Shell Programming and Scripting

Are you running sox once per loop? Why not run it once and just read from it as long as you need?

Code:

sox | while true
do
        DATA=$(dd bs=8192 count=1) # Get 8K bytes from the stream
done

That might solve some timing too, since you'd only get data as fast as sox captures it...

This is a new updated version of the AudioScope with limited scans which can be from single shot storage to pseudo-continuous.

Beware when altering any values as there is NO error detection as yet.

Now starting to create _functions_ as this becomes more mature...

Corona688, just eyeballed your upload and will see if it fits my requirements...

Well this script code is now taking shape...

Apologies for any typos...

It has some _command_line_ stuff to alter certain parameters and a few status line updates.

The _commands_ are not case sensitive and there is a starter help window built in also.

HELP and some other commands will auto switch from capture mode to rerun mode displaying the last capture taken. READ THE CODE for more information...

There is a commented out section that would, if edited according to the instructions in the code, generate another shell script and run it in a separate terminal. This will generate an 8 second burst of 1KHz sinewave and as long as the window is the active one just press ENTER to keep rerunning or Ctrl-C to quit the script and close down the terminal...

Connecting the output of the signal to the mic input and adjusting the audio system's gain controls will display a nice sinewave within the limits of standard text mode...

If you edit and run the generator script then SOX must be installed...

I hope this will become a practical kids learning tool along with some simple HW construction in the not too distant future...

As you will see almost EVERYTHING that is needed for testing, calibration running and the like will be generated from this one script only...

Enjoy...

Changes made so that the start position of any timebase scan can be altered. The effective speed is not yet done. This will be the next upload. After that timebase calibration, then a means of saving the settings and being able to start the script with "your" user settings...

Minor changes made including removal of UPPERCASE conversion mode as this could still create a user error.

I hope you guys and gals taking a look at this are enjoying what you are seeing...

What started as a fun upload is evolving into something serious that 10 year olds could understand. I have tried to write the code in such a way as to make it as visaully noise free as possible to youngsters.

Watch for wordwrapping, etc...

To the Admins; thanks for allowing me to do this...

Bazza...

User variable timebase is now complete.

The command is TBVAR then follow the prompts...

It is _in_theory_ error free when typing in wrong characters and/or values... ;o)

The "scan_jump" variable sets the timebase speed, number 1 is the fastest and the maximium allowable number is the slowest and varies according to the "scan_start" position.

The next upload will be the _calibrated_ timebase ranges...

The oscillator commented out of the code will probably be used for checking this...

These will probably be FASTEST, 1mS, 10mS, 100mS and SLOW...

DEMO will not be able to have FASTEST and 1mS.

/dev/dsp will not be able to have FASTEST...

This is slowly evolving into fully working tool...

Watch for wordwrapping, etc...

I think case statements could shrink your kbinput and status checks a lot:

Code:

case "$kbinput" in
TEN)  echo "got TEN"
        ;;
HUNDRED) echo "got HUNDRED"
        ;;
...
*) echo "Unknown kbinput"
    ;;
esac

For tiny if-statements with only one thing in them, you can use && shortcuts:

Code:

[ $hold -eq 1 ] && waveform

Hi Corona688...

Apologies for any typos...

Thanks for the heads up, you are a real pal...

I will fork the existing code and shorten it in your style but upload the longhand version for the foreseeable future.

I have done it this way so that a 10 year old can see the progression.

Does it show that much that I am an amateur... ;o)

BTW, calibrated timebase ranges are finished and coupling the earphone output to the mic input and setting the levels accordingly works a treat. Also added a small offset to ensure that the trace is linear when there is zero input. This removes the sound card's mid point bit error...

Working on a config file now so that user parameters can be saved and auto-restarted. This will take a while...

Bazza...