mbradley612 / hhscstartline Goto Github PK

Sometimes when FitPC starts, it does not start in HiRes mode. Can we force it into HiRes?

Also, would be nice to have some kind of splash screen while it is starting.

Warnings about cards.pcm.rear etc

Instructions to resolve at

http://stackoverflow.com/questions/7088672/pyaudio-working-but-spits-out-error-messages-each-time

Following feedback from the club's neighbours, Luke has asked for the finish horn to be shorter

If you disable the lights in the ini file then controllers.py crashes at line 819 because comPort isn’t defined. This is because when you check for the lights being enabled at line 773, comPort is only becomes defined if Lights enabled = Y. I got round this by adding a line to the else to set comPort=0 which will cause the any if statements to return a False

startline.sh does not provide the location of the startline.ini file.

This causes the application to crash at startup when running in Linux

Makes it really obvious which start you're doing.

The log file just grows forever. Need some kind of cycling log.

its not needed because there is only one fleet to choose from!!!

Reset button should clear the fleets and finishes.

After implementing issue N to delay the General Recall of the last fleet by 60 seconds, we now have the problem that there is no horn or beeps for the five minute warning.

This is due to the slightly cacky way that horns and beeps are implemented for the five minute warning.

PYTHONPATH does not contain a valid directory if the script is launched automatically i.e. as part of an xsession file.

2 Changes are requried

1. PYTHONPATH needs to be exported as an absolute path name
2. $PYTHONPATH needs to be prefixed to the call to start controllers.py and the location of startline.ini

e.g.

export PYTHONPATH=
python $PYTHONPATH/controllers/controllers.py $PYTHONPATH/../startline.ini

The primary purpose of this enhancement is to support an external Human Interface Device (HID) as an alternative to pressing the buttons on the screen. HHSC is buying a Delcom handheld button which we will bind to the Gun and Finish button.

http://www.delcomproducts.com/productdetails.asp?PartNumber=706400

As per lights hardware, refactor audio manager to call audio API in its own thread. This insulates the Tk event thread from IO,

From Matthew West:

if the start is delayed, we should be raising the AP (postponed) flag. The warning signal should be sounded 1 minute after removal. The sequence should therefore be:

-6: gun
-5: gun

and then as a normal start.

We are no longer using the F flag start, so the simple change is to repurpose the F flag start to a "Postponed" start and make the first gun at 6 minutes instead of 10. The label underneath should show "(6,5,4,1,0 minutes)".

Also change the Class flag start label to "(5,4,1,0)"

Refactor the lights module so that it runs in its own thread instead of using the Tk event loop for all serial port IO.

http://stackoverflow.com/questions/2846653/python-multithreading-for-dummies gives a simple example of how to use a python Queue as the interface between threads, and explains that Queues are threadsafe.

General recall for a single race restarts the race automatically. This is not the behaviour we want. It should sound the horns then behave more like an abandon start.

I think we need to disable power management. See http://ubuntuforums.org/showthread.php?t=788276

When shutting down at line 739 there is no guard condition to check the lights were enabled before sending the off command to them. This causes the shutdown sequence to fail if running with lights = N

There is currently no way to delete a finish. A delete button would enable a race officer to keep the finishes nice and neat, for what should be very little work.

We are running sensorsd to capture temperature information. But the config isn't very good, and we are getting continuous errors in the logs for a null temperature on a fan. Need to tune the sensors module configuration down so that we are monitoring only the sensors that are available to us.

Startine currently only attempts to create a PyAudio connection object on startup. if the audio is not connected at startup, or becomes disconnected after startup, the only way to recover is to restart Startline.

The fix is to:

1. catch any PyAudio exceptions
2. track the state of the connection (are we connected or not? if not, try to create a connection object before playing a sound)

Try
http://askubuntu.com/questions/362722/how-to-fix-splash-screen-in-all-ubuntu-releases

These buttons are a bit fiddly small at the moment on the touch screen.

For example:

Jun 30 22:01:48 startline kernel: [350816.004273] hub 2-0:1.0: port 1 disabled by hub (EMI?), re-enabling...
Jun 30 22:01:48 startline kernel: [350816.004302] usb 2-1: USB disconnect, device number 2
Jun 30 22:01:48 startline kernel: [350816.004628] ftdi_sio ttyUSB0: usb_serial_generic_submit_read_urb - usb_submit_urb failed: -19
Jun 30 22:01:48 startline kernel: [350816.004659] ftdi_sio ttyUSB0: error from flowcontrol urb
Jun 30 22:01:48 startline kernel: [350816.005044] ftdi_sio ttyUSB0: FTDI USB Serial Device converter now disconnected from ttyUSB0
Jun 30 22:01:48 startline kernel: [350816.005126] ftdi_sio 2-1:1.0: device disconnected
Jun 30 22:01:48 startline kernel: [350816.244147] usb 2-1: new full-speed USB device number 3 using uhci_hcd
Jun 30 22:01:48 startline kernel: [350816.422411] usb 2-1: New USB device found, idVendor=0403, idProduct=6001
Jun 30 22:01:48 startline kernel: [350816.422423] usb 2-1: New USB device strings: Mfr=1, Product=2, SerialNumber=0
Jun 30 22:01:48 startline kernel: [350816.422430] usb 2-1: Product: USB <-> Serial
Jun 30 22:01:48 startline kernel: [350816.422436] usb 2-1: Manufacturer: FTDI
Jun 30 22:01:48 startline kernel: [350816.428556] ftdi_sio 2-1:1.0: FTDI USB Serial Device converter detected
Jun 30 22:01:48 startline kernel: [350816.428759] usb 2-1: Detected FT232BM
Jun 30 22:01:48 startline kernel: [350816.428781] usb 2-1: Number of endpoints 2
Jun 30 22:01:48 startline kernel: [350816.428800] usb 2-1: Endpoint 1 MaxPacketSize 64
Jun 30 22:01:48 startline kernel: [350816.428819] usb 2-1: Endpoint 2 MaxPacketSize 64
Jun 30 22:01:48 startline kernel: [350816.428837] usb 2-1: Setting MaxPacketSize 64
Jun 30 22:01:48 startline kernel: [350816.430655] usb 2-1: FTDI USB Serial Device converter now attached to ttyUSB1
Jun 30 22:01:49 startline mtp-probe: checking bus 2, device 3: "/sys/devices/pci0000:00/0000:00:1d.0/usb2/2-1"
Jun 30 22:01:49 startline mtp-probe: bus: 2, device: 3 was not an MTP device
Jun 30 22:01:49 startline kernel: [350816.996275] hub 2-0:1.0: port 1 disabled by hub (EMI?), re-enabling...
Jun 30 22:01:49 startline kernel: [350816.996302] usb 2-1: USB disconnect, device number 3
Jun 30 22:01:49 startline kernel: [350816.997107] ftdi_sio ttyUSB1: FTDI USB Serial Device converter now disconnected from ttyUSB1
Jun 30 22:01:49 startline kernel: [350816.997211] ftdi_sio 2-1:1.0: device disconnected
Jun 30 22:01:49 startline kernel: [350817.244101] usb 2-1: new full-speed USB device number 4 using uhci_hcd
Jun 30 22:01:49 startline kernel: [350817.420376] usb 2-1: New USB device found, idVendor=0403, idProduct=6001
Jun 30 22:01:49 startline kernel: [350817.420389] usb 2-1: New USB device strings: Mfr=1, Product=2, SerialNumber=0
Jun 30 22:01:49 startline kernel: [350817.420396] usb 2-1: Product: USB <-> Serial
Jun 30 22:01:49 startline kernel: [350817.420402] usb 2-1: Manufacturer: FTDI
Jun 30 22:01:49 startline kernel: [350817.425494] ftdi_sio 2-1:1.0: FTDI USB Serial Device converter detected
Jun 30 22:01:49 startline kernel: [350817.425614] usb 2-1: Detected FT232BM
Jun 30 22:01:49 startline kernel: [350817.425625] usb 2-1: Number of endpoints 2
Jun 30 22:01:49 startline kernel: [350817.425634] usb 2-1: Endpoint 1 MaxPacketSize 64
Jun 30 22:01:49 startline kernel: [350817.425643] usb 2-1: Endpoint 2 MaxPacketSize 64
Jun 30 22:01:49 startline kernel: [350817.425652] usb 2-1: Setting MaxPacketSize 64
Jun 30 22:01:49 startline kernel: [350817.427523] usb 2-1: FTDI USB Serial Device converter now attached to ttyUSB1
Jun 30 22:01:49 startline mtp-probe: checking bus 2, device 4: "/sys/devices/pci0000:00/0000:00:1d.0/usb2/2-1"
Jun 30 22:01:49 startline mtp-probe: bus: 2, device: 4 was not an MTP device

e.g.

File "C:\Users\mbradley\git\HHSCStartLine\HHSCStartLine\src\lightsui\hardware.py", line 221, in reconnect
self.connect()
File "C:\Users\mbradley\git\HHSCStartLine\HHSCStartLine\src\lightsui\hardware.py", line 211, in connect
self.reconnect()
File "C:\Users\mbradley\git\HHSCStartLine\HHSCStartLine\src\lightsui\hardware.py", line 221, in reconnect
self.connect()
File "C:\Users\mbradley\git\HHSCStartLine\HHSCStartLine\src\lightsui\hardware.py", line 211, in connect
self.reconnect()
File "C:\Users\mbradley\git\HHSCStartLine\HHSCStartLine\src\lightsui\hardware.py", line 221, in reconnect
self.connect()
File "C:\Users\mbradley\git\HHSCStartLine\HHSCStartLine\src\lightsui\hardware.py", line 210, in connect
self.beReconnecting()
File "C:\Users\mbradley\git\HHSCStartLine\HHSCStartLine\src\lightsui\hardware.py", line 105, in beReconnecting
self.setSessionState(RECONNECTING)
File "C:\Users\mbradley\git\HHSCStartLine\HHSCStartLine\src\lightsui\hardware.py", line 94, in setSessionState
logging.info("Session state is %s" % self.sessionStateDescription())
File "C:\Python27\lib\logging__init.py", line 1592, in info
root.info(msg, args, *kwargs)
File "C:\Python27\lib\logging__init.py", line 1132, in info
self.log(INFO, msg, args, **kwargs)
File "C:\Python27\lib\logging__init.py", line 1250, in log
self.handle(record)
File "C:\Python27\lib\logging__init.py", line 1260, in handle
self.callHandlers(record)
File "C:\Python27\lib\logging__init.py", line 1300, in callHandlers
hdlr.handle(record)
File "C:\Python27\lib\logging__init__.py", line 744, in handle
self.emit(record)
File "C:\Python27\lib\logging__init__.py", line 926, in emit
StreamHandler.emit(self, record)
File "C:\Python27\lib\logging__init__.py", line 870, in emit
self.handleError(record)
File "C:\Python27\lib\logging__init__.py", line 797, in handleError
None, sys.stderr)
File "C:\Python27\lib\traceback.py", line 125, in print_exception
print_tb(tb, limit, file)
File "C:\Python27\lib\traceback.py", line 69, in print_tb
line = linecache.getline(filename, lineno, f.f_globals)
File "C:\Python27\lib\linecache.py", line 14, in getline
lines = getlines(filename, module_globals)
RuntimeError: maximum recursion depth exceeded

Integrate with weewx for live wind speed, five minute average, five minute gusts

We have logic to enable the fleet finish buttons when the first finish is created, but no logic to disable the buttons if all finishes are removed, e.g. on a reset start event.

https://pypi.python.org/pypi/Tka11y/0.1.1

Luke reported that we experienced a Python windows crash approximately 40 seconds before the start.

There is currently no way to store or print out the finish list. On Windows, it was possible to do a screen grab. On the new Startline kiosk, even this isn't available.

Test case

1. two fleet start, e.g. small handicap and large handicap
2. general recall of first fleet, e.g. small handicap
3. start (i.e. last ten seconds) of second fleet e.g. large handicap start has double beeps and horns.

Luke would like a 1 second delay between horns for a general recall

A quick test of Startline on a Raspberry Pi (original model B) revealed that PyAudio with PortAudio on Raspbian results in audio that breaks up - even in simple in the simplest use case as per the "play" example on http://people.csail.mit.edu/hubert/pyaudio/

In contrast, an equivalent sample test of the PyGame pygame.mixer API worked perfectly. PyGame also includes a built in queuing mechanism, so it looks like we can greatly simplify the audio manager in Startline, see http://www.nerdparadise.com/tech/python/pygame/basics/part3/.

As part of this, test the behaviour on Windows and Linux if the line out is disconnected and reconnected.

Big banner at the top
We could also put wind info there - may be bramblemet too.

The list of fleets is to short for all the fleets configured in the ini file.

Fix is to make the frame bigger - also good to have scroll bars.

probably one for implementing when changing the UI.