Merge pull request #81 from GitBib/perf/optimize-mkv-loading

perf: optimize MKV file loading performance by reducing redundant mkv…

Author: GitBib

.pre-commit-config.yaml

@@ -6,7 +6,7 @@ default_language_version:
 
 repos:
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v5.0.0
+    rev: v6.0.0
     hooks:
       - id: trailing-whitespace
       - id: end-of-file-fixer
@@ -21,14 +21,14 @@ repos:
       - id: detect-private-key
 
   - repo: https://github.com/astral-sh/uv-pre-commit
-    rev: '0.6.10'
+    rev: '0.8.23'
     hooks:
       - id: uv-lock
       - id: uv-sync
 
   # Run the Ruff linter.
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.11.2
+    rev: v0.13.3
     hooks:
       # Linter
       - id: ruff
@@ -37,7 +37,7 @@ repos:
       - id: ruff-format
 
   - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v1.15.0
+    rev: v1.18.2
     hooks:
       - id: mypy
 

Makefile

@@ -15,7 +15,7 @@ test: $(TEST_FILE) $(TEST_TWO_FILE)
 	@mkvmerge -J $(TEST_FILE)
 	@echo "Running mkvmerge -J $(TEST_TWO_FILE)..."
 	@mkvmerge -J $(TEST_TWO_FILE)
-	pytest --cov=pymkv $(TEST_DIR) --cov-report=xml --junitxml=test-results/junit.xml
+	uv run pytest --cov=pymkv $(TEST_DIR) --cov-report=xml --junitxml=test-results/junit.xml
 
 $(TEST_FILE):
 	@if [ ! -f $(TEST_FILE) ]; then \

pymkv/MKVFile.py

@@ -55,7 +55,6 @@
     checking_file_path,
     get_file_info,
     verify_mkvmerge,
-    verify_supported,
 )
 
 T = TypeVar("T")
@@ -123,10 +122,6 @@ def __init__(
             raise FileNotFoundError(msg)
 
         if file_path is not None:
-            if not verify_supported(file_path, mkvmerge_path=self.mkvmerge_path):
-                msg = f"The file '{file_path}' is not a valid Matroska file or is not supported."
-                raise ValueError(msg)
-            # add file title
             file_path = checking_file_path(file_path)
             try:
                 info_json = get_file_info(
@@ -142,6 +137,11 @@ def __init__(
                     e.cmd,
                     output=error_output,
                 ) from e
+
+            if not info_json["container"]["supported"]:
+                msg = f"The file '{file_path}' is not a valid Matroska file or is not supported."
+                raise ValueError(msg)
+
             if self.title is None and "title" in info_json["container"]["properties"]:
                 self.title = info_json["container"]["properties"]["title"]
 

pymkv/MKVTrack.py

@@ -142,7 +142,7 @@ def __init__(  # noqa: PLR0913
         tag_entries: int = 0,
         compression: bool | None = None,
     ) -> None:
-        from pymkv.TypeTrack import get_track_extension
+        from pymkv.TypeTrack import get_track_extension  # noqa: PLC0415
 
         # track info
         self._track_codec: str | None = None
@@ -215,6 +215,8 @@ def file_path(self, file_path: str) -> None:
 
         This method checks if the provided file path is valid and supported by mkvmerge.
         If the file is valid, it sets the file path and resets the track_id to 0.
+        If existing_info is already set, skip the verify_supported check since the file
+        was already verified.
 
         Args:
             file_path (str): The path to the file containing the track.
@@ -223,7 +225,7 @@ def file_path(self, file_path: str) -> None:
             ValueError: If the file is not a valid Matroska file or is not supported.
         """
         fp = checking_file_path(file_path)
-        if not verify_supported(fp, mkvmerge_path=self.mkvmerge_path):
+        if not self._info_json and not verify_supported(fp, mkvmerge_path=self.mkvmerge_path):
             msg = f"The file '{file_path}' is not a valid Matroska file or is not supported."
             raise ValueError(msg)
         self._file_path = fp

pymkv/Verifications.py

@@ -45,6 +45,7 @@
 import os
 import subprocess as sp
 from collections.abc import Iterable, Sequence
+from functools import cache
 from pathlib import Path
 from re import match
 from typing import Any
@@ -122,6 +123,18 @@ def get_file_info(
     return json.loads(sp.check_output(cmds).decode())  # noqa: S603
 
 
+@cache
+def _verify_mkvmerge_cached(mkvmerge_path_tuple: tuple[str, ...]) -> bool:
+    """Internal cached function to verify mkvmerge availability."""
+    try:
+        mkvmerge_command = list(mkvmerge_path_tuple)
+        mkvmerge_command.append("-V")
+        output = sp.check_output(mkvmerge_command).decode()  # noqa: S603
+    except (sp.CalledProcessError, FileNotFoundError):
+        return False
+    return bool(match("mkvmerge.*", output))
+
+
 def verify_mkvmerge(
     mkvmerge_path: str | os.PathLike | Iterable[str] = "mkvmerge",
 ) -> bool:
@@ -137,13 +150,10 @@ def verify_mkvmerge(
     bool
         True if mkvmerge is available at the specified path, False otherwise.
     """
-    try:
-        mkvmerge_command = list(prepare_mkvtoolnix_path(mkvmerge_path))
-        mkvmerge_command.append("-V")
-        output = sp.check_output(mkvmerge_command).decode()  # noqa: S603
-    except (sp.CalledProcessError, FileNotFoundError):
-        return False
-    return bool(match("mkvmerge.*", output))
+    mkvmerge_command = prepare_mkvtoolnix_path(mkvmerge_path)
+    if isinstance(mkvmerge_command, list):
+        mkvmerge_command = tuple(mkvmerge_command)
+    return _verify_mkvmerge_cached(mkvmerge_command)
 
 
 def verify_matroska(

pyproject.toml

@@ -43,6 +43,7 @@ dev = [
     "mypy<2.0.0,>=1.14.1",
     "ruff<1.0.0,>=0.9.0",
     "sphinx-immaterial<0.14.0,>=0.12.4; python_version >= \"3.10\"",
+    "pyinstrument>=5.1.1",
 ]
 
 [project.urls]

tests/conftest.py

@@ -3,6 +3,9 @@
 from pathlib import Path
 
 import pytest
+from pyinstrument import Profiler
+
+TESTS_ROOT = Path.cwd()
 
 
 @pytest.fixture
@@ -67,3 +70,18 @@ def get_path_test_srt(get_base_path: Path) -> Path:
     srt_path = get_base_path / "test_file.srt"
     create_srt_file_with_random_text(srt_path)
     return srt_path
+
+
+@pytest.fixture(autouse=True)
+def auto_profile(request: pytest.FixtureRequest) -> Generator[None, None, None]:
+    profile_root = TESTS_ROOT / ".profiles"
+    profiler = Profiler()
+    profiler.start()
+
+    yield
+
+    profiler.stop()
+    profile_root.mkdir(exist_ok=True)
+    test_name = request.node.name
+    results_file = profile_root / f"{test_name}.html"
+    profiler.write_html(results_file)

tests/test_open_files.py

@@ -31,15 +31,15 @@ def test_file_not_found() -> None:
 def test_file_not_support() -> None:
     with pytest.raises(
         ValueError,
-        match="The file 'tests/conftest.py' is not a valid Matroska file or is not supported.",
+        match=r"The file 'tests[/\\]conftest\.py' is not a valid Matroska file or is not supported\.",
     ):
         MKVFile("tests/conftest.py")
 
 
 def test_track_not_support() -> None:
     with pytest.raises(
         ValueError,
-        match="The file 'tests/conftest.py' is not a valid Matroska file or is not supported.",
+        match=r"The file 'tests[/\\]conftest\.py' is not a valid Matroska file or is not supported\.",
     ):
         MKVTrack("tests/conftest.py")