Generalise slicer

src/interviewkit/slicer.py

@@ -18,102 +18,90 @@
     data/sampled-70m40s-80m40s-MartineBarrat_FINAL.mp3
     data/sampled-2m-4m-MartineBarrat_FINAL.mp3
 
-Note: 
-The Script currently doesn't support input in hours format, so hours will need to converted to mins:secs format.
-Example: If we want audio from 1hr 20mins 30 secs to 1hr 30mins 40secs, it can be done as shown below:
-1hr 20mins 30 secs = 80mins 30secs
-1hr 30mins 40sec = 90mins 40secs
-
 python interviewkit/slicer.py data/Martine+Barrat_FINAL.mp3 80:30 90:40
 
 """
 import sys
 from pathlib import Path
 import shutil
+from datetime import timedelta, datetime
+from dataclasses import dataclass
+import pydub
 
-try:
-    import pydub
-except ImportError:
-    print("Please install pydub: pip install pydub")
-    exit(1)
 
 if shutil.which("ffmpeg") is None:
     print("Please install ffmpeg: https://ffmpeg.org/download.html")
     print("  On mac you can: brew install ffmpeg")
     exit(1)
 
-def convert_audio_time_to_msec(audio_time_split_list):
-    """ Converting mins and secs to msecs for pydub computation """
-
-    if(audio_time_split_list):
-        if(len(audio_time_split_list) == 1):
-            return int(audio_time_split_list[0]) * 60 * 1000
-        elif(len(audio_time_split_list) == 2):
-            return int(audio_time_split_list[0]) * 60 * 1000 + int(audio_time_split_list[1]) * 1000
-        else:
-            print("Error! Audio slice input params invalid. Audio slice supports start/end time in mins or mins:secs format. Please try again with correct input times.")
-            print("Error inside convert_audio_time_to_msec(audio_time_split_list) funtion.")
-            exit(1)
-    else:
-        print("Error! Audio slice input params invalid. Please try again with correct parameters.")
-        print("Error inside convert_audio_time_to_msec(audio_time_split_list) funtion.")
-        exit(1)
-
-def export_filename(audio_time_list):
-    """ Filename for exported file """
-
-    if audio_time_list and len(audio_time_list) == 2:
-        return f"{audio_time_list[0]}m{audio_time_list[1]}s"
-    elif audio_time_list and len(audio_time_list) == 1:
-        return f"{audio_time_list[0]}m"
-    else:
-        print("Error! Audio slice input params invalid. Please try again with correct parameters.")
-        print("Error inside export_filename(audio_time_list) funtion.")
-        exit(1)
-
-
-def audio_slicing(path, audio_slice_start_time, audio_slice_end_time):
-    """ It reads the original audio and uses start and end input time params to generate sliced audio. """
-
-    print("Sampling {} from {} to {}".format(path, audio_slice_start_time, audio_slice_end_time))
 
-    # Reading original audio file
-    audio = pydub.AudioSegment.from_file(path)
-    original_audio_size_ms = audio.duration_seconds * 1000
+EXPECTED_TIME_FORMAT = "%H:%M:%S"
 
-    # Fetching mins and secs from audio input
-    audio_start_time_list = audio_slice_start_time.split(":")
-    audio_end_time_list = audio_slice_end_time.split(":")
+@dataclass
+class SlicerInput:
+    path: Path
+    start_time: timedelta
+    end_time: timedelta
 
-    # Converting audio start and end times in msecs
-    audio_start_time = convert_audio_time_to_msec(audio_start_time_list)
-    audio_end_time = convert_audio_time_to_msec(audio_end_time_list)
+def convert_time_input_to_time_delta(time_input: str):
+
+    time_info_values = zip(reversed(time_input.split(":")), reversed(EXPECTED_TIME_FORMAT.split(":")))
+
+    full_time_value, time_info = zip(*time_info_values)
+
+    time =  datetime.strptime(":".join(full_time_value), ":".join(time_info))
+
+    return timedelta(hours=time.hour, minutes=time.minute, seconds=time.second)
+
+
+def parse_input(argv: list[str]):
+    if len(argv) != 4:
+        raise ValueError("Usage: python3 slicer.py <filepath> <audio start time in minutes> <audio end time in minutes>")
 
-    # Check if audio start and end times are within original audio size limits
-    if(audio_start_time > original_audio_size_ms or audio_end_time > original_audio_size_ms):
-        print("Error! Audio slice input params cannot be greater than original audio size. Please try again with correct parameters.")
-        exit(1)
+    path = Path(argv[1])
+
+    # Converting audio start and end times in msecs
+    audio_start_time = convert_time_input_to_time_delta(argv[2])
+    audio_end_time = convert_time_input_to_time_delta(argv[3])
+
+    return SlicerInput(path, audio_start_time, audio_end_time)
+
+def create_output_path(input: SlicerInput):
+    start_minutes, start_seconds = divmod(input.start_time.seconds, 60)
+    end_minutes, end_seconds = divmod(input.end_time.seconds, 60)
+    start_time_output = f'{start_minutes}m{start_seconds}s'
+    end_time_output = f'{end_minutes}m{end_seconds}s'
+    new_filepath =   input.path.parent / f"sampled-{start_time_output}-{end_time_output}-{input.path.name}"
+    return new_filepath
 
-    # Audio slicing process
-    audio = audio[audio_start_time:audio_end_time]
+def slice_audio(audio: pydub.AudioSegment, start_time: timedelta , end_time: timedelta):
+
+
+    if(start_time.seconds > audio.duration_seconds or end_time.seconds > audio.duration_seconds):
+        raise ValueError("slice not within audio length")
+
+    start_time_ms = int(start_time.total_seconds() * 1000)
+    end_time_ms = int(end_time.total_seconds() * 1000)
+
+    return audio[start_time_ms:end_time_ms]
 
-    # Filename for exported file
-    audio_start_time_name = export_filename(audio_start_time_list)
-    audio_end_time_name = export_filename(audio_end_time_list)
-    new_filename =  f"{path.parent}/sampled-{audio_start_time_name}-{audio_end_time_name}-{path.name}"
-    audio.export(new_filename, format="mp3")
-    print("Created new file: ", new_filename)
 
 def main():
-    if len(sys.argv) != 4:
-        print("Usage: python3 slicer.py <filepath> <audio start time in minutes> <audio end time in minutes>")
-        return
 
-    path = Path(sys.argv[1])
-    audio_slice_start_time = (sys.argv[2])
-    audio_slice_end_time = (sys.argv[3])
+    argv = parse_input(sys.argv)
+
+    audio = pydub.AudioSegment.from_file(argv.path)
+
+    try:
+       sliced_audio =  slice_audio(audio, argv.start_time, argv.end_time)
+    except ValueError:
+        raise ValueError("Error! Audio slice input params cannot be greater than original audio size. Please try again with correct parameters.")
+
+    output_file_path = create_output_path(argv)
+    sliced_audio.export(output_file_path, "mp3")
+
+    print(f'Created new file: {output_file_path}')
 
-    audio_slicing(path, audio_slice_start_time, audio_slice_end_time)
 
 if __name__ == '__main__':
     main()