Add control on ffmpeg CPU threads consumption, with a default of 1

CHANGELOG.md

@@ -13,6 +13,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Suuport for Python 3.12, drop Python 3.7 #118
 - Replace "iso-369" iso639-lang by "iso639-lang" library
 - Rework the VideoWebmLow preset for faster encoding and smaller file size (preset has been bumped to version 2)
+- When reencoding a video, ffmpeg now uses only 1 CPU thread by default (new arg to `reencode` allows to override this default value)
 
 ## [3.2.0] - 2023-12-16
 

src/zimscraperlib/video/encoding.py

@@ -6,41 +6,64 @@
 import shutil
 import subprocess
 import tempfile
+from typing import List, Optional
 
 from zimscraperlib import logger
 from zimscraperlib.logging import nicer_args_join
 
 
+def _build_ffmpeg_args(
+    src_path: pathlib.Path,
+    tmp_path: pathlib.Path,
+    ffmpeg_args: List[str],
+    threads: Optional[int],
+) -> List[str]:
+    if threads:
+        if "-threads" in ffmpeg_args:
+            raise AttributeError("Cannot set the number of threads, already set")
+        else:
+            ffmpeg_args += ["-threads", str(threads)]
+    args = [
+        "/usr/bin/env",
+        "ffmpeg",
+        "-y",
+        "-i",
+        f"file:{src_path}",
+        *ffmpeg_args,
+        f"file:{tmp_path}",
+    ]
+    return args
+
+
 def reencode(
     src_path,
     dst_path,
     ffmpeg_args,
     delete_src=False,  # noqa: FBT002
     with_process=False,  # noqa: FBT002
     failsafe=True,  # noqa: FBT002
+    threads: Optional[int] = 1,
 ):
     """Runs ffmpeg with given ffmpeg_args
 
     Arguments -
         src_path - Path to source file
         dst_path - Path to destination file
         ffmpeg_args - A list of ffmpeg arguments
+        threads - Number of encoding threads used by ffmpeg
         delete_src - Delete source file after convertion
         with_process - Optionally return the output from ffmpeg (stderr and stdout)
         failsafe - Run in failsafe mode
     """
 
     with tempfile.TemporaryDirectory() as tmp_dir:
         tmp_path = pathlib.Path(tmp_dir).joinpath(f"video.tmp{dst_path.suffix}")
-        args = [
-            "/usr/bin/env",
-            "ffmpeg",
-            "-y",
-            "-i",
-            f"file:{src_path}",
-            *ffmpeg_args,
-            f"file:{tmp_path}",
-        ]
+        args = _build_ffmpeg_args(
+            src_path=src_path,
+            tmp_path=tmp_path,
+            ffmpeg_args=ffmpeg_args,
+            threads=threads,
+        )
         logger.debug(
             f"Encode {src_path} -> {dst_path} video format = {dst_path.suffix}"
         )

tests/video/test_encoding.py

@@ -0,0 +1,94 @@
+import re
+from pathlib import Path
+from typing import List, Optional
+
+import pytest
+
+from zimscraperlib.video.encoding import _build_ffmpeg_args
+
+
+@pytest.mark.parametrize(
+    "src_path,tmp_path,ffmpeg_args,threads,expected",
+    [
+        (
+            Path("path1/file1.mp4"),
+            Path("path1/fileout.mp4"),
+            [
+                "-codec:v",
+                "libx265",
+            ],
+            None,
+            [
+                "/usr/bin/env",
+                "ffmpeg",
+                "-y",
+                "-i",
+                "file:path1/file1.mp4",
+                "-codec:v",
+                "libx265",
+                "file:path1/fileout.mp4",
+            ],
+        ),
+        (
+            Path("path2/file2.mp4"),
+            Path("path12/tmpfile.mp4"),
+            [
+                "-b:v",
+                "300k",
+            ],
+            1,
+            [
+                "/usr/bin/env",
+                "ffmpeg",
+                "-y",
+                "-i",
+                "file:path2/file2.mp4",
+                "-b:v",
+                "300k",
+                "-threads",
+                "1",
+                "file:path12/tmpfile.mp4",
+            ],
+        ),
+        (
+            Path("path2/file2.mp4"),
+            Path("path12/tmpfile.mp4"),
+            [
+                "-b:v",
+                "300k",
+                "-threads",
+                "1",
+            ],
+            1,
+            None,
+        ),
+    ],
+)
+def test_build_ffmpeg_args(
+    src_path: Path,
+    tmp_path: Path,
+    ffmpeg_args: List[str],
+    threads: Optional[int],
+    expected: Optional[List[str]],
+):
+    if expected:
+        assert (
+            _build_ffmpeg_args(
+                src_path=src_path,
+                tmp_path=tmp_path,
+                ffmpeg_args=ffmpeg_args,
+                threads=threads,
+            )
+            == expected
+        )
+    else:
+        with pytest.raises(
+            AttributeError,
+            match=re.escape("Cannot set the number of threads, already set"),
+        ):
+            _build_ffmpeg_args(
+                src_path=src_path,
+                tmp_path=tmp_path,
+                ffmpeg_args=ffmpeg_args,
+                threads=threads,
+            )