue until "Message" has been reset to an empty string - see Sequencer: scripting for more details about sequencer messages.

Running multiple instances

You can run multiples instances of the sequencers by supplying the -c argument when starting the sequencer program. This defines a "suffix" which is used in the program name, ODB, and webpages.

$MIDASSYS/bin/msequencer -c MySuffix

odbedit -c "create string '/Alias/My new sequencer'"
odbedit -c "set '/Alias/My new sequencer' '?cmd=Sequencer&SeqODB=/SequencerMySuffix'"

python $MIDASSYS/python/midas/sequencer.py -c MySuffix

odbedit -c "create string '/Alias/My new pysequencer'"
odbedit -c "set '/Alias/My new pysequencer' '?cmd=Sequencer&SeqODB=/PySequencerMySuffix'"