Pin lxml to < 5.2.0

.github/workflows/test.yml

@@ -21,10 +21,10 @@ jobs:
       uses: actions/setup-python@v4
       with:
         python-version: ${{ matrix.python-version }}
-    - uses: FedericoCarboni/setup-ffmpeg@v2
+    - uses: FedericoCarboni/setup-ffmpeg@v3
       id: setup-ffmpeg
       with:
-        token: ${{ secrets.GH_API_TOKEN }}
+        token: ${{ github.token }}
     - name: Install dependencies
       run: |
         python -m pip install --upgrade pip

pyproject.toml

@@ -23,7 +23,7 @@ dependencies = [
     "jinja2",
     "click>=6.0",
     "progressbar2",
-    "lxml>=2.0",
+    "lxml>=2.0,<5.2.0",
     "SQLAlchemy>=1.4,<2",
     "attrs>=22.0",
     "backports.strenum>=1.1.1;python_version<'3.11'",

scripts/bundle_fa_subset.py

@@ -69,7 +69,7 @@ def extract_font_awesome(icon: str, css: str) -> str:
     name = f".fa-{icon}"
 
     m = re.search(
-        name + ":+before {\s+content: " "['\"]+(?P<codepoint>[^'\"]+)",
+        name + r":+before {\s+content: " "['\"]+(?P<codepoint>[^'\"]+)",
         css,
         re.MULTILINE,
     )
@@ -176,9 +176,9 @@ def bad_options(message: str) -> typing.NoReturn:
 ExistingFileOrDir = click.Path(
     dir_okay=True, file_okay=True, exists=True, path_type=pathlib.Path
 )  # type: ignore
-FA_VERSION_TEMPLATE: typing.Final[
-    str
-] = "https://use.fontawesome.com/releases/v{version}/fontawesome-free-{version}-web.zip"
+FA_VERSION_TEMPLATE: typing.Final[str] = (
+    "https://use.fontawesome.com/releases/v{version}/fontawesome-free-{version}-web.zip"
+)
 LATEST_FA_VERSION: typing.Final[str] = "6.2.0"
 
 

src/audio_feeder/_compat.py

@@ -6,6 +6,6 @@
 try:
     from enum import StrEnum  # type: ignore[attr-defined]
 except ImportError:
-    from backports.strenum import StrEnum  # type: ignore[import]
+    from backports.strenum import StrEnum  # type: ignore[import,no-redef]
 
 __all__ = ("Self", "StrEnum")

src/audio_feeder/_db_types.py

@@ -2,6 +2,7 @@
 
 This exists for compatibility while the old database_handler code still exists.
 """
+
 import typing
 
 from ._object_types import ID, SchemaObject
@@ -16,17 +17,12 @@
 
 # pragma: nocover
 class DatabaseHandler(typing.Protocol):
-    def __init__(self, db_loc: PathType):
-        ...
+    def __init__(self, db_loc: PathType): ...
 
-    def save_table(self, table_name: TableName, table_contents: Table) -> None:
-        ...
+    def save_table(self, table_name: TableName, table_contents: Table) -> None: ...
 
-    def load_table(self, table_name: TableName) -> Table:
-        ...
+    def load_table(self, table_name: TableName) -> Table: ...
 
-    def save_database(self, database: Database) -> None:
-        ...
+    def save_database(self, database: Database) -> None: ...
 
-    def load_database(self) -> Database:
-        ...
+    def load_database(self) -> Database: ...

src/audio_feeder/_object_types.py

@@ -3,6 +3,7 @@
 This is partially to allow typing without circular imports and partially to
 allow typing on the generated schema types.
 """
+
 import typing
 
 from ._compat import Self
@@ -15,13 +16,11 @@
 class SchemaObject(typing.Protocol):
     id: ID
 
-    def to_dict(self) -> typing.Mapping[str, typing.Any]:
-        ...
+    def to_dict(self) -> typing.Mapping[str, typing.Any]: ...
 
     def to_dict_sparse(
         self, *, _filter: typing.Callable = ...
-    ) -> typing.Mapping[str, typing.Any]:
-        ...
+    ) -> typing.Mapping[str, typing.Any]: ...
 
     def copy(self: Self) -> Self:
         """Make a new copy of this object."""

src/audio_feeder/cache_utils.py

@@ -1,6 +1,7 @@
 """
 Functions useful for caching.
 """
+
 import functools
 import typing
 

src/audio_feeder/cli.py

@@ -1,6 +1,7 @@
 """
 Command line scripts
 """
+
 import pathlib
 
 import click

src/audio_feeder/config.py

@@ -1,6 +1,7 @@
 """
 Configuration manager - handles the application's global configuration.
 """
+
 import base64
 import functools
 import graphlib
@@ -64,13 +65,11 @@ class TemplateStr(str):
 
 
 @typing.overload
-def _path_converter(s: typing.Union[TemplatePath, TemplateStr]) -> TemplatePath:
-    ...
+def _path_converter(s: typing.Union[TemplatePath, TemplateStr]) -> TemplatePath: ...
 
 
 @typing.overload
-def _path_converter(s: typing.Union[str, Path]) -> Path:
-    ...
+def _path_converter(s: typing.Union[str, Path]) -> Path: ...
 
 
 def _path_converter(s):
@@ -288,25 +287,25 @@ def find_requirements(attr_name: str) -> typing.Sequence[str]:
     @typing.overload
     def _make_single_replacement(
         self, value: typing.Union[str, TemplateStr], replacements: _ReplacementsMapType
-    ) -> str:
-        ...
+    ) -> str: ...
 
     @typing.overload
     def _make_single_replacement(
         self,
         value: typing.Union[Path, TemplatePath],
         replacements: _ReplacementsMapType,
-    ) -> Path:
-        ...
+    ) -> Path: ...
 
     def _make_single_replacement(self, value, replacements):
         if isinstance(value, TemplatePath):
             return Path(
                 *(
                     self._make_single_replacement(
-                        TemplateStr(component)
-                        if component.startswith("{{")
-                        else component,
+                        (
+                            TemplateStr(component)
+                            if component.startswith("{{")
+                            else component
+                        ),
                         replacements=replacements,
                     )
                     for component in value.parts

src/audio_feeder/directory_parser.py

@@ -107,14 +107,14 @@ class AudiobookLoader(BaseAudioLoader):
     )
 
     COVER_PATTERNS: typing.Final[typing.Sequence[re.Pattern]] = [
-        re.compile("$.*\-Cover^", re.IGNORECASE),
+        re.compile(r"$.*-Cover^", re.IGNORECASE),
         re.compile("cover", re.IGNORECASE),
     ]
 
     DIR_NAME_RE = re.compile(
-        "(?P<authors>.+?)(?= \- )"
-        + "(?: \- \[(?P<series_name>.*?) (?P<series_number>\d+)\])? \- "
-        + "(?P<title>.*$)"
+        r"(?P<authors>.+?)(?= - )"
+        + r"(?: - \[(?P<series_name>.*?) (?P<series_number>\d+)\])? - "
+        + r"(?P<title>.*$)"
     )
 
     @classmethod

src/audio_feeder/file_probe.py

@@ -1,4 +1,5 @@
 """Utilities for determining information about the files."""
+
 import copy
 import datetime
 import functools
@@ -66,6 +67,7 @@ class OverdriveMediaMarker:
     @classmethod
     def from_xml(cls, xml: str) -> typing.Sequence[Self]:
         root = lxml.etree.fromstring(xml)
+        assert isinstance(root.tag, str)
         if root.tag != "Markers":
             raise ValueError(
                 f"Unknown media marker tag: {root.tag}. " + "Should be 'Markers'"
@@ -196,6 +198,7 @@ def from_json(cls, json_dict: FFProbeReturnJSON) -> Self:
             if omm_tag in format_info.tags:
                 omm_tags = OverdriveMediaMarker.from_xml(format_info.tags[omm_tag])
                 omm_tags = sorted(omm_tags)
+                assert format_info.duration is not None  # TODO: Handle this gracefully
 
                 for num, start_tag, end_tag in zip(
                     itertools.count(1),

src/audio_feeder/hash_utils.py

@@ -7,14 +7,11 @@
 
 
 class _Hash(typing.Protocol):
-    def update(self, data: bytes) -> None:
-        ...
+    def update(self, data: bytes) -> None: ...
 
-    def hexdigest(self) -> str:
-        ...
+    def hexdigest(self) -> str: ...
 
-    def digest(self) -> bytes:
-        ...
+    def digest(self) -> bytes: ...
 
 
 _HashFunc = typing.Callable[[], _Hash]

src/audio_feeder/html_utils.py

@@ -1,6 +1,7 @@
 """
 Module for handling HTML-related operations
 """
+
 import bisect
 import io
 from html.parser import HTMLParser

src/audio_feeder/m4btools.py

@@ -1,4 +1,5 @@
 """Tools to split or create chaptered m4b files."""
+
 import copy
 import functools
 import logging
@@ -270,7 +271,7 @@ def _merge_subsets(
             "\n".join(_to_file_list_entry(x.path, x.start, x.end) for x in subsets)
         )
 
-        cmd_before = [
+        cmd_before: Sequence[str] = (
             "ffmpeg",
             "-loglevel",
             "error",
@@ -288,9 +289,9 @@ def _merge_subsets(
             "0",
             "-c:a",
             audio_codec,
-        ]
+        )
 
-        cmd_after = []
+        cmd_after: Sequence[str] = ()
 
         logging.debug("file list:\n%s", file_list.read_text())
         _try_video_variations(
@@ -331,7 +332,7 @@ def _extract_subset(
     if subset.end:
         subset_directives.extend(("-to", str(subset.end)))
 
-    cmd_before = [
+    cmd_before: Sequence[str] = [
         "ffmpeg",
         "-loglevel",
         "error",
@@ -348,7 +349,7 @@ def _extract_subset(
         "copy",
     ]
 
-    cmd_after = []
+    cmd_after: Sequence[str] = ()
 
     _try_video_variations(
         cmd_before,
@@ -412,7 +413,9 @@ def chapter_split_jobs(
 
             job_pool.append(new_job)
 
-        if abs(chapter.end_time - file_infos[fpath].format_info.duration) < 0.25:
+        duration = file_infos[fpath].format_info.duration
+        assert duration is not None  # TODO: Handle this gracefully
+        if abs(chapter.end_time - duration) < 0.25:
             end_time = None
         else:
             end_time = chapter.end_time

src/audio_feeder/media_renderer.py

@@ -1,4 +1,5 @@
 """Utilties for handling media rendering for derived feeds."""
+
 import functools
 import hashlib
 import json

src/audio_feeder/object_handler.py

@@ -50,16 +50,16 @@ class BaseObject:
     id: ID
 
     def to_dict(self) -> typing.Mapping[str, typing.Any]:
-        return attrs.asdict(self)
+        return attrs.asdict(self)  # type: ignore[arg-type]
 
     def to_dict_sparse(
         self, *, _filter=_filter_sparse
     ) -> typing.Mapping[str, typing.Any]:
-        return attrs.asdict(self, filter=_filter)
+        return attrs.asdict(self, filter=_filter)  # type: ignore[arg-type]
 
     def copy(self: Self) -> Self:
         """Make a new copy of this object."""
-        return attrs.evolve(self)
+        return attrs.evolve(self)  # type: ignore[misc]
 
 
 # SqlAlchemy has problems with slots ☹

src/audio_feeder/page_generator.py

@@ -1,6 +1,7 @@
 """
 Page generator
 """
+
 import math
 import os
 import string

src/audio_feeder/resources.py

@@ -1,4 +1,5 @@
 """Modules for handling package data resources."""
+
 import hashlib
 import importlib.resources
 import os

src/audio_feeder/rss_feeds.py

@@ -1,6 +1,7 @@
 """
 RSS Feed generators
 """
+
 import datetime
 import functools
 import json
@@ -207,9 +208,10 @@ def _executor() -> futures.Executor:
             file_hash = None
 
         if entry_obj.file_metadata is None or m_rel_path not in entry_obj.file_metadata:
+            # TODO: Use an assertion or fix fl.path possibly being None
             metadata_futures.append(
                 _executor().submit(
-                    lambda fl, fh: (fl, fp.FileInfo.from_file(fl.path), fh),
+                    lambda fl, fh: (fl, fp.FileInfo.from_file(fl.path), fh),  # type: ignore[arg-type]
                     file_loc,
                     file_hash,
                 )

src/audio_feeder/sql_database_handler.py

@@ -11,6 +11,7 @@
 
 import attrs
 import sqlalchemy as sa
+import sqlalchemy.types  # pylint: disable=unused-import
 from sqlalchemy import orm
 
 from . import _object_types as ot

src/audio_feeder/updater.py

@@ -30,7 +30,7 @@
 # This technically violates PEP 484, but it works. See
 # https://github.com/python/mypy/issues/14023 for more details.
 _PBar = typing.TypeVar(
-    "_PBar", bound=typing.Callable[[typing.Iterable[_T]], typing.Iterable[_T]]
+    "_PBar", bound=typing.Callable[[typing.Iterable[_T]], typing.Iterable[_T]]  # type: ignore[valid-type]
 )
 
 

tests/conftest.py

@@ -36,9 +36,9 @@ def change_config_loc(new_config_loc: _OptPath = None) -> typing.Iterator[_OptPa
     class SentinelEnum(enum.Enum):
         Sentinel = enum.auto
 
-    old_config_dir: typing.Union[
-        str, typing.Literal[SentinelEnum.Sentinel]
-    ] = os.environ.get("AF_CONFIG_DIR", SentinelEnum.Sentinel)
+    old_config_dir: typing.Union[str, typing.Literal[SentinelEnum.Sentinel]] = (
+        os.environ.get("AF_CONFIG_DIR", SentinelEnum.Sentinel)
+    )
     try:
         if new_config_loc is None:
             del os.environ["AF_CONFIG_DIR"]

tests/test_updater.py

@@ -1,4 +1,5 @@
 """Tests for audio_feeder.updater"""
+
 import contextlib
 import json
 import pathlib