fix: get rid of useless caching (#9)

* fix: get rid of caching

* fix: remove dead code

* fix: remove more dead code

pyproject.toml

@@ -25,11 +25,8 @@ httpx = ">=0.27.2"
 pydantic = ">=2.9.1"
 beautifulsoup4 = ">=4.12.3"
 lxml = ">=5.3.0"
-hishel = ">=0.0.30"
 torf = ">=4.2.7"
 typing-extensions = ">=4.12.2"
-platformdirs = ">=4.3.3"
-xmltodict = ">=0.13.0"
 strenum = { version = ">=0.4.15", python = "<3.11" }
 
 [tool.poetry.group.dev.dependencies]

src/pynyaa/_clients/_async.py

@@ -1,22 +1,20 @@
 from __future__ import annotations
 
 from io import BytesIO
-from pathlib import Path
 from typing import Any
 from urllib.parse import urljoin
 
-from hishel import AsyncCacheClient, AsyncFileStorage
+from httpx import AsyncClient
 from torf import Torrent
 from typing_extensions import AsyncGenerator
 
 from pynyaa._enums import Category, Filter, SortBy
 from pynyaa._models import NyaaTorrentPage
 from pynyaa._parser import parse_nyaa_search_results, parse_nyaa_torrent_page
-from pynyaa._utils import get_user_cache_path
 
 
 class AsyncNyaa:
-    def __init__(self, base_url: str = "https://nyaa.si/", cache: bool = True, **kwargs: Any) -> None:
+    def __init__(self, base_url: str = "https://nyaa.si/", **kwargs: Any) -> None:
         """
         Async Nyaa client.
 
@@ -25,22 +23,12 @@ def __init__(self, base_url: str = "https://nyaa.si/", cache: bool = True, **kwa
         base_url : str, optional
             The base URL of Nyaa. Default is `https://nyaa.si/`.
             This is used for constructing the full URL from relative URLs.
-        cache : bool, optional
-            Whether to enable caching. Default is `True`.
-            This will cache the page upon it's first request and then use the cached result
-            for any subsequent requests for the same page.
-            This helps in avoiding [HTTP 429 Error](https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/429) but
-            do note some fields like seeders, leechers, and completed are constantly changing and thus caching would
-            mean you won't get the latest data on said fields.
         kwargs : Any, optional
             Keyword arguments to pass to the underlying [`httpx.AsyncClient`](https://www.python-httpx.org/api/#asyncclient)
             used to make the GET request.
         """
         self._base_url = base_url
-        self._cache = cache
         self._kwargs = kwargs
-        self._extensions = {"force_cache": self._cache, "cache_disabled": not self._cache}
-        self._storage = AsyncFileStorage(base_path=get_user_cache_path())
 
     @property
     def base_url(self) -> str:
@@ -49,13 +37,6 @@ def base_url(self) -> str:
         """
         return self._base_url
 
-    @property
-    def cache_path(self) -> Path:
-        """
-        Path where cache files are stored.
-        """
-        return get_user_cache_path()
-
     async def get(self, page: int | str) -> NyaaTorrentPage:
         """
         Retrieve information from a Nyaa torrent page.
@@ -85,14 +66,14 @@ async def get(self, page: int | str) -> NyaaTorrentPage:
             url = page
             nyaaid = page.split("/")[-1]  # type: ignore
 
-        async with AsyncCacheClient(storage=self._storage, **self._kwargs) as client:
-            nyaa = await client.get(url, extensions=self._extensions)
+        async with AsyncClient(**self._kwargs) as client:
+            nyaa = await client.get(url)
             nyaa.raise_for_status()
 
             parsed = parse_nyaa_torrent_page(self._base_url, nyaa.text)
 
             # Get the torrent file and convert it to a torf.Torrent object
-            response = await client.get(parsed["torrent_file"], extensions=self._extensions)
+            response = await client.get(parsed["torrent_file"])
             response.raise_for_status()
             torrent = Torrent.read_stream(BytesIO(response.content))
 
@@ -133,7 +114,7 @@ async def search(
         NyaaTorrentPage
             A NyaaTorrentPage object representing the retrieved data.
         """
-        async with AsyncCacheClient(storage=self._storage, **self._kwargs) as client:
+        async with AsyncClient(**self._kwargs) as client:
             params: dict[str, Any] = dict(
                 f=filter,
                 c=category.id,
@@ -142,7 +123,7 @@ async def search(
                 o="asc" if reverse else "desc",
             )
 
-            nyaa = await client.get(self._base_url, params=params, extensions=self._extensions)
+            nyaa = await client.get(self._base_url, params=params)
             nyaa.raise_for_status()
             results = parse_nyaa_search_results(nyaa.text)
 

src/pynyaa/_clients/_sync.py

@@ -1,22 +1,20 @@
 from __future__ import annotations
 
 from io import BytesIO
-from pathlib import Path
 from typing import Any
 from urllib.parse import urljoin
 
-from hishel import CacheClient, FileStorage
+from httpx import Client
 from torf import Torrent
 from typing_extensions import Generator
 
 from pynyaa._enums import Category, Filter, SortBy
 from pynyaa._models import NyaaTorrentPage
 from pynyaa._parser import parse_nyaa_search_results, parse_nyaa_torrent_page
-from pynyaa._utils import get_user_cache_path
 
 
 class Nyaa:
-    def __init__(self, base_url: str = "https://nyaa.si/", cache: bool = True, **kwargs: Any) -> None:
+    def __init__(self, base_url: str = "https://nyaa.si/", **kwargs: Any) -> None:
         """
         Nyaa client.
 
@@ -25,22 +23,12 @@ def __init__(self, base_url: str = "https://nyaa.si/", cache: bool = True, **kwa
         base_url : str, optional
             The base URL of Nyaa. Default is `https://nyaa.si/`.
             This is used for constructing the full URL from relative URLs.
-        cache : bool, optional
-            Whether to enable caching. Default is `True`.
-            This will cache the page upon it's first request and then use the cached result
-            for any subsequent requests for the same page.
-            This helps in avoiding [HTTP 429 Error](https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/429) but
-            do note some fields like seeders, leechers, and completed are constantly changing and thus caching would
-            mean you won't get the latest data on said fields.
         kwargs : Any, optional
             Keyword arguments to pass to the underlying [`httpx.Client`](https://www.python-httpx.org/api/#client)
             used to make the GET request.
         """
         self._base_url = base_url
-        self._cache = cache
         self._kwargs = kwargs
-        self._extensions = {"force_cache": self._cache, "cache_disabled": not self._cache}
-        self._storage = FileStorage(base_path=get_user_cache_path())
 
     @property
     def base_url(self) -> str:
@@ -49,13 +37,6 @@ def base_url(self) -> str:
         """
         return self._base_url
 
-    @property
-    def cache_path(self) -> Path:
-        """
-        Path where cache files are stored.
-        """
-        return get_user_cache_path()
-
     def get(self, page: int | str) -> NyaaTorrentPage:
         """
         Retrieve information from a Nyaa torrent page.
@@ -85,12 +66,12 @@ def get(self, page: int | str) -> NyaaTorrentPage:
             url = page
             nyaaid = page.split("/")[-1]  # type: ignore
 
-        with CacheClient(storage=self._storage, **self._kwargs) as client:
-            nyaa = client.get(url, extensions=self._extensions).raise_for_status()
+        with Client(**self._kwargs) as client:
+            nyaa = client.get(url).raise_for_status()
             parsed = parse_nyaa_torrent_page(self._base_url, nyaa.text)
 
             # Get the torrent file and convert it to a torf.Torrent object
-            torrent_file = client.get(parsed["torrent_file"], extensions=self._extensions).raise_for_status().content
+            torrent_file = client.get(parsed["torrent_file"]).raise_for_status().content
             torrent = Torrent.read_stream(BytesIO(torrent_file))
 
             return NyaaTorrentPage(id=nyaaid, url=url, torrent=torrent, **parsed)  # type: ignore
@@ -130,7 +111,7 @@ def search(
         NyaaTorrentPage
             A NyaaTorrentPage object representing the retrieved data.
         """
-        with CacheClient(storage=self._storage, **self._kwargs) as client:
+        with Client(**self._kwargs) as client:
             params: dict[str, Any] = dict(
                 f=filter,
                 c=category.id,
@@ -139,7 +120,7 @@ def search(
                 o="asc" if reverse else "desc",
             )
 
-            nyaa = client.get(self._base_url, params=params, extensions=self._extensions).raise_for_status()
+            nyaa = client.get(self._base_url, params=params).raise_for_status()
             results = parse_nyaa_search_results(nyaa.text)
 
             for link in results:

src/pynyaa/_parser.py

@@ -6,7 +6,6 @@
 
 from bs4 import BeautifulSoup
 from typing_extensions import Generator
-from xmltodict import parse as xmltodict_parse
 
 
 def parse_nyaa_torrent_page(base_url: str, html: str) -> dict[str, Any]:
@@ -119,33 +118,6 @@ def parse_nyaa_torrent_page(base_url: str, html: str) -> dict[str, Any]:
     )
 
 
-def parse_nyaa_rss_page(xml: str) -> Generator[str]:
-    """
-    Parse the torrent links out of the RSS page
-
-    Parameters
-    ----------
-    xml : str
-        Nyaa's RSS XML data as a string
-
-    Yields
-    ------
-    str
-        The full URL of torrent page, in the order they were present.
-    """
-    try:
-        items = xmltodict_parse(xml, encoding="utf-8")["rss"]["channel"]["item"]
-    except KeyError:
-        yield from ()
-        return
-
-    if isinstance(items, dict):  # RSS returns single results as a dict instead of a list
-        items = [items]
-
-    for item in items:
-        yield item["guid"]["#text"]
-
-
 def parse_nyaa_search_results(html: str, base_url: str = "https://nyaa.si") -> Generator[str]:
     """
     Parses the HTML of a Nyaa search results page to extract torrent links.

src/pynyaa/_utils.py

@@ -1,29 +1,11 @@
 from __future__ import annotations
 
-from functools import cache
 from typing import TYPE_CHECKING
 
-from platformdirs import user_cache_path
-
 if TYPE_CHECKING:  # pragma: no cover
-    from pathlib import Path
-    from typing import Callable, ParamSpec, TypeVar
-
     from pynyaa._types import CategoryID
 
-    P = ParamSpec("P")
-    T = TypeVar("T")
-
-    # functools.cache destroys the signature of whatever it wraps, so we use this to fix it.
-    # This is to only "fool" typecheckers and IDEs, this doesn't exist at runtime.
-    def cache(user_function: Callable[P, T], /) -> Callable[P, T]: ...  # type: ignore
-
-
-def get_user_cache_path() -> Path:
-    return user_cache_path(appname="pynyaa", ensure_exists=True).resolve()
-
 
-@cache
 def _get_category_id_from_name(key: str) -> CategoryID:
     mapping: dict[str, CategoryID] = {
         # All, c=0_0

tests/test_async.py

@@ -1,7 +1,5 @@
 from __future__ import annotations
 
-from platformdirs import user_cache_path
-
 from pynyaa import AsyncNyaa, Category
 
 headers = {
@@ -12,7 +10,6 @@
 
 async def test_properties() -> None:
     assert client.base_url == "https://nyaa.si/"
-    assert client.cache_path == user_cache_path(appname="pynyaa").resolve()
 
 
 async def test_nyaa_default() -> None:

tests/test_sync.py

@@ -1,7 +1,5 @@
 from __future__ import annotations
 
-from platformdirs import user_cache_path
-
 from pynyaa import Category, Nyaa
 
 headers = {
@@ -12,7 +10,6 @@
 
 def test_properties() -> None:
     assert client.base_url == "https://nyaa.si/"
-    assert client.cache_path == user_cache_path(appname="pynyaa").resolve()
 
 
 def test_nyaa_default() -> None: