eaf-base-api 4.2.1__tar.gz → 4.2.2__tar.gz

{eaf_base_api-4.2.1 → eaf_base_api-4.2.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: eaf_base_api
-Version: 4.2.1
+Version: 4.2.2
 Summary: A base API for EchterAlsFake's Porn APIs
 Author: Johannes Habel
 Author-email: Johannes Habel <EchterAlsFake@proton.me>

{eaf_base_api-4.2.1 → eaf_base_api-4.2.2}/base_api/__init__.py

@@ -1,4 +1,8 @@
 __all__ = [
+    "append_url_path",
+    "profile_base_url",
+    "replace_url_path",
+    "update_url_query",
     "BaseCore",
     "BaseMedia",
     "Cache",
@@ -69,6 +73,7 @@ __all__ = [
 
 
 from base_api.modules import errors
+from base_api.modules.urls import append_url_path, profile_base_url, replace_url_path, update_url_query
 from base_api.modules.progress_bars import Callback
 from base_api.modules.errors import (
     DataNotLoadedError,

{eaf_base_api-4.2.1 → eaf_base_api-4.2.2}/base_api/base.py

@@ -29,6 +29,7 @@ from typing import (
 
 
 # 1. Standardize on relative imports
+from base_api.modules.urls import FetchedText
 from base_api.modules.errors import *
 from base_api.modules.type_hints import (
     DownloadReport, ResultOrder, ErrorMode, ErrorAction, ScrapeStage, 
@@ -422,6 +423,9 @@ class BaseMedia:
     url: str
     core: object
 
+    # Transport metadata is separate from lazily loaded fields and the caller's URL.
+    resolved_url: str | None = field(default=None, init=False, repr=False, compare=False)
+
     loader_methods: ClassVar[Mapping[str, str]] = {}
 
     _load_lock: asyncio.Lock = field(
@@ -2097,7 +2101,10 @@ class BaseCore:
                 params=params,
                 retry_non_idempotent=retry_non_idempotent,
             )
-            content = self._decode_response(response, url, self.logger)
+            content = FetchedText(
+                self._decode_response(response, url, self.logger),
+                url=getattr(response, "url", None) or url,
+            )
             if key is not None:
                 self.cache.set_response(key, content)
                 assert pending is not None

{eaf_base_api-4.2.1 → eaf_base_api-4.2.2}/base_api/modules/provider.py

@@ -7,6 +7,7 @@ from functools import wraps
 
 from base_api.modules import errors
 from base_api.modules.logger import class_name, get_logger, log_context
+from base_api.modules.urls import same_site_url
 
 
 async def fetch_content(core, url: str, *, logger, owner=None, get_json=False,
@@ -20,6 +21,10 @@ async def fetch_content(core, url: str, *, logger, owner=None, get_json=False,
     with log_context(owner, url):
         try:
             content = await core.fetch_text(url)
+            resolved = getattr(content, "url", None)
+            if (isinstance(resolved, str) and owner is not None
+                    and getattr(owner, "url", None) == url and same_site_url(url, resolved)):
+                owner.resolved_url = resolved
             return json.loads(content, strict=False) if get_json else content
         except errors.DownloadCancelled:
             raise

eaf_base_api-4.2.2/base_api/modules/urls.py

@@ -0,0 +1,51 @@
+"""Small URL operations shared by provider pagination and profile adapters."""
+
+from urllib.parse import parse_qsl, urlencode, urlsplit, urlunsplit
+
+
+def update_url_query(url: str, **changes: str | int | None) -> str:
+    """Replace named query keys without losing filters or treating anchors as HTTP."""
+    parsed = urlsplit(url)
+    pairs = [(key, value) for key, value in parse_qsl(parsed.query, keep_blank_values=True) if key not in changes]
+    pairs.extend((key, str(value)) for key, value in changes.items() if value is not None)
+    return urlunsplit(parsed._replace(query=urlencode(pairs), fragment=""))
+
+
+def replace_url_path(url: str, path: str) -> str:
+    parsed = urlsplit(url)
+    return urlunsplit(parsed._replace(path=path, fragment=""))
+
+
+def append_url_path(url: str, *parts: str | int) -> str:
+    path = urlsplit(url).path.rstrip("/")
+    return replace_url_path(url, path + "/" + "/".join(str(part).strip("/") for part in parts))
+
+
+def profile_base_url(url: str, *, depth: int = 2) -> str:
+    """Remove a provider's tab/page suffix, retaining its host, slug and filters."""
+    parsed = urlsplit(url)
+    parts = [part for part in parsed.path.split("/") if part]
+    if len(parts) < depth:
+        raise ValueError("An individual profile URL with a name or ID is required")
+    return update_url_query(replace_url_path(url, "/" + "/".join(parts[:depth])), page=None)
+
+
+def same_site_url(original: str, resolved: str) -> bool:
+    """Accept response metadata on the same site, including localized subdomains."""
+    before, after = urlsplit(original), urlsplit(resolved)
+    if after.scheme not in {"http", "https"} or not before.hostname or not after.hostname:
+        return False
+    # The supported providers use these two-label domains, except RedTube Brazil.
+    depth = 3 if before.hostname.endswith(".com.br") else 2
+    return before.hostname.split(".")[-depth:] == after.hostname.split(".")[-depth:]
+
+
+class FetchedText(str):
+    """A string that retains response identity through text caches and parsers."""
+
+    url: str
+
+    def __new__(cls, content: str, *, url: str):
+        value = super().__new__(cls, content)
+        value.url = url
+        return value

{eaf_base_api-4.2.1 → eaf_base_api-4.2.2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "uv_build"
 
 [project]
 name = "eaf_base_api"
-version = "4.2.1"
+version = "4.2.2"
 description = "A base API for EchterAlsFake's Porn APIs"
 requires-python = ">=3.12"
 license = "AGPL-3.0-or-later"

{eaf_base_api-4.2.1 → eaf_base_api-4.2.2}/pyproject.toml.orig

@@ -4,7 +4,7 @@ build-backend = "uv_build"
 
 [project]
 name = "eaf_base_api"
-version = "4.2.1"
+version = "4.2.2"
 description = "A base API for EchterAlsFake's Porn APIs"
 readme = { file = "README.md", content-type = "text/markdown" }
 requires-python = ">=3.12"