gestion de l'audio

Author: Samy Laumonier

celery/README.md

@@ -0,0 +1,5 @@
+*.avi
+*.mp4
+*.wav
+*.mp3
+*.txt

celery/core/__init__.py

@@ -0,0 +1,17 @@
+# Comment tester ?
+
+```
+cd tests
+
+# Lancement du core et des workers
+./test.sh start
+
+# Lancement de la conversion d'une vidéo
+python ./test-video.py
+
+# Lancement de la conversion d'un fichier audio
+python ./test-audio.py
+```
+
+La vidéo `tests/resources/input.avi` sera convertie, le résultat sera disponible dans le dossier `tests/.tmp`.
+Le fichier audio `tests/resources/input.wav` sera converti, le résultat sera disponible dans le dossier `tests/.tmp`.

celery/worker/__init__.py

@@ -7,15 +7,13 @@
 from celery import Celery
 from celery.result import ResultSet
 
-from worker.worker import process_video_part
+from worker.worker import process_part
 
 app = Celery('core', backend='amqp', broker='amqp://')
-WORKERS = 10
 
 
-# Video and audio
 def get_total_workers():
-    return WORKERS
+    return 10
 
 
 def get_file_length(input_path):
@@ -37,26 +35,25 @@ def get_file_length(input_path):
         raise Exception("File length not found {}".format(input_path))
 
 
-# Video
-@app.task(name='core.process_video')
-def process_video(input_path, output_format):
-    video_length = get_file_length(input_path)
-    part_length = int(video_length / WORKERS)
+@app.task(name='core.process_file')
+def process_file(is_video, input_path, output_format):
+    file_length = get_file_length(input_path)
+    part_length = int(file_length / get_total_workers())
 
-    output_part_paths = convert_video(input_path, output_format, video_length, part_length)
-    output_path = concatenate_video_parts(input_path, output_format, output_part_paths)
+    output_part_paths = convert_file(is_video, input_path, output_format, file_length, part_length)
+    output_path = concatenate_parts(input_path, output_format, output_part_paths)
 
     return output_path
 
 
-def convert_video(input_path, output_format, video_length, part_length):
+def convert_file(is_video, input_path, output_format, video_length, part_length):
     rs = ResultSet([])
 
-    for i in range(WORKERS):
+    for i in range(get_total_workers()):
         start_at = i * part_length
-        stop_at = start_at + part_length if i != WORKERS - 1 else video_length
+        stop_at = start_at + part_length if i != get_total_workers() - 1 else video_length
         print("worker {} will process from {}s to {}s".format(i + 1, start_at, stop_at))
-        rs.add(process_video_part.delay(input_path, output_format, start_at, stop_at))
+        rs.add(process_part.delay(is_video, input_path, output_format, start_at, stop_at))
 
     return rs.get()
 
@@ -73,7 +70,7 @@ def create_parts_list(input_name, output_part_paths):
     return parts_list_path
 
 
-def concatenate_video_parts(input_path, output_format, output_part_paths):
+def concatenate_parts(input_path, output_format, output_part_paths):
     base_name = os.path.basename(input_path)
     input_name = os.path.splitext(base_name)[0]
 
@@ -90,15 +87,3 @@ def concatenate_video_parts(input_path, output_format, output_part_paths):
         os.remove(output_part_path)
 
     return output_path
-
-
-# Audio
-@app.task(name='core.process_audio')
-def process_audio(input_path, output_format):
-    audio_length = get_file_length(input_path)
-    part_length = int(audio_length / WORKERS)
-
-    output_part_paths = convert_audio(input_path, output_format, audio_length, part_length)
-    output_path = concatenate_audio_parts(input_path, output_format, output_part_paths)
-
-    return output_path

tests/.tmp/.gitignore

@@ -4,7 +4,7 @@
 
 
 start_time = time.time()
-output_path = core.process_audio("resources/input.wav", "mp3")
+output_path = core.process_file(False, "resources/input.wav", "mp3")
 end_time = time.time()
 elapsed = end_time - start_time
 print("file generated in {:4.3f}s, path: {}".format(elapsed, output_path))