Add macOS recording script example (#4)

* Add recording example MacOS, sox and pysox

* Add more documentation on full system

README.md

@@ -5,55 +5,84 @@
 ### Raspberry Pi 4 - 64bit
 
 Bring it up and run:
-docker compose -f docker-compose.rpi4.yml up -d --build
+`docker compose -f docker-compose.rpi4.yml up -d --build`
 
 To run the watcher/analyzer:
-docker compose -f docker-compose.rpi4.yml exec web python manage.py runscript analyze
+`docker compose -f docker-compose.rpi4.yml exec web python manage.py runscript analyze`
 
 To run the Django test cases:
-docker compose -f docker-compose.rpi4.yml exec web python manage.py test
+`docker compose -f docker-compose.rpi4.yml exec web python manage.py test`
 
 To take it down:
-docker compose -f docker-compose.rpi4.yml down
+`docker compose -f docker-compose.rpi4.yml down`
 
 To run bash within the docker instance:
-docker compose -f docker-compose.rpi4.yml exec web bash
+`docker compose -f docker-compose.rpi4.yml exec web bash`
 
 To rebuild the image for a reason (e.g. after pip change)
-docker compose -f docker-compose.rpi4.yml down; docker compose build --no-cache
+`docker compose -f docker-compose.rpi4.yml down; docker compose build --no-cache`
 
 ### MacOS M1
 
 Bring it up and run:
-docker compose -f docker-compose.macm1.yml up -d --build
+`docker compose -f docker-compose.macm1.yml up -d --build`
 
 To run the watcher/analyzer:
-docker compose -f docker-compose.macm1.yml exec web python manage.py runscript analyze
+`docker compose -f docker-compose.macm1.yml exec web python manage.py runscript analyze`
 
 To run the Django test cases:
-docker compose -f docker-compose.macm1.yml exec web python manage.py test
+`docker compose -f docker-compose.macm1.yml exec web python manage.py test`
 
 To take it down:
-docker compose -f docker-compose.macm1.yml down
+`docker compose -f docker-compose.macm1.yml down`
 
 To run bash within the docker instance:
-docker compose -f docker-compose.macm1.yml exec web bash
+`docker compose -f docker-compose.macm1.yml exec web bash`
 
 To rebuild the image for a reason (e.g. after pip change)
-docker compose -f docker-compose.macm1.yml down; docker compose build --no-cache
+`docker compose -f docker-compose.macm1.yml down; docker compose build --no-cache`
 
 ## Recording audio
 
-The docker container itself does not record audio to audio_inbox directory. See below for basic examples for recording 30-second long WAV files to audio_inbox.
+The docker container itself does not record audio to the `audio_inbox` directory. See below for basic examples for recording 30-second properly-dated WAV files to audio_inbox.
 
 ### Raspberry Pi 4
 
+Prerequisites: arecord
+
 To record an audio stream to "audio_inbox", run the following.
 `python script_examples/audio_recording_rpi.py`
 
 ### MacOS M1
 
-Prerequisites: sox
+Prerequisites: sox, pysox
 
 To record an audio stream to "audio_inbox", run the following.
-`python script_examples/audio_recording_macosm1.py`
+`python script_examples/audio_recording_macos.py`
+
+## Putting it all together
+
+In order to create a self-running, restart-tolerant installation, you'll need to setup a few extra things. For example, the setup for a Raspberry Pi 4 is provided below. Other systems will require other methods of starting and restarting the processes.
+
+Prerequisites: Raspberry Pi 4 with clean 64-bit system, `docker`, `docker-compose` and `arecord`.
+
+### Confirm that the system works
+
+In one terminal, run:
+
+```
+mkdir audio_inbox
+mkdir audio_post_analyze
+docker compose -f docker-compose.rpi4.yml up -d --build
+docker compose -f docker-compose.rpi4.yml exec web python manage.py runscript analyze
+```
+
+In another terminal, start the recording:
+
+```
+python script_examples/audio_recording_macos.py
+```
+
+### Setup systemd to auto-start the analyze and recording processes on boot
+
+<snip> More to come ...

script_examples/audio_recording_macos.py

@@ -0,0 +1,30 @@
+# Requires sox and pysox: install with:
+#  brew install sox
+#  pip install sox
+
+import sox
+from datetime import datetime
+
+
+def main():
+
+    duration_secs = 15
+    output_directory = "audio_inbox"
+
+    while True:
+        filename = datetime.now().strftime("%Y-%m-%d-%H:%M:%S.wav")
+        print(f"{output_directory}/{filename}")
+        args = [
+            "-d",
+            "-t",
+            "wav",
+            f"{output_directory}/{filename}",
+            "trim",
+            "0",
+            f"{duration_secs}",
+        ]
+        sox.core.sox(args)
+
+
+if __name__ == "__main__":
+    main()