Rebuild audio streamer event handlers

README.md

@@ -1,10 +1,10 @@
 # Croaker
 
-A shoutcast audio player designed for serving D&D session music.
+A shoutcast server designed primarily for streaming D&D session music.
 
 ### Features
 
-* Native streaming of MP3 sources direct to your shoutcast / icecast server
+* Native streaming of MP3 sources direct to your clients
 * Transcoding of anything your local `ffmpeg` installation can convert to mp3
 * Playlists are built using symlinks
 * Randomizes playlist order the first time it is cached
@@ -21,7 +21,7 @@ A shoutcast audio player designed for serving D&D session music.
 
 ## What? Why?
 
-Because I run an online D&D game, which includes a background music stream for my players. The stream used to be served by liquidsoap and controlled by a bunch of bash scripts I cobbled together which are functional but brittle, and liquidsoap is a nightmare for the small use case. Also, this required me to have a terminal window open to my media server to control liquidsoap directly, and I'd rather integrate the music controls directly with the rest of my DM tools, all of which run on my laptop.
+Because I run an online D&D game, which includes a background music stream for my players. The stream used to be served by liquidsoap and icecast and controlled by a bunch of bash scripts I cobbled together which are functional but brittle, and liquidsoap is a nightmare for the small use case. Also, this required me to have a terminal window open to my media server to control liquidsoap directly, and I'd rather integrate the music controls directly with the rest of my DM tools, all of which run on my laptop.
 
 *Now that is a powerful yak! -- Aesop Rock (misquoted)*
 

src/croaker/cli.py

@@ -85,11 +85,12 @@ def setup(context: typer.Context):
 def start(
     context: typer.Context,
     daemonize: bool = typer.Option(True, help="Daemonize the server."),
+    shoutcast: bool = typer.Option(True, help="Stream to shoutcast."),
 ):
     """
     Start the Croaker command and control server.
     """
-    server.start(daemonize=daemonize)
+    server.start(daemonize=daemonize, shoutcast_enabled=shoutcast)
 
 
 @app.command()

src/croaker/server.py

@@ -2,7 +2,6 @@ import logging
 import os
 import queue
 import socketserver
-import threading
 from pathlib import Path
 from time import sleep
 
@@ -93,7 +92,7 @@ class RequestHandler(socketserver.StreamRequestHandler):
         return self.send("\n".join(f"{cmd} {txt}" for cmd, txt in self.supported_commands.items()))
 
     def handle_STOP(self, args):
-        return self.server.stop_event.set()
+        return self.streamer.stop_requested.set()
 
     def handle_STFU(self, args):
         self.send("Shutting down.")
@@ -110,9 +109,6 @@ class CroakerServer(socketserver.TCPServer):
     def __init__(self):
         self._context = daemon.DaemonContext()
         self._queue = queue.Queue()
-        self.skip_event = threading.Event()
-        self.stop_event = threading.Event()
-        self.load_event = threading.Event()
         self._streamer = None
         self.playlist = None
 
@@ -121,8 +117,6 @@ class CroakerServer(socketserver.TCPServer):
 
     @property
     def streamer(self):
-        if not self._streamer:
-            self._streamer = AudioStreamer(self._queue, self.skip_event, self.stop_event, self.load_event)
         return self._streamer
 
     def bind_address(self):
@@ -143,7 +137,7 @@ class CroakerServer(socketserver.TCPServer):
         self._context.files_preserve = [self.fileno()]
         self._context.open()
 
-    def start(self, daemonize: bool = True) -> None:
+    def start(self, daemonize: bool = True, shoutcast_enabled: bool = True) -> None:
         """
         Start the shoutcast controller background thread, then begin listening for connections.
         """
@@ -153,11 +147,13 @@ class CroakerServer(socketserver.TCPServer):
             self._daemonize()
         try:
             logger.debug("Starting AudioStreamer...")
+            self._streamer = AudioStreamer(self._queue, shoutcast_enabled=shoutcast_enabled)
             self.streamer.start()
             self.load("session_start")
             self.serve_forever()
         except KeyboardInterrupt:
-            logger.info("Shutting down.")
+            logger.info("Keyboard interrupt detected.")
+            self.streamer.shutdown_requested.set()
             self.stop()
 
     def stop(self):
@@ -165,12 +161,14 @@ class CroakerServer(socketserver.TCPServer):
 
     def ffwd(self):
         logger.debug("Sending SKIP signal to streamer...")
-        self.skip_event.set()
+        self.streamer.skip_requested.set()
 
     def clear_queue(self):
-        logger.debug("Requesting a reload...")
+        logger.debug("Requesting a clear...")
-        self.streamer.load_requested.set()
+        self.streamer.clear_requested.set()
-        sleep(0.5)
+        while self.streamer.clear_requested.is_set():
+            sleep(0.001)
+        logger.debug("Cleared")
 
     def list(self, playlist_name: str = None):
         if playlist_name:
@@ -179,12 +177,14 @@ class CroakerServer(socketserver.TCPServer):
 
     def load(self, playlist_name: str):
         logger.debug(f"Switching to {playlist_name = }")
+        self.streamer.stop_requested.set()
         if self.playlist:
             self.clear_queue()
         self.playlist = load_playlist(playlist_name)
         logger.debug(f"Loaded new playlist {self.playlist = }")
         for track in self.playlist.tracks:
             self._queue.put(str(track).encode())
+        self.streamer.start_requested.set()
 
 
 server = CroakerServer()

src/croaker/streamer.py

@@ -2,6 +2,7 @@ import logging
 import os
 import queue
 import threading
+from dataclasses import dataclass
 from functools import cached_property
 from pathlib import Path
 from time import sleep
@@ -19,21 +20,27 @@ class AudioStreamer(threading.Thread):
     those files to the icecast server.
     """
 
-    def __init__(self, queue, skip_event, stop_event, load_event, chunk_size=4096):
+    def __init__(self, queue: queue.Queue = queue.Queue(), chunk_size: int = 8092, shoutcast_enabled: bool = True):
         super().__init__()
         self.queue = queue
-        self.skip_requested = skip_event
-        self.stop_requested = stop_event
-        self.load_requested = load_event
         self.chunk_size = chunk_size
-
+        self._shoutcast_enabled = shoutcast_enabled
-    @property
+        self.skip_requested = threading.Event()
-    def silence(self):
+        self.stop_requested = threading.Event()
-        return FrameAlignedStream.from_source(Path(__file__).parent / "silence.mp3", chunk_size=self.chunk_size)
+        self.start_requested = threading.Event()
+        self.clear_requested = threading.Event()
+        self.shutdown_requested = threading.Event()
 
     @cached_property
-    def _shout(self):
+    def silence(self):
+        return FrameAlignedStream(Path(__file__).parent / "silence.mp3", chunk_size=self.chunk_size)
+
+    @cached_property
+    def _out(self):
+        if self._shoutcast_enabled:
             s = shout.Shout()
+        else:
+            s = debugServer()
         s.name = "Croaker Radio"
         s.url = os.environ["ICECAST_URL"]
         s.mount = os.environ["ICECAST_MOUNT"]
@@ -45,23 +52,31 @@ class AudioStreamer(threading.Thread):
         return s
 
     def run(self):  # pragma: no cover
-        while True:
+        while not self.shutdown_requested.is_set():
             try:
-                logger.debug(f"Connecting to shoutcast server at {self._shout.host}:{self._shout.port}")
+                self.connect()
-                self._shout.open()
+                self.stream_forever()
+                break
             except shout.ShoutException as e:
                 logger.error("Error connecting to shoutcast server. Will sleep and try again.", exc_info=e)
                 sleep(3)
-                continue
+        self.shutdown()
+        self.shutdown_requested.clear()
 
-            try:
+    def connect(self):
-                self.stream_queued_audio()
+        logger.info(f"Connecting to downstream server at {self._out}")
-            except Exception as exc:
+        self._out.close()
-                logger.error("Caught exception.", exc_info=exc)
+        self._out.open()
-        self._shout.close()
+
+    def shutdown(self):
+        if hasattr(self, "_out"):
+            self._out.close()
+            del self._out
+        self.clear_queue()
+        logger.info("Shutting down.")
 
     def clear_queue(self):
-        logger.debug("Clearing queue...")
+        logger.info("Clearing queue...")
         while not self.queue.empty():
             self.queue.get()
 
@@ -72,45 +87,82 @@ class AudioStreamer(threading.Thread):
         try:
             track = Path(self.queue.get(block=False).decode())
             logger.debug(f"Streaming {track.stem = }")
-            return FrameAlignedStream.from_source(track, chunk_size=self.chunk_size), track.stem
+            return FrameAlignedStream(track, chunk_size=self.chunk_size), track.stem
         except queue.Empty:
             logger.debug("Nothing queued; enqueing silence.")
         except Exception as exc:
             logger.error("Caught exception; falling back to silence.", exc_info=exc)
         return self.silence, "[NOTHING PLAYING]"
 
-    def stream_queued_audio(self):
+    def pause_if_necessary(self):
-        stream = None
+        while self.stop_requested.is_set():
-        title = None
+            if self.start_requested.is_set():
-        next_stream = None
+                self.stop_requested.clear()
-        next_title = None
+                self.start_requested.clear()
-
+                return
-        while True:
+            sleep(0.001)
-            stream, title = (next_stream, next_title) if next_stream else self.queued_audio_source()
-            logging.debug(f"Starting stream of {title = }, {stream = }")
-            self._shout.set_metadata({"song": title})
-            next_stream, next_title = self.queued_audio_source()
 
+    def stream_forever(self):
+        while not self.shutdown_requested.is_set():
+            self.pause_if_necessary()
+            stream, title = self.queued_audio_source()
+            logging.debug(f"Starting stream of {title = }")
+            self._out.set_metadata({"song": title})
             for chunk in stream:
-                self._shout.send(chunk)
-                self._shout.sync()
-
-                # play the next source immediately
                 if self.skip_requested.is_set():
-                    logger.debug("Skip was requested.")
+                    logger.info("EVENT: Skip")
                     self.skip_requested.clear()
                     break
 
-                # clear the queue
+                if self.clear_requested.is_set():
-                if self.load_requested.is_set():
+                    logger.info("EVENT: Clear")
-                    logger.debug("Load was requested.")
                     self.clear_queue()
-                    self.load_requested.clear()
+                    self.clear_requested.clear()
                     break
 
-                # Stop streaming and clear the queue
                 if self.stop_requested.is_set():
-                    logger.debug("Stop was requested.")
+                    logger.info("EVENT: Stop")
-                    self.clear_queue()
-                    self.stop_requested.clear()
                     break
+
+                if self.start_requested.is_set():
+                    self.start_requested.clear()
+                    break
+
+                if self.shutdown_requested.is_set():
+                    logger.info("EVENT: Shutdown")
+                    break
+
+                logger.debug(f"{title}: {len(chunk)} bytes")
+                self._out.send(chunk)
+                self._out.sync()
+
+
+@dataclass
+class debugServer:
+    name: str = "Croaker Debugger"
+    url: str = None
+    mount: str = None
+    host: str = None
+    port: str = None
+    password: str = None
+    format: str = None
+
+    _output_file: Path = Path("/dev/null")  # Path("./croaker.stream.output.mp3")
+    _filehandle = None
+
+    def open(self):
+        self._filehandle = self._output_file.open("wb")
+
+    def close(self):
+        if self._filehandle:
+            self._filehandle.close()
+            self._filehandle = None
+
+    def set_metadata(self, metadata: dict):
+        logger.info(f"debugServer: {metadata = }")
+
+    def send(self, chunk: bytes):
+        self._filehandle.write(chunk)
+
+    def sync(self):
+        self._filehandle.flush()

src/croaker/transcoder.py

@@ -1,8 +1,8 @@
+import io
 import logging
 import os
 import subprocess
 from dataclasses import dataclass
-from io import BufferedReader
 from pathlib import Path
 
 import ffmpeg
@@ -26,11 +26,23 @@ class FrameAlignedStream:
             ...
     """
 
-    source: BufferedReader
+    source_file: Path
     chunk_size: int = 1024
     bit_rate: int = 192000
     sample_rate: int = 44100
 
+    _transcoder: subprocess.Popen = None
+    _buffer: io.BufferedReader = None
+
+    @property
+    def source(self):
+        if self._buffer:
+            return self._buffer
+        if self._transcoder:
+            return self._transcoder.stdout
+        logger.info("Source is empty")
+        return None
+
     @property
     def frames(self):
         while True:
@@ -86,6 +98,8 @@ class FrameAlignedStream:
         Generate approximately chunk_size segments of audio data by iterating over the
         frames, buffering them, and then yielding several as a single bytes object.
         """
+        try:
+            self._start_transcoder()
             buf = b""
             for frame in self.frames:
                 if len(buf) >= self.chunk_size:
@@ -96,16 +110,21 @@ class FrameAlignedStream:
                 buf += frame
             if buf:
                 yield buf
+        finally:
+            self._stop_transcoder()
 
-    @classmethod
+    def _stop_transcoder(self):
-    def from_source(cls, infile: Path, **kwargs):
+        if self._transcoder:
-        """
+            logger.debug(f"Killing {self._transcoder = }")
-        Create a FrameAlignedStream instance by transcoding an audio source on disk.
+            self._transcoder.kill()
-        """
+            self._transcoder = None
+        self._buffer = None
 
+    def _start_transcoder(self):
         args = [] if os.environ.get("DEBUG") else ["-hide_banner", "-loglevel", "quiet"]
-        ffmpeg_args = (
+        self._transcoder = subprocess.Popen(
-            ffmpeg.input(str(infile))
+            (
+                ffmpeg.input(str(self.source_file))
                 .output(
                     "pipe:",
                     map="a",
@@ -115,18 +134,18 @@ class FrameAlignedStream:
                     id3v2_version=0,
                     # force sample and bit rates
                     **{
-                    "b:a": kwargs.get("bit_rate", cls.bit_rate),
+                        "b:a": self.bit_rate,
-                    "ar": kwargs.get("sample_rate", cls.sample_rate),
+                        "ar": self.sample_rate,
                     },
                 )
                 .global_args("-vn", *args)
                 .compile()
+            ),
+            bufsize=self.chunk_size,
+            stdout=subprocess.PIPE,
+            stdin=subprocess.PIPE,
         )
 
         # Force close STDIN to prevent ffmpeg from trying to read from it. silly ffmpeg.
-        proc = subprocess.Popen(
+        self._transcoder.stdin.close()
-            ffmpeg_args, bufsize=kwargs.get("chunk_size", cls.chunk_size), stdout=subprocess.PIPE, stdin=subprocess.PIPE
+        logger.debug(f"Spawned ffmpeg (PID {self._transcoder.pid}): {' '.join(self._transcoder.args)}")
-        )
-        proc.stdin.close()
-        logger.debug(f"Spawned ffmpeg (PID {proc.pid}) with args {ffmpeg_args = }")
-        return cls(proc.stdout, **kwargs)

test/conftest.py

@@ -1,3 +1,4 @@
+import logging
 from pathlib import Path
 
 import pytest
@@ -14,3 +15,7 @@ def mock_env(monkeypatch):
     monkeypatch.setenv("ICECAST_PORT", "6523")
     monkeypatch.setenv("ICECAST_PASSWORD", "password")
     monkeypatch.setenv("DEBUG", "1")
+
+    logging.basicConfig(format="%(asctime)s - %(name)s - %(levelname)s - %(message)s", level=logging.DEBUG)
+    # logging.getLogger('transcoder').setLevel(logging.INFO)
+    # logging.getLogger('root').setLevel(logging.INFO)

test/test_streamer.py

@@ -1,7 +1,7 @@
 import io
-import queue
 import threading
 from pathlib import Path
+from time import sleep
 from unittest.mock import MagicMock
 
 import pytest
@@ -10,13 +10,10 @@ import shout
 from croaker import playlist, streamer
 
 
-def get_stream_output(stream):
-    return stream.read()
-
-
 @pytest.fixture(scope="session")
 def silence_bytes():
-    return (Path(streamer.__file__).parent / "silence.mp3").read_bytes()
+    # return (Path(streamer.__file__).parent / "silence.mp3").read_bytes()
+    return (Path(__file__).parent / "fixtures" / "transcoded_silence.mp3").read_bytes()
 
 
 @pytest.fixture
@@ -27,6 +24,7 @@ def output_stream():
 @pytest.fixture
 def mock_shout(output_stream, monkeypatch):
     def handle_send(buf):
+        print(f"buffering {len(buf)} bytes to output_stream.")
         output_stream.write(buf)
 
     mm = MagicMock(spec=shout.Shout, **{"return_value.send.side_effect": handle_send})
@@ -35,98 +33,60 @@ def mock_shout(output_stream, monkeypatch):
 
 
 @pytest.fixture
-def input_queue():
+def audio_streamer(monkeypatch, mock_shout):
-    return queue.Queue()
+    return streamer.AudioStreamer()
 
 
 @pytest.fixture
-def skip_event():
+def thread(audio_streamer):
-    return threading.Event()
+    thread = threading.Thread(target=audio_streamer.run)
+    thread.daemon = True
+    yield thread
+    audio_streamer.shutdown_requested.set()
+    thread.join()
 
 
-@pytest.fixture
+def wait_for(condition, timeout=2.0):
-def stop_event():
+    elapsed = 0.0
-    return threading.Event()
+    while not condition() and elapsed < 2.0:
+        elapsed += 0.01
+        sleep(0.01)
+    return elapsed <= timeout
 
 
-@pytest.fixture
+def wait_for_not(condition, timeout=2.0):
-def load_event():
+    return wait_for(lambda: not condition(), timeout=timeout)
-    return threading.Event()
 
 
-@pytest.fixture
+def test_streamer_clear(audio_streamer, thread):
-def audio_streamer(mock_shout, input_queue, skip_event, stop_event, load_event):
+    # enqueue some tracks
-    return streamer.AudioStreamer(input_queue, skip_event, stop_event, load_event)
-
-
-def test_streamer_stop(audio_streamer, stop_event, output_stream):
-    stop_event.set()
-    audio_streamer.stream_queued_audio()
-    assert not stop_event.is_set()
-
-
-def test_streamer_skip(audio_streamer, skip_event, output_stream):
-    skip_event.set()
-    audio_streamer.stream_queued_audio()
-    assert not skip_event.is_set()
-
-
-def test_streamer_load(audio_streamer, load_event, output_stream):
-    load_event.set()
-    audio_streamer.stream_queued_audio()
-    assert not load_event.is_set()
-
-
-def test_clear_queue(audio_streamer, input_queue):
     pl = playlist.Playlist(name="test_playlist")
     for track in pl.tracks:
-        input_queue.put(bytes(track))
+        audio_streamer.queue.put(bytes(track))
-    assert input_queue.not_empty
+    assert not audio_streamer.queue.empty()
-    audio_streamer.clear_queue()
+
-    assert input_queue.empty
+    # start the server and send it a clear request
+    thread.start()
+    audio_streamer.clear_requested.set()
+    assert wait_for(audio_streamer.queue.empty)
+    assert wait_for_not(audio_streamer.clear_requested.is_set)
 
 
-@pytest.mark.skip
+def test_streamer_shutdown(audio_streamer, thread):
-def test_streamer_defaults_to_silence(audio_streamer, input_queue, output_stream, silence_bytes):
+    thread.start()
-    audio_streamer.stream_queued_audio()
+    audio_streamer.shutdown_requested.set()
-    track = playlist.Playlist(name="test_playlist").tracks[0]
+    assert wait_for_not(audio_streamer.shutdown_requested.is_set)
-    input_queue.put(bytes(track))
-    audio_streamer.stream_queued_audio()
-    audio_streamer.stream_queued_audio()
-    assert get_stream_output(output_stream) == silence_bytes + track.read_bytes() + silence_bytes
 
 
-@pytest.mark.skip
+def test_streamer_skip(audio_streamer, thread):
-def test_streamer_plays_silence_on_error(monkeypatch, audio_streamer, input_queue, output_stream, silence_bytes):
+    thread.start()
-    monkeypatch.setattr(audio_streamer.queue, "get", MagicMock(side_effect=Exception))
+    audio_streamer.skip_requested.set()
-    track = playlist.Playlist(name="test_playlist").tracks[0]
+    assert wait_for_not(audio_streamer.skip_requested.is_set)
-    input_queue.put(bytes(track))
-    audio_streamer.stream_queued_audio()
-    assert get_stream_output(output_stream) == silence_bytes
 
 
-@pytest.mark.skip
+def test_streamer_defaults_to_silence(audio_streamer, thread, output_stream, silence_bytes):
-def test_streamer_plays_from_queue(audio_streamer, input_queue, output_stream):
+    thread.start()
-    pl = playlist.Playlist(name="test_playlist")
+    thread.join(timeout=1)
-    expected = b""
+    output_stream.seek(0, 0)
-    for track in pl.tracks:
+    out = output_stream.read()
-        input_queue.put(bytes(track))
+    assert silence_bytes in out
-        expected += track.read_bytes()
-    while not input_queue.empty():
-        audio_streamer.stream_queued_audio()
-    assert get_stream_output(output_stream) == expected
-
-
-def test_streamer_handles_stop_interrupt(audio_streamer, output_stream, stop_event):
-    stop_event.set()
-    audio_streamer.stream_queued_audio()
-    assert get_stream_output(output_stream) == b""
-
-
-def test_streamer_handles_load_interrupt(audio_streamer, input_queue, output_stream, load_event):
-    pl = playlist.Playlist(name="test_playlist")
-    input_queue.put(bytes(pl.tracks[0]))
-    load_event.set()
-    audio_streamer.stream_queued_audio()
-    assert get_stream_output(output_stream) == b""
-    assert input_queue.empty