Fix error in Popen call of ffmpeg (#327)

Make system calls to ffmpeg are more robust
Allows sounds extension to work on Linux

Author: natir

docs/changelog.rst

@@ -16,6 +16,11 @@ Master
     - Bug fixes
         - Make sure double-quotes are properly tokenized for bot commands
 
+
+- ext.sound
+    - Bug fixes
+        - Make system calls to ffmpeg are more robust (works on windows and linux)
+
 - ext.eventsub
     - Additions
         - Goal subscriptions have been Added
@@ -25,9 +30,11 @@ Master
             - :func:`~twitchio.ext.eventsub.event_eventsub_notification_channel_goal_begin`
             - :func:`~twitchio.ext.eventsub.event_eventsub_notification_channel_goal_progress`
             - :func:`~twitchio.ext.eventsub.event_eventsub_notification_channel_goal_end`
+
     - Bug fixes
         Correct typo in :class:`~twitchio.ext.eventsub.HypeTrainBeginProgressData` attribute :attr:`~twitchio.ext.eventsub.HypeTrainBeginProgressData.expires` 
 
+
 2.4.0
 ======
 - TwitchIO
@@ -74,7 +81,7 @@ Master
     - Bug fixes
         - Add type conversion for variable positional arguments
         - Fixed message content while handling commands in reply messages
-      
+
 - ext.pubsub
     - Bug fixes
         - :class:`~twitchio.ext.pubsub.PubSubModerationAction` now handles missing keys

twitchio/ext/sounds/__init__.py

@@ -43,13 +43,18 @@
 
 AP = TypeVar("AP", bound="AudioPlayer")
 
-has_ffmpeg: bool
+ffmpeg_bin: Optional[str] = None
 try:
-    proc = subprocess.Popen("ffmpeg -version", stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+    proc = subprocess.Popen(["ffmpeg", "-version"], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
 except FileNotFoundError:
-    has_ffmpeg = False
+    try:
+        proc = subprocess.Popen(["ffmpeg.exe", "-version"])
+    except FileNotFoundError:
+        ffmpeg_bin = None
+    else:
+        ffmpeg_bin = "ffmpeg.exe"
 else:
-    has_ffmpeg = True
+    ffmpeg_bin = "ffmpeg"
 
 
 __all__ = ("Sound", "AudioPlayer")
@@ -127,19 +132,24 @@ class Sound:
 
     YTDL = YoutubeDL(YTDLOPTS)
 
-    def __init__(self, source: Optional[Union[str, io.BufferedIOBase]] = None, *, info: Optional[dict] = None):
+    def __init__(
+        self,
+        source: Optional[Union[str, io.BufferedIOBase]] = None,
+        *,
+        info: Optional[dict] = None,
+    ):
         self.title = None
         self.url = None
 
         self.proc = None
 
-        if not has_ffmpeg:
-            raise RuntimeError("ffmpeg is required to create and play Sounds. For more information visit: ...")
+        if ffmpeg_bin is None:
+            raise RuntimeError("ffmpeg is required to create and play Sounds. Check your is present in your Path")
 
         if info:
             self.proc = subprocess.Popen(
                 [
-                    "ffmpeg.exe",
+                    ffmpeg_bin,
                     "-reconnect",
                     "1",
                     "-reconnect_streamed",
@@ -165,7 +175,17 @@ def __init__(self, source: Optional[Union[str, io.BufferedIOBase]] = None, *, in
             self.title = source
 
             self.proc = subprocess.Popen(
-                ["ffmpeg.exe", "-i", source, "-loglevel", "panic", "-vn", "-f", "s16le", "pipe:1"],
+                [
+                    ffmpeg_bin,
+                    "-i",
+                    source,
+                    "-loglevel",
+                    "panic",
+                    "-vn",
+                    "-f",
+                    "s16le",
+                    "pipe:1",
+                ],
                 stdout=subprocess.PIPE,
             )
 
@@ -260,7 +280,9 @@ def _get_devices(self):
                 continue
 
             self._devices[index] = OutputDevice(
-                name=device["name"], index=device["index"], channels=device["maxOutputChannels"]
+                name=device["name"],
+                index=device["index"],
+                channels=device["maxOutputChannels"],
             )
 
     def play(self, sound: Sound, *, replace: bool = False) -> None:
@@ -284,7 +306,11 @@ def _play_run(self, sound: Sound):
 
         device = self._use_device.index if self._use_device else None
         self._stream = self._pa.open(
-            format=pyaudio.paInt16, output=True, channels=sound.channels, rate=sound.rate, output_device_index=device
+            format=pyaudio.paInt16,
+            output=True,
+            channels=sound.channels,
+            rate=sound.rate,
+            output_device_index=device,
         )
 
         bytes_ = sound.proc.stdout.read(4096)