ffmpeg-normalize 1.29.1__tar.gz → 1.30.0__tar.gz

{ffmpeg-normalize-1.29.1 → ffmpeg_normalize-1.30.0}/CHANGELOG.md

@@ -1,6 +1,36 @@
 # Changelog
 
 
+## v1.30.0 (2024-11-22)
+
+* Change lower-only message to warning.
+
+* Make setup name PEP 625 compliant.
+
+* Docs: add @ahmetsait as a contributor.
+
+* Implement `--lower-only`
+
+* Fix: `--print-stats` only outputs the last stream.
+
+* More robust `loudnorm` output parsing.
+
+* Remove unnecessary conversions.
+
+* Update .editorconfig.
+
+* Remove python 3.8, add python 3.12, 3.13.
+
+* Add README on file size.
+
+
+## v1.29.2 (2024-11-18)
+
+* Fix: show percentage with two decimal digits in progress.
+
+* Chore: add python 12.
+
+
 ## v1.29.1 (2024-10-22)
 
 * Fix: override argparse usage.

{ffmpeg-normalize-1.29.1 → ffmpeg_normalize-1.30.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
-Name: ffmpeg-normalize
-Version: 1.29.1
+Name: ffmpeg_normalize
+Version: 1.30.0
 Summary: Normalize audio via ffmpeg
 Home-page: https://github.com/slhck/ffmpeg-normalize
 Author: Werner Robitza
@@ -15,11 +15,12 @@ Classifier: Topic :: Multimedia :: Sound/Audio :: Conversion
 Classifier: License :: OSI Approved :: MIT License
 Classifier: Natural Language :: English
 Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.8
 Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
-Requires-Python: >=3.8
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Requires-Python: >=3.9
 Description-Content-Type: text/markdown
 License-File: LICENSE
 
@@ -30,7 +31,7 @@ License-File: LICENSE
 ![GitHub Actions Workflow Status](https://img.shields.io/github/actions/workflow/status/slhck/ffmpeg-normalize/python-package.yml)
 
 <!-- ALL-CONTRIBUTORS-BADGE:START - Do not remove or modify this section -->
-[![All Contributors](https://img.shields.io/badge/all_contributors-19-orange.svg?style=flat-square)](#contributors-)
+[![All Contributors](https://img.shields.io/badge/all_contributors-20-orange.svg?style=flat-square)](#contributors-)
 <!-- ALL-CONTRIBUTORS-BADGE:END -->
 
 A utility for batch-normalizing audio using ffmpeg.
@@ -68,6 +69,7 @@ Read on for more info.
   - [Environment Variables](#environment-variables)
 - [API](#api)
 - [FAQ](#faq)
+  - [My output file is too large?](#my-output-file-is-too-large)
   - [What options should I choose for the EBU R128 filter? What is linear and dynamic mode?](#what-options-should-i-choose-for-the-ebu-r128-filter-what-is-linear-and-dynamic-mode)
   - [The program doesn't work because the "loudnorm" filter can't be found](#the-program-doesnt-work-because-the-loudnorm-filter-cant-be-found)
   - [Should I use this to normalize my music collection?](#should-i-use-this-to-normalize-my-music-collection)
@@ -87,7 +89,7 @@ Read on for more info.
 
 ## Requirements
 
-You need Python 3.8 or higher, and ffmpeg.
+You need Python 3.9 or higher, and ffmpeg.
 
 ### ffmpeg
 
@@ -404,6 +406,16 @@ For more information see the [API documentation](https://htmlpreview.github.io/?
 
 ## FAQ
 
+### My output file is too large?
+
+This is because the default output codec is PCM, which is uncompressed. If you want to reduce the file size, you can specify an audio codec with `-c:a` (e.g., `-c:a aac` for ffmpeg's built-in AAC encoder), and optionally a bitrate with `-b:a`.
+
+For example:
+
+```bash
+ffmpeg-normalize input.wav -o output.m4a -c:a aac -b:a 192k
+```
+
 ### What options should I choose for the EBU R128 filter? What is linear and dynamic mode?
 
 EBU R128 is a method for normalizing audio loudness across different tracks or programs. It works by analyzing the audio content and adjusting it to meet specific loudness targets. The main components are:
@@ -547,6 +559,7 @@ If you found this program useful and feel like giving back, feel free to send a
       <td align="center" valign="top" width="14.28%"><a href="https://github.com/psavva"><img src="https://avatars.githubusercontent.com/u/1454758?v=4?s=100" width="100px;" alt="Panayiotis Savva"/><br /><sub><b>Panayiotis Savva</b></sub></a><br /><a href="https://github.com/slhck/ffmpeg-normalize/commits?author=psavva" title="Code">💻</a></td>
       <td align="center" valign="top" width="14.28%"><a href="https://github.com/HighMans"><img src="https://avatars.githubusercontent.com/u/42877729?v=4?s=100" width="100px;" alt="HighMans"/><br /><sub><b>HighMans</b></sub></a><br /><a href="https://github.com/slhck/ffmpeg-normalize/commits?author=HighMans" title="Code">💻</a></td>
       <td align="center" valign="top" width="14.28%"><a href="https://github.com/kanjieater"><img src="https://avatars.githubusercontent.com/u/32607317?v=4?s=100" width="100px;" alt="kanjieater"/><br /><sub><b>kanjieater</b></sub></a><br /><a href="#ideas-kanjieater" title="Ideas, Planning, & Feedback">🤔</a></td>
+      <td align="center" valign="top" width="14.28%"><a href="https://ahmetsait.com/"><img src="https://avatars.githubusercontent.com/u/8372246?v=4?s=100" width="100px;" alt="Ahmet Sait"/><br /><sub><b>Ahmet Sait</b></sub></a><br /><a href="https://github.com/slhck/ffmpeg-normalize/commits?author=ahmetsait" title="Code">💻</a></td>
     </tr>
   </tbody>
   <tfoot>
@@ -594,6 +607,36 @@ SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 # Changelog
 
 
+## v1.30.0 (2024-11-22)
+
+* Change lower-only message to warning.
+
+* Make setup name PEP 625 compliant.
+
+* Docs: add @ahmetsait as a contributor.
+
+* Implement `--lower-only`
+
+* Fix: `--print-stats` only outputs the last stream.
+
+* More robust `loudnorm` output parsing.
+
+* Remove unnecessary conversions.
+
+* Update .editorconfig.
+
+* Remove python 3.8, add python 3.12, 3.13.
+
+* Add README on file size.
+
+
+## v1.29.2 (2024-11-18)
+
+* Fix: show percentage with two decimal digits in progress.
+
+* Chore: add python 12.
+
+
 ## v1.29.1 (2024-10-22)
 
 * Fix: override argparse usage.

{ffmpeg-normalize-1.29.1 → ffmpeg_normalize-1.30.0}/README.md

@@ -5,7 +5,7 @@
 ![GitHub Actions Workflow Status](https://img.shields.io/github/actions/workflow/status/slhck/ffmpeg-normalize/python-package.yml)
 
 <!-- ALL-CONTRIBUTORS-BADGE:START - Do not remove or modify this section -->
-[![All Contributors](https://img.shields.io/badge/all_contributors-19-orange.svg?style=flat-square)](#contributors-)
+[![All Contributors](https://img.shields.io/badge/all_contributors-20-orange.svg?style=flat-square)](#contributors-)
 <!-- ALL-CONTRIBUTORS-BADGE:END -->
 
 A utility for batch-normalizing audio using ffmpeg.
@@ -43,6 +43,7 @@ Read on for more info.
   - [Environment Variables](#environment-variables)
 - [API](#api)
 - [FAQ](#faq)
+  - [My output file is too large?](#my-output-file-is-too-large)
   - [What options should I choose for the EBU R128 filter? What is linear and dynamic mode?](#what-options-should-i-choose-for-the-ebu-r128-filter-what-is-linear-and-dynamic-mode)
   - [The program doesn't work because the "loudnorm" filter can't be found](#the-program-doesnt-work-because-the-loudnorm-filter-cant-be-found)
   - [Should I use this to normalize my music collection?](#should-i-use-this-to-normalize-my-music-collection)
@@ -62,7 +63,7 @@ Read on for more info.
 
 ## Requirements
 
-You need Python 3.8 or higher, and ffmpeg.
+You need Python 3.9 or higher, and ffmpeg.
 
 ### ffmpeg
 
@@ -379,6 +380,16 @@ For more information see the [API documentation](https://htmlpreview.github.io/?
 
 ## FAQ
 
+### My output file is too large?
+
+This is because the default output codec is PCM, which is uncompressed. If you want to reduce the file size, you can specify an audio codec with `-c:a` (e.g., `-c:a aac` for ffmpeg's built-in AAC encoder), and optionally a bitrate with `-b:a`.
+
+For example:
+
+```bash
+ffmpeg-normalize input.wav -o output.m4a -c:a aac -b:a 192k
+```
+
 ### What options should I choose for the EBU R128 filter? What is linear and dynamic mode?
 
 EBU R128 is a method for normalizing audio loudness across different tracks or programs. It works by analyzing the audio content and adjusting it to meet specific loudness targets. The main components are:
@@ -522,6 +533,7 @@ If you found this program useful and feel like giving back, feel free to send a
       <td align="center" valign="top" width="14.28%"><a href="https://github.com/psavva"><img src="https://avatars.githubusercontent.com/u/1454758?v=4?s=100" width="100px;" alt="Panayiotis Savva"/><br /><sub><b>Panayiotis Savva</b></sub></a><br /><a href="https://github.com/slhck/ffmpeg-normalize/commits?author=psavva" title="Code">💻</a></td>
       <td align="center" valign="top" width="14.28%"><a href="https://github.com/HighMans"><img src="https://avatars.githubusercontent.com/u/42877729?v=4?s=100" width="100px;" alt="HighMans"/><br /><sub><b>HighMans</b></sub></a><br /><a href="https://github.com/slhck/ffmpeg-normalize/commits?author=HighMans" title="Code">💻</a></td>
       <td align="center" valign="top" width="14.28%"><a href="https://github.com/kanjieater"><img src="https://avatars.githubusercontent.com/u/32607317?v=4?s=100" width="100px;" alt="kanjieater"/><br /><sub><b>kanjieater</b></sub></a><br /><a href="#ideas-kanjieater" title="Ideas, Planning, & Feedback">🤔</a></td>
+      <td align="center" valign="top" width="14.28%"><a href="https://ahmetsait.com/"><img src="https://avatars.githubusercontent.com/u/8372246?v=4?s=100" width="100px;" alt="Ahmet Sait"/><br /><sub><b>Ahmet Sait</b></sub></a><br /><a href="https://github.com/slhck/ffmpeg-normalize/commits?author=ahmetsait" title="Code">💻</a></td>
     </tr>
   </tbody>
   <tfoot>

{ffmpeg-normalize-1.29.1 → ffmpeg_normalize-1.30.0}/ffmpeg_normalize/__main__.py

@@ -235,6 +235,20 @@ def create_parser() -> argparse.ArgumentParser:
         default=0.0,
     )
 
+    group_ebu.add_argument(
+        "--lower-only",
+        action="store_true",
+        help=textwrap.dedent(
+            """\
+        Whether the audio should not increase in loudness.
+
+        If the measured loudness from the first pass is lower than the target
+        loudness then normalization pass will be skipped for the measured audio
+        source.
+        """
+        ),
+    )
+
     group_ebu.add_argument(
         "--dual-mono",
         action="store_true",
@@ -514,6 +528,7 @@ def main() -> None:
         keep_lra_above_loudness_range_target=cli_args.keep_lra_above_loudness_range_target,
         true_peak=cli_args.true_peak,
         offset=cli_args.offset,
+        lower_only=cli_args.lower_only,
         dual_mono=cli_args.dual_mono,
         dynamic=cli_args.dynamic,
         audio_codec=cli_args.audio_codec,

{ffmpeg-normalize-1.29.1 → ffmpeg_normalize-1.30.0}/ffmpeg_normalize/_ffmpeg_normalize.py

@@ -3,6 +3,8 @@ from __future__ import annotations
 import json
 import logging
 import os
+import sys
+from itertools import chain
 from typing import TYPE_CHECKING, Literal
 
 from tqdm import tqdm
@@ -58,6 +60,7 @@ class FFmpegNormalize:
         keep_lra_above_loudness_range_target (bool, optional): Keep input loudness range above loudness range target. Defaults to False.
         true_peak (float, optional): True peak. Defaults to -2.0.
         offset (float, optional): Offset. Defaults to 0.0.
+        lower_only (bool, optional): Whether the audio should not increase in loudness. Defaults to False.
         dual_mono (bool, optional): Dual mono. Defaults to False.
         dynamic (bool, optional): Dynamic. Defaults to False.
         audio_codec (str, optional): Audio codec. Defaults to "pcm_s16le".
@@ -94,6 +97,7 @@ class FFmpegNormalize:
         keep_lra_above_loudness_range_target: bool = False,
         true_peak: float = -2.0,
         offset: float = 0.0,
+        lower_only: bool = False,
         dual_mono: bool = False,
         dynamic: bool = False,
         audio_codec: str = "pcm_s16le",
@@ -164,6 +168,7 @@ class FFmpegNormalize:
 
         self.true_peak = check_range(true_peak, -9, 0, name="true_peak")
         self.offset = check_range(offset, -99, 99, name="offset")
+        self.lower_only = lower_only
 
         # Ensure library user is passing correct types
         assert isinstance(dual_mono, bool), "dual_mono must be bool"
@@ -254,5 +259,6 @@ class FFmpegNormalize:
 
             _logger.info(f"Normalized file written to {media_file.output_file}")
 
-        if self.print_stats and self.stats:
-            print(json.dumps(self.stats, indent=4))
+        if self.print_stats:
+            json.dump(list(chain.from_iterable(media_file.get_stats() for media_file in self.media_files)), sys.stdout, indent=4)
+            print()

{ffmpeg-normalize-1.29.1 → ffmpeg_normalize-1.30.0}/ffmpeg_normalize/_media_file.py

@@ -6,13 +6,18 @@ import re
 import shlex
 from shutil import move, rmtree
 from tempfile import mkdtemp
-from typing import TYPE_CHECKING, Iterator, Literal, TypedDict
+from typing import TYPE_CHECKING, Iterable, Iterator, Literal, TypedDict
 
 from tqdm import tqdm
 
 from ._cmd_utils import DUR_REGEX, NUL, CommandRunner
 from ._errors import FFmpegNormalizeError
-from ._streams import AudioStream, SubtitleStream, VideoStream
+from ._streams import (
+    AudioStream,
+    LoudnessStatisticsWithMetadata,
+    SubtitleStream,
+    VideoStream,
+)
 
 if TYPE_CHECKING:
     from ffmpeg_normalize import FFmpegNormalize
@@ -24,6 +29,8 @@ _logger = logging.getLogger(__name__)
 AUDIO_ONLY_FORMATS = {"aac", "ast", "flac", "mka", "oga", "ogg", "opus", "wav"}
 ONE_STREAM = {"aac", "ast", "flac", "mp3", "wav"}
 
+TQDM_BAR_FORMAT = "{desc}: {percentage:3.2f}% |{bar}{r_bar}"
+
 
 def _to_ms(**kwargs: str) -> int:
     hour = int(kwargs.get("hour", 0))
@@ -189,7 +196,12 @@ class MediaFile:
 
         # run the second pass as a whole
         if self.ffmpeg_normalize.progress:
-            with tqdm(total=100, position=1, desc="Second Pass") as pbar:
+            with tqdm(
+                total=100,
+                position=1,
+                desc="Second Pass",
+                bar_format=TQDM_BAR_FORMAT,
+            ) as pbar:
                 for progress in self._second_pass():
                     pbar.update(progress - pbar.n)
         else:
@@ -225,6 +237,7 @@ class MediaFile:
                     total=100,
                     position=1,
                     desc=f"Stream {index + 1}/{len(self.streams['audio'].values())}",
+                    bar_format=TQDM_BAR_FORMAT,
                 ) as pbar:
                     for progress in fun():
                         pbar.update(progress - pbar.n)
@@ -232,11 +245,6 @@ class MediaFile:
                 for _ in fun():
                     pass
 
-        # set initial stats (for dry-runs, this is the only thing we need to do)
-        self.ffmpeg_normalize.stats = [
-            audio_stream.get_stats() for audio_stream in self.streams["audio"].values()
-        ]
-
     def _get_audio_filter_cmd(self) -> tuple[str, list[str]]:
         """
         Return the audio filter command and output labels needed.
@@ -248,10 +256,40 @@ class MediaFile:
         output_labels = []
 
         for audio_stream in self.streams["audio"].values():
-            if self.ffmpeg_normalize.normalization_type == "ebu":
-                normalization_filter = audio_stream.get_second_pass_opts_ebu()
+            skip_normalization = False
+            if self.ffmpeg_normalize.lower_only:
+                if self.ffmpeg_normalize.normalization_type == "ebu":
+                    if (
+                        audio_stream.loudness_statistics["ebu_pass1"] is not None
+                        and audio_stream.loudness_statistics["ebu_pass1"]["input_i"]
+                        < self.ffmpeg_normalize.target_level
+                    ):
+                        skip_normalization = True
+                elif self.ffmpeg_normalize.normalization_type == "peak":
+                    if (
+                        audio_stream.loudness_statistics["max"] is not None
+                        and audio_stream.loudness_statistics["max"]
+                        < self.ffmpeg_normalize.target_level
+                    ):
+                        skip_normalization = True
+                elif self.ffmpeg_normalize.normalization_type == "rms":
+                    if (
+                        audio_stream.loudness_statistics["mean"] is not None
+                        and audio_stream.loudness_statistics["mean"]
+                        < self.ffmpeg_normalize.target_level
+                    ):
+                        skip_normalization = True
+
+            if skip_normalization:
+                _logger.warn(
+                    f"Stream {audio_stream.stream_id} had measured input loudness lower than target, skipping normalization."
+                )
+                normalization_filter = "acopy"
             else:
-                normalization_filter = audio_stream.get_second_pass_opts_peakrms()
+                if self.ffmpeg_normalize.normalization_type == "ebu":
+                    normalization_filter = audio_stream.get_second_pass_opts_ebu()
+                else:
+                    normalization_filter = audio_stream.get_second_pass_opts_peakrms()
 
             input_label = f"[0:{audio_stream.stream_id}]"
             output_label = f"[norm{audio_stream.stream_id}]"
@@ -413,16 +451,10 @@ class MediaFile:
         # in the second pass, we do not normalize stream-by-stream, so we set the stats based on the
         # overall output (which includes multiple loudnorm stats)
         if self.ffmpeg_normalize.normalization_type == "ebu":
-            all_stats = AudioStream.prune_and_parse_loudnorm_output(
-                output, num_stats=len(self.streams["audio"])
-            )
-            for idx, audio_stream in enumerate(self.streams["audio"].values()):
-                audio_stream.set_second_pass_stats(all_stats[idx])
-
-        # collect all stats for the final report, again (overwrite the input)
-        self.ffmpeg_normalize.stats = [
-            audio_stream.get_stats() for audio_stream in self.streams["audio"].values()
-        ]
+            all_stats = AudioStream.prune_and_parse_loudnorm_output(output)
+            for stream_id, audio_stream in self.streams["audio"].items():
+                if stream_id in all_stats:
+                    audio_stream.set_second_pass_stats(all_stats[stream_id])
 
         # warn if self.media_file.ffmpeg_normalize.dynamic == False and any of the second pass stats contain "normalization_type" == "dynamic"
         if self.ffmpeg_normalize.dynamic is False:
@@ -438,3 +470,8 @@ class MediaFile:
                     )
 
         _logger.debug("Normalization finished")
+
+    def get_stats(self) -> Iterable[LoudnessStatisticsWithMetadata]:
+        return (
+            audio_stream.get_stats() for audio_stream in self.streams["audio"].values()
+        )

{ffmpeg-normalize-1.29.1 → ffmpeg_normalize-1.30.0}/ffmpeg_normalize/_streams.py

@@ -15,6 +15,7 @@ if TYPE_CHECKING:
 
 _logger = logging.getLogger(__name__)
 
+_loudnorm_pattern = re.compile(r"\[Parsed_loudnorm_(\d+)")
 
 class EbuLoudnessStatistics(TypedDict):
     input_i: float
@@ -320,58 +321,36 @@ class AudioStream(MediaStream):
             f"Loudnorm first pass command output: {CommandRunner.prune_ffmpeg_progress_from_output(output)}"
         )
 
-        self.loudness_statistics["ebu_pass1"] = (
-            AudioStream.prune_and_parse_loudnorm_output(
-                output, num_stats=1
-            )[0]  # only one stream
-        )
+        # only one stream
+        self.loudness_statistics["ebu_pass1"] = next(iter(AudioStream.prune_and_parse_loudnorm_output(output).values()))
 
     @staticmethod
     def prune_and_parse_loudnorm_output(
-        output: str, num_stats: int = 1
-    ) -> List[EbuLoudnessStatistics]:
+        output: str
+    ) -> dict[int, EbuLoudnessStatistics]:
         """
         Prune ffmpeg progress lines from output and parse the loudnorm filter output.
         There may be multiple outputs if multiple streams were processed.
 
         Args:
             output (str): The output from ffmpeg.
-            num_stats (int): The number of loudnorm statistics to parse.
 
         Returns:
             list: The EBU loudness statistics.
         """
         pruned_output = CommandRunner.prune_ffmpeg_progress_from_output(output)
         output_lines = [line.strip() for line in pruned_output.split("\n")]
-
-        ret = []
-        idx = 0
-        while True:
-            _logger.debug(f"Parsing loudnorm stats for stream {idx}")
-            loudnorm_stats = AudioStream._parse_loudnorm_output(
-                output_lines, stream_index=idx
-            )
-            idx += 1
-
-            if loudnorm_stats is None:
-                continue
-            ret.append(loudnorm_stats)
-
-            if len(ret) >= num_stats:
-                break
-
-        return ret
+        return AudioStream._parse_loudnorm_output(output_lines)
 
     @staticmethod
     def _parse_loudnorm_output(
-        output_lines: list[str], stream_index: Optional[int] = None
-    ) -> Optional[EbuLoudnessStatistics]:
+        output_lines: list[str]
+    ) -> dict[int, EbuLoudnessStatistics]:
         """
         Parse the output of a loudnorm filter to get the EBU loudness statistics.
 
         Args:
             output_lines (list[str]): The output lines of the loudnorm filter.
-            stream_index (int): The stream index, optional to filter out the correct stream. If unset, the first stream is used.
 
         Raises:
             FFmpegNormalizeError: When the output could not be parsed.
@@ -379,64 +358,58 @@ class AudioStream(MediaStream):
         Returns:
             EbuLoudnessStatistics: The EBU loudness statistics, if found.
         """
+        result = dict[int, EbuLoudnessStatistics]()
+        stream_index = -1
         loudnorm_start = 0
-        loudnorm_end = 0
         for index, line in enumerate(output_lines):
-            if line.startswith(f"[Parsed_loudnorm_{stream_index}"):
-                loudnorm_start = index + 1
-                continue
-            if loudnorm_start and line.startswith("}"):
-                loudnorm_end = index + 1
-                break
-
-        if not (loudnorm_start and loudnorm_end):
-            if stream_index is not None:
-                # not an error
-                return None
-
-            raise FFmpegNormalizeError(
-                "Could not parse loudnorm stats; no loudnorm-related output found"
-            )
-
-        try:
-            loudnorm_stats = json.loads(
-                "\n".join(output_lines[loudnorm_start:loudnorm_end])
-            )
-
-            _logger.debug(
-                f"Loudnorm stats for stream {stream_index} parsed: {json.dumps(loudnorm_stats)}"
-            )
-
-            for key in [
-                "input_i",
-                "input_tp",
-                "input_lra",
-                "input_thresh",
-                "output_i",
-                "output_tp",
-                "output_lra",
-                "output_thresh",
-                "target_offset",
-                "normalization_type",
-            ]:
-                if key not in loudnorm_stats:
-                    continue
-                if key == "normalization_type":
-                    loudnorm_stats[key] = loudnorm_stats[key].lower()
-                # handle infinite values
-                elif float(loudnorm_stats[key]) == -float("inf"):
-                    loudnorm_stats[key] = -99
-                elif float(loudnorm_stats[key]) == float("inf"):
-                    loudnorm_stats[key] = 0
-                else:
-                    # convert to floats
-                    loudnorm_stats[key] = float(loudnorm_stats[key])
-
-            return cast(EbuLoudnessStatistics, loudnorm_stats)
-        except Exception as e:
-            raise FFmpegNormalizeError(
-                f"Could not parse loudnorm stats; wrong JSON format in string: {e}"
-            )
+            if stream_index < 0:
+                if m := _loudnorm_pattern.match(line):
+                    loudnorm_start = index + 1
+                    stream_index = int(m.group(1))
+            else:
+                if line.startswith("}"):
+                    loudnorm_end = index + 1
+                    loudnorm_data = "\n".join(output_lines[loudnorm_start:loudnorm_end])
+
+                    try:
+                        loudnorm_stats = json.loads(loudnorm_data)
+
+                        _logger.debug(
+                            f"Loudnorm stats for stream {stream_index} parsed: {loudnorm_data}"
+                        )
+
+                        for key in [
+                            "input_i",
+                            "input_tp",
+                            "input_lra",
+                            "input_thresh",
+                            "output_i",
+                            "output_tp",
+                            "output_lra",
+                            "output_thresh",
+                            "target_offset",
+                            "normalization_type",
+                        ]:
+                            if key not in loudnorm_stats:
+                                continue
+                            if key == "normalization_type":
+                                loudnorm_stats[key] = loudnorm_stats[key].lower()
+                            # handle infinite values
+                            elif float(loudnorm_stats[key]) == -float("inf"):
+                                loudnorm_stats[key] = -99
+                            elif float(loudnorm_stats[key]) == float("inf"):
+                                loudnorm_stats[key] = 0
+                            else:
+                                # convert to floats
+                                loudnorm_stats[key] = float(loudnorm_stats[key])
+
+                        result[stream_index] = cast(EbuLoudnessStatistics, loudnorm_stats)
+                        stream_index = -1
+                    except Exception as e:
+                        raise FFmpegNormalizeError(
+                            f"Could not parse loudnorm stats; wrong JSON format in string: {e}"
+                        )
+        return result
 
     def get_second_pass_opts_ebu(self) -> str:
         """
@@ -515,19 +488,19 @@ class AudioStream(MediaStream):
             "lra": self.media_file.ffmpeg_normalize.loudness_range_target,
             "tp": self.media_file.ffmpeg_normalize.true_peak,
             "offset": self._constrain(
-                float(stats["target_offset"]), -99, 99, name="target_offset"
+                stats["target_offset"], -99, 99, name="target_offset"
             ),
             "measured_i": self._constrain(
-                float(stats["input_i"]), -99, 0, name="input_i"
+                stats["input_i"], -99, 0, name="input_i"
             ),
             "measured_lra": self._constrain(
-                float(stats["input_lra"]), 0, 99, name="input_lra"
+                stats["input_lra"], 0, 99, name="input_lra"
             ),
             "measured_tp": self._constrain(
-                float(stats["input_tp"]), -99, 99, name="input_tp"
+                stats["input_tp"], -99, 99, name="input_tp"
             ),
             "measured_thresh": self._constrain(
-                float(stats["input_thresh"]), -99, 0, name="input_thresh"
+                stats["input_thresh"], -99, 0, name="input_thresh"
             ),
             "linear": "false" if self.media_file.ffmpeg_normalize.dynamic else "true",
             "print_format": "json",

ffmpeg_normalize-1.30.0/ffmpeg_normalize/_version.py

@@ -0,0 +1 @@
+__version__ = "1.30.0"

{ffmpeg-normalize-1.29.1 → ffmpeg_normalize-1.30.0}/ffmpeg_normalize.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: ffmpeg-normalize
-Version: 1.29.1
+Version: 1.30.0
 Summary: Normalize audio via ffmpeg
 Home-page: https://github.com/slhck/ffmpeg-normalize
 Author: Werner Robitza
@@ -15,11 +15,12 @@ Classifier: Topic :: Multimedia :: Sound/Audio :: Conversion
 Classifier: License :: OSI Approved :: MIT License
 Classifier: Natural Language :: English
 Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.8
 Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
-Requires-Python: >=3.8
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Requires-Python: >=3.9
 Description-Content-Type: text/markdown
 License-File: LICENSE
 
@@ -30,7 +31,7 @@ License-File: LICENSE
 ![GitHub Actions Workflow Status](https://img.shields.io/github/actions/workflow/status/slhck/ffmpeg-normalize/python-package.yml)
 
 <!-- ALL-CONTRIBUTORS-BADGE:START - Do not remove or modify this section -->
-[![All Contributors](https://img.shields.io/badge/all_contributors-19-orange.svg?style=flat-square)](#contributors-)
+[![All Contributors](https://img.shields.io/badge/all_contributors-20-orange.svg?style=flat-square)](#contributors-)
 <!-- ALL-CONTRIBUTORS-BADGE:END -->
 
 A utility for batch-normalizing audio using ffmpeg.
@@ -68,6 +69,7 @@ Read on for more info.
   - [Environment Variables](#environment-variables)
 - [API](#api)
 - [FAQ](#faq)
+  - [My output file is too large?](#my-output-file-is-too-large)
   - [What options should I choose for the EBU R128 filter? What is linear and dynamic mode?](#what-options-should-i-choose-for-the-ebu-r128-filter-what-is-linear-and-dynamic-mode)
   - [The program doesn't work because the "loudnorm" filter can't be found](#the-program-doesnt-work-because-the-loudnorm-filter-cant-be-found)
   - [Should I use this to normalize my music collection?](#should-i-use-this-to-normalize-my-music-collection)
@@ -87,7 +89,7 @@ Read on for more info.
 
 ## Requirements
 
-You need Python 3.8 or higher, and ffmpeg.
+You need Python 3.9 or higher, and ffmpeg.
 
 ### ffmpeg
 
@@ -404,6 +406,16 @@ For more information see the [API documentation](https://htmlpreview.github.io/?
 
 ## FAQ
 
+### My output file is too large?
+
+This is because the default output codec is PCM, which is uncompressed. If you want to reduce the file size, you can specify an audio codec with `-c:a` (e.g., `-c:a aac` for ffmpeg's built-in AAC encoder), and optionally a bitrate with `-b:a`.
+
+For example:
+
+```bash
+ffmpeg-normalize input.wav -o output.m4a -c:a aac -b:a 192k
+```
+
 ### What options should I choose for the EBU R128 filter? What is linear and dynamic mode?
 
 EBU R128 is a method for normalizing audio loudness across different tracks or programs. It works by analyzing the audio content and adjusting it to meet specific loudness targets. The main components are:
@@ -547,6 +559,7 @@ If you found this program useful and feel like giving back, feel free to send a
       <td align="center" valign="top" width="14.28%"><a href="https://github.com/psavva"><img src="https://avatars.githubusercontent.com/u/1454758?v=4?s=100" width="100px;" alt="Panayiotis Savva"/><br /><sub><b>Panayiotis Savva</b></sub></a><br /><a href="https://github.com/slhck/ffmpeg-normalize/commits?author=psavva" title="Code">💻</a></td>
       <td align="center" valign="top" width="14.28%"><a href="https://github.com/HighMans"><img src="https://avatars.githubusercontent.com/u/42877729?v=4?s=100" width="100px;" alt="HighMans"/><br /><sub><b>HighMans</b></sub></a><br /><a href="https://github.com/slhck/ffmpeg-normalize/commits?author=HighMans" title="Code">💻</a></td>
       <td align="center" valign="top" width="14.28%"><a href="https://github.com/kanjieater"><img src="https://avatars.githubusercontent.com/u/32607317?v=4?s=100" width="100px;" alt="kanjieater"/><br /><sub><b>kanjieater</b></sub></a><br /><a href="#ideas-kanjieater" title="Ideas, Planning, & Feedback">🤔</a></td>
+      <td align="center" valign="top" width="14.28%"><a href="https://ahmetsait.com/"><img src="https://avatars.githubusercontent.com/u/8372246?v=4?s=100" width="100px;" alt="Ahmet Sait"/><br /><sub><b>Ahmet Sait</b></sub></a><br /><a href="https://github.com/slhck/ffmpeg-normalize/commits?author=ahmetsait" title="Code">💻</a></td>
     </tr>
   </tbody>
   <tfoot>
@@ -594,6 +607,36 @@ SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 # Changelog
 
 
+## v1.30.0 (2024-11-22)
+
+* Change lower-only message to warning.
+
+* Make setup name PEP 625 compliant.
+
+* Docs: add @ahmetsait as a contributor.
+
+* Implement `--lower-only`
+
+* Fix: `--print-stats` only outputs the last stream.
+
+* More robust `loudnorm` output parsing.
+
+* Remove unnecessary conversions.
+
+* Update .editorconfig.
+
+* Remove python 3.8, add python 3.12, 3.13.
+
+* Add README on file size.
+
+
+## v1.29.2 (2024-11-18)
+
+* Fix: show percentage with two decimal digits in progress.
+
+* Chore: add python 12.
+
+
 ## v1.29.1 (2024-10-22)
 
 * Fix: override argparse usage.

{ffmpeg-normalize-1.29.1 → ffmpeg_normalize-1.30.0}/setup.py

@@ -19,7 +19,7 @@ with open(path.join(here, "CHANGELOG.md"), encoding="utf8") as f:
     history = f.read()
 
 setup(
-    name="ffmpeg-normalize",
+    name="ffmpeg_normalize",
     version=version,
     description="Normalize audio via ffmpeg",
     long_description=long_description + "\n\n" + history,
@@ -50,12 +50,13 @@ setup(
         "License :: OSI Approved :: MIT License",
         "Natural Language :: English",
         "Programming Language :: Python :: 3",
-        "Programming Language :: Python :: 3.8",
         "Programming Language :: Python :: 3.9",
         "Programming Language :: Python :: 3.10",
         "Programming Language :: Python :: 3.11",
+        "Programming Language :: Python :: 3.12",
+        "Programming Language :: Python :: 3.13",
     ],
-    python_requires=">=3.8",
+    python_requires=">=3.9",
     entry_points={
         "console_scripts": ["ffmpeg-normalize = ffmpeg_normalize.__main__:main"]
     },

ffmpeg-normalize-1.29.1/ffmpeg_normalize/_version.py

@@ -1 +0,0 @@
-__version__ = "1.29.1"