Use objects as codecs instead of modules

.github/dependabot.yml

@@ -9,7 +9,6 @@ updates:
         patterns:
           - "pytest*"
           - "ruff"
-          - "refurb"
 
   - package-ecosystem: "github-actions"
     directory: "/"

alternative_encodings/__init__.py

@@ -2,7 +2,7 @@
 
 
 def register_all() -> None:
-    cp859.register()
-    cp866i.register()
-    romaji.register()
-    viscii.register()
+    cp859.codec.register()
+    cp866i.codec.register()
+    romaji.codec.register()
+    viscii.codec.register()

alternative_encodings/common.py

@@ -1,4 +1,6 @@
 import codecs
+import contextlib
+from abc import ABC, abstractmethod
 from typing import TYPE_CHECKING, Union
 
 if TYPE_CHECKING:
@@ -44,3 +46,23 @@ class StreamReader(codec, codecs.StreamReader):
         pass
 
     return StreamReader
+
+
+class CodecBase(ABC):
+    @abstractmethod
+    def get_codec_info(self) -> codecs.CodecInfo:
+        raise NotImplementedError
+
+    def search_function(self, encoding: str) -> codecs.CodecInfo | None:
+        codec_info = self.get_codec_info()
+        if codec_info.name == encoding:
+            return codec_info
+
+        return None
+
+    def register(self) -> None:
+        codecs.register(self.search_function)
+
+    def unregister(self) -> None:
+        with contextlib.suppress(AttributeError):
+            codecs.unregister(self.search_function)

alternative_encodings/cp859.py

@@ -1,7 +1,7 @@
 import codecs
-import contextlib
 
 from .common import (
+    CodecBase,
     get_codec,
     get_incremental_decoder,
     get_incremental_encoder,
@@ -46,29 +46,20 @@
 
 # encodings module API
 
-codec = Codec()
+_codec = Codec()
 
 
-regentry = codecs.CodecInfo(
-    name="cp859",
-    encode=codec.encode,
-    decode=codec.decode,
-    incrementalencoder=IncrementalEncoder,
-    incrementaldecoder=IncrementalDecoder,
-)
-
-
-def search_function(encoding: str) -> codecs.CodecInfo | None:
-    if regentry.name == encoding:
-        return regentry
-
-    return None
-
+class Cp859Codec(CodecBase):
+    codec_info = codecs.CodecInfo(
+        name="cp859",
+        encode=_codec.encode,
+        decode=_codec.decode,
+        incrementalencoder=IncrementalEncoder,
+        incrementaldecoder=IncrementalDecoder,
+    )
 
-def register() -> None:
-    codecs.register(search_function)
+    def get_codec_info(self) -> codecs.CodecInfo:
+        return self.codec_info
 
 
-def unregister() -> None:
-    with contextlib.suppress(AttributeError):
-        codecs.unregister(search_function)
+codec = Cp859Codec()

alternative_encodings/cp866i.py

@@ -1,7 +1,8 @@
 import codecs
-import contextlib
 from encodings import cp866
 
+from alternative_encodings.common import CodecBase
+
 
 # Codec APIs
 class Codec(cp866.Codec):
@@ -19,29 +20,20 @@ def encode(self, input_string: str, final: bool = False) -> bytes:  # noqa: FBT0
 IncrementalDecoder = cp866.IncrementalDecoder
 
 # encodings module API
-codec = Codec()
-
-
-regentry = codecs.CodecInfo(
-    name="cp866i",
-    encode=codec.encode,
-    decode=codec.decode,
-    incrementalencoder=IncrementalEncoder,
-    incrementaldecoder=IncrementalDecoder,
-)
-
-
-def search_function(encoding: str) -> codecs.CodecInfo | None:
-    if regentry.name == encoding:
-        return regentry
+_codec = Codec()
 
-    return None
 
+class Cp866iCodec(CodecBase):
+    codec_info = codecs.CodecInfo(
+        name="cp866i",
+        encode=_codec.encode,
+        decode=_codec.decode,
+        incrementalencoder=IncrementalEncoder,
+        incrementaldecoder=IncrementalDecoder,
+    )
 
-def register() -> None:
-    codecs.register(search_function)
+    def get_codec_info(self) -> codecs.CodecInfo:
+        return self.codec_info
 
 
-def unregister() -> None:
-    with contextlib.suppress(AttributeError):
-        codecs.unregister(search_function)
+codec = Cp866iCodec()

alternative_encodings/romaji.py

@@ -1,11 +1,12 @@
 import codecs
-import contextlib
 import re
 from collections.abc import Iterator
 from encodings import cp437
 
 import cutlet
 
+from alternative_encodings.common import CodecBase
+
 katsu = cutlet.Cutlet(use_foreign_spelling=False)
 
 
@@ -48,28 +49,20 @@ def encode(self, input_string: str, final: bool = False) -> bytes:  # noqa: FBT0
 IncrementalDecoder = cp437.IncrementalDecoder
 
 # encodings module API
-codec = Codec()
-
-regentry = codecs.CodecInfo(
-    name="romaji",
-    encode=codec.encode,
-    decode=codec.decode,
-    incrementalencoder=IncrementalEncoder,
-    incrementaldecoder=IncrementalDecoder,
-)
-
-
-def search_function(encoding: str) -> codecs.CodecInfo | None:
-    if regentry.name == encoding:
-        return regentry
+_codec = Codec()
 
-    return None
 
+class RomajiCodec(CodecBase):
+    codec_info = codecs.CodecInfo(
+        name="romaji",
+        encode=_codec.encode,
+        decode=_codec.decode,
+        incrementalencoder=IncrementalEncoder,
+        incrementaldecoder=IncrementalDecoder,
+    )
 
-def register() -> None:
-    codecs.register(search_function)
+    def get_codec_info(self) -> codecs.CodecInfo:
+        return self.codec_info
 
 
-def unregister() -> None:
-    with contextlib.suppress(AttributeError):
-        codecs.unregister(search_function)
+codec = RomajiCodec()

alternative_encodings/viscii.py

@@ -1,7 +1,7 @@
 import codecs
-import contextlib
 
 from .common import (
+    CodecBase,
     get_codec,
     get_incremental_decoder,
     get_incremental_encoder,
@@ -46,29 +46,20 @@
 
 # encodings module API
 
-codec = Codec()
+_codec = Codec()
 
 
-regentry = codecs.CodecInfo(
-    name="viscii",
-    encode=codec.encode,
-    decode=codec.decode,
-    incrementalencoder=IncrementalEncoder,
-    incrementaldecoder=IncrementalDecoder,
-)
-
-
-def search_function(encoding: str) -> codecs.CodecInfo | None:
-    if regentry.name == encoding:
-        return regentry
-
-    return None
-
+class VisciiCodec(CodecBase):
+    codec_info = codecs.CodecInfo(
+        name="viscii",
+        encode=_codec.encode,
+        decode=_codec.decode,
+        incrementalencoder=IncrementalEncoder,
+        incrementaldecoder=IncrementalDecoder,
+    )
 
-def register() -> None:
-    codecs.register(search_function)
+    def get_codec_info(self) -> codecs.CodecInfo:
+        return self.codec_info
 
 
-def unregister() -> None:
-    with contextlib.suppress(AttributeError):
-        codecs.unregister(search_function)
+codec = VisciiCodec()

pyproject.toml

@@ -26,7 +26,7 @@ line-length = 120
 
 [tool.ruff.lint]
 select = ["ALL"]
-ignore = ["D"]  # Ignore all documentation warnings. Probably not the best option in real projects.
+ignore = ["D"]  # Ignore all documentation warnings
 fixable = ["ALL"]
 
 [tool.ruff.lint.per-file-ignores]

tests/test_cp859.py

@@ -10,13 +10,12 @@
 
 @pytest.fixture(scope="module", autouse=True)
 def _register_codec_fixture() -> Generator[None, None, None]:
-    with register_codec(cp859):
+    with register_codec(cp859.codec):
         yield
 
 
 source_data = (
-    "J'aime l'idée que le plus grand des Sex-symbols new-yorkais, "
-    "n'était qu'un chien dans un foyer de Brazzaville"
+    "J'aime l'idée que le plus grand des Sex-symbols new-yorkais, n'était qu'un chien dans un foyer de Brazzaville"
 )
 
 encoded = (
@@ -26,7 +25,8 @@ def _register_codec_fixture() -> Generator[None, None, None]:
 )
 
 
-@pytest.mark.parametrize(("source_data", "encoded"),
+@pytest.mark.parametrize(
+    ("source_data", "encoded"),
     [
         (source_data, encoded),
         ("\r\n", b"\r\n"),

tests/test_cp866i.py

@@ -10,7 +10,7 @@
 
 @pytest.fixture(scope="module", autouse=True)
 def _register_codec_fixture() -> Generator[None, None, None]:
-    with register_codec(cp866i):
+    with register_codec(cp866i.codec):
         yield
 
 

tests/test_general.py

@@ -2,18 +2,20 @@
 
 import pytest
 
-from alternative_encodings import cp859, cp866i, viscii
+from alternative_encodings import cp859, cp866i, romaji, viscii
 
 from .utils import register_codec
 
 codecs_pairs = [
-    ("cp859", cp859),
-    ("cp866i", cp866i),
-    ("viscii", viscii),
+    ("cp859", cp859.codec),
+    ("cp866i", cp866i.codec),
+    ("romaji", romaji.codec),
+    ("viscii", viscii.codec),
 ]
 
 
-@pytest.mark.parametrize("string",
+@pytest.mark.parametrize(
+    "string",
     [
         "\r\n",
     ],

tests/test_romaji.py

@@ -10,7 +10,7 @@
 
 @pytest.fixture(scope="module", autouse=True)
 def _register_codec_fixture() -> Generator[None, None, None]:
-    with register_codec(romaji):
+    with register_codec(romaji.codec):
         yield
 
 
@@ -25,7 +25,8 @@ def _register_codec_fixture() -> Generator[None, None, None]:
 """
 
 
-@pytest.mark.parametrize(("source_data", "encoded"),
+@pytest.mark.parametrize(
+    ("source_data", "encoded"),
     [
         (source_data, encoded),
         ("\r\n", b"\r\n"),

tests/test_viscii.py

@@ -10,7 +10,7 @@
 
 @pytest.fixture(scope="module", autouse=True)
 def _register_codec_fixture() -> Generator[None, None, None]:
-    with register_codec(viscii):
+    with register_codec(viscii.codec):
         yield