Some useful updates

README.md

@@ -44,7 +44,7 @@ which simplifies installing python applications into isolated virtualenvs:
 
     $ pipx install git+https://github.com/MawKKe/audiobook-split-ffmpeg
 
-afterwards, the `audiobook-split-ffmpeg` command should be available via your `PATH`. 
+afterwards, the `audiobook-split-ffmpeg` command should be available via your `PATH`.
 
 Next, see Usage below.
 
@@ -72,6 +72,14 @@ the concurrency might not increase the throughput. (We specifically instruct `ff
 re-encoding, so most of the processing work consists of copying the existing encoded audio data from the
 input file to the output file(s) - this kind of processing is more I/O bounded than CPU-bounded).
 
+You may specify the desired output file format.
+For example, if you need to transcode a video file into a folder containing audio files.
+
+    $ audiobook-split-ffmpeg --infile /path/to/video.mp4 --out-ext mp3
+
+This command will create the directory containing audio files:
+    `/path/to/video/audio_1.mp3`, `/path/to/video/audio_2.mp3`, etc.
+
 # Dependencies
 
 This application has no 3rd party library dependencies, as everything is
@@ -121,4 +129,3 @@ See file `LICENSE` for more information.
 This project is hosted at https://github.com/MawKKe/audiobook-split-ffmpeg
 
 You are welcome to leave bug reports, fixes and feature requests. Thanks!
-

pyproject.toml

@@ -31,6 +31,7 @@ Issues = 'https://github.com/MawKKe/audiobook-split-ffmpeg/issues'
 
 [project.scripts]
 audiobook-split-ffmpeg = 'audiobook_split_ffmpeg.cli:main'
+audiobook-split-ffmpeg-mp3 = 'audiobook_split_ffmpeg.cli:main_mp3_mode'
 
 [tool.pytest.ini_options]
 pythonpath = [

src/audiobook_split_ffmpeg/__init__.py

@@ -52,4 +52,4 @@
     'compute_workitems',
 ]
 __author__ = 'Markus Holmström (MawKKe) <markus@mawkke.fi>'
-__version__ = '0.2.0'
+__version__ = '0.2.5'

src/audiobook_split_ffmpeg/cli.py

@@ -16,6 +16,7 @@
 CLI application implementation for audiobook-split-ffmpeg
 """
 
+import os
 import sys
 import shlex
 import argparse
@@ -28,7 +29,7 @@
 from .workers import process_workitems
 
 
-def parse_args(argv: t.List[str]) -> argparse.Namespace:
+def parse_args(argv: t.List[str], mp3_default: bool = False) -> argparse.Namespace:
     """
     Parse argv into argparse.Namespace
 
@@ -37,22 +38,40 @@ def parse_args(argv: t.List[str]) -> argparse.Namespace:
     argv
         a list of strings, usually the value of sys.argv
 
+    mp3_default
+        set default value of out-ext to `mp3`
+
     WARNING:
         If argv is malformed, the process will exit. Avoid using this function in tests.
     """
     parser = argparse.ArgumentParser(
         description='Split audiobook chapters using ffmpeg', epilog=f'version {__version__}'
     )
-    parser.add_argument(
+
+    infile_excl = parser.add_mutually_exclusive_group(required=True)
+    infile_excl.add_argument(
+        'infile_positional',
+        nargs='*',
+        help='Input file. Chapter information must be present in file metadata',
+    )
+    infile_excl.add_argument(
         '--infile',
-        required=True,
+        nargs='*',
         help='Input file. Chapter information must be present in file metadata',
     )
+
     parser.add_argument(
         '--outdir',
-        required=True,
+        required=False,
+        default=None,
         help='Output directory. Created if does not exist yet.',
     )
+    parser.add_argument(
+        '--out-ext',
+        required=False,
+        default=("mp3" if mp3_default else None),
+        help='New file extension. If conversion to another format is required.',
+    )
     parser.add_argument(
         '--concurrency',
         required=False,
@@ -95,6 +114,7 @@ def parse_args(argv: t.List[str]) -> argparse.Namespace:
     )
 
     args = parser.parse_args(argv[1:])
+    args.infile = (args.infile_positional or args.infile)
 
     return args
 
@@ -113,32 +133,54 @@ def _main(args: argparse.Namespace) -> int:
     if args.verbose:
         print('args:', args)
 
-    work_items = list(
-        compute_workitems(
-            args.infile,
-            args.outdir,
-            enumerate_files=args.enumerate_files,
-            use_title_in_filenames=args.use_title,
+    def parse_path(*pathes):
+        for _infile in map(os.path.abspath, pathes):
+            if os.path.isfile(_infile):
+                yield _infile
+            elif os.path.isdir(_infile):
+                yield from parse_path(
+                    *map(lambda p: os.path.join(_infile, p), os.listdir(_infile))
+                )
+            else:
+                raise RuntimeError("No path \"{0}\" was found".format(_infile))
+
+    for infile in parse_path(*args.infile):
+
+        outdir = args.outdir
+        if outdir is None:
+            outdir, _ = os.path.splitext(infile)
+
+        work_items = list(
+            compute_workitems(
+                infile,
+                outdir,
+                enumerate_files=args.enumerate_files,
+                use_title_in_filenames=args.use_title,
+                out_ext=args.out_ext
+            )
         )
-    )
-    if args.verbose:
-        print('Found: {0} chapters to be processed'.format(len(work_items)))
-
-    if args.dry_run:
-        print('# NOTE: dry-run requested')
-        print(shlex.join(['mkdir', '-p', args.outdir]))
-        commands = (workitem_to_ffmpeg_cmd(wi) for wi in work_items)
-        escaped_cmds = (shlex.join(cmd) for cmd in commands)
-        for cmd in escaped_cmds:
-            print(cmd)
-        return 0
-
-    return process_workitems(
-        work_items,
-        args.outdir,
-        args.concurrency,
-        args.verbose,
-    )
+        if args.verbose:
+            print('Found: {0} chapters to be processed'.format(len(work_items)))
+
+        if args.dry_run:
+            print('# NOTE: dry-run requested')
+            print(shlex.join(['mkdir', '-p', outdir]))
+            commands = (workitem_to_ffmpeg_cmd(wi) for wi in work_items)
+            escaped_cmds = (shlex.join(cmd) for cmd in commands)
+            for cmd in escaped_cmds:
+                print(cmd)
+            return 0
+
+        return_code = process_workitems(
+            work_items,
+            outdir,
+            args.concurrency,
+            args.verbose,
+        )
+        if return_code != 0:
+            return return_code
+
+    return 0
 
 
 def main() -> t.NoReturn:
@@ -148,5 +190,9 @@ def main() -> t.NoReturn:
     sys.exit(_main(parse_args(sys.argv)))
 
 
+def main_mp3_mode() -> t.NoReturn:
+    sys.exit(_main(parse_args(sys.argv, True)))
+
+
 if __name__ == '__main__':
     main()

src/audiobook_split_ffmpeg/ffmpeg.py

@@ -77,6 +77,9 @@ def workitem_to_ffmpeg_cmd(w_item: WorkItem) -> t.List[str]:
     # and/or 'stty sane'.  If corruption still occurs, let me know (email is at
     # the top of the file).
 
+    _, in_ext = os.path.splitext(w_item.infile)
+    _, out_ext = os.path.splitext(w_item.outfile)
+
     base_cmd = [
         'ffmpeg',
         '-nostdin',
@@ -86,9 +89,17 @@ def workitem_to_ffmpeg_cmd(w_item: WorkItem) -> t.List[str]:
         'error',
         '-map_chapters',
         '-1',
-        '-vn',
-        '-c',
-        'copy',
+        '-vn'
+    ]
+
+    if in_ext.lower() == out_ext.lower():
+        # Same format. No conversion needed. Copy the stream.
+        base_cmd += [
+            '-c',
+            'copy'
+        ]
+
+    base_cmd += [
         '-ss',
         w_item.start,
         '-to',
@@ -128,6 +139,10 @@ def ffmpeg_split_chapter(w_item: WorkItem) -> t.Dict:
     # cmd is a a flat list of strings
     cmd = workitem_to_ffmpeg_cmd(w_item)
 
+    _dir = os.path.dirname(w_item.outfile)
+    if not os.path.isdir(_dir):
+        os.makedirs(_dir, exist_ok=True)
+
     try:
         proc = sub.run(
             cmd,
@@ -153,7 +168,11 @@ def ffmpeg_split_chapter(w_item: WorkItem) -> t.Dict:
 
 
 def compute_workitems(
-    infile: Path, outdir: Path, enumerate_files: bool = True, use_title_in_filenames: bool = True
+    infile: Path,
+    outdir: t.Optional[Path] = None,
+    enumerate_files: bool = True,
+    use_title_in_filenames: bool = True,
+    out_ext: t.Optional[str] = None
 ) -> t.Iterator[WorkItem]:
     """
     Compute WorkItem's for each chapter to be processed. These WorkItems can be then used
@@ -164,11 +183,15 @@ def compute_workitems(
     infile
         Path to an audio(book) file. Must contain chapter information in its metadata.
     outdir
-        Path to a directory where chapter files will be written. Must exist already.
+        Path to a directory where chapter files will be written.
+        If the value is `None`, a directory named `infile` (without an extension) will be used.
     enumerate_files
         Include chapter numbers in output filenames?
     use_title_in_filenames
         Include chapter titles in output filenames?
+    out_ext
+        A new file extension, in case conversion to another format is required.
+        If `None`, the extension remains unchanged.
     """
 
     in_root, in_ext = os.path.splitext(os.path.basename(infile))
@@ -178,6 +201,13 @@ def compute_workitems(
 
     # Make sure extension has no leading dots.
     in_ext = in_ext[1:] if in_ext.startswith('.') else in_ext
+    if out_ext:
+        out_ext = out_ext[1:] if out_ext.startswith('.') else out_ext
+    else:
+        out_ext = in_ext
+
+    if outdir is None:
+        outdir, _ = os.path.splitext(infile)
 
     # Get chapter metadata
     info = ffprobe_read_chapters(infile)
@@ -210,11 +240,11 @@ def chnum_fmt(n: int) -> str:
         # Otherwise, use the root part of input filename
         title = title_maybe if (use_title_in_filenames and title_maybe) else in_root
 
-        out_base = '{title}.{ext}'.format(title=title, ext=in_ext)
+        out_base = '{title}.{ext}'.format(title=title, ext=out_ext)
 
         # Prepend chapter number if requested
         if enumerate_files:
-            out_base = '{0} - {1}'.format(chnum_fmt(ch_num), out_base)
+            out_base = '{0}. {1}'.format(chnum_fmt(ch_num), out_base)
 
         yield WorkItem(
             infile=infile,