Strip links from meta fields

[paperdoorknob] / fetch.py

diff --git a/fetch.py b/fetch.py
index 3c0c6a3aa9c47bc63999e0adde5cb750df8c85e6..eb904a4f4b0e525756c977d442fb0e4144dc0825 100644 (file)
--- a/fetch.py
+++ b/fetch.py
@@ -7,11 +7,18 @@
 
 from abc import ABC, abstractmethod
 from contextlib import contextmanager
 
 from abc import ABC, abstractmethod
 from contextlib import contextmanager

-from typing import Iterator
+from sys import stderr
+from typing import Callable, Iterator

 
 import requests
 import requests_cache
 
 
 import requests
 import requests_cache
 

+from version import paperdoorknob_version
+
+
+_headers = {'User-Agent': f'paperdoorknob/{paperdoorknob_version} ' +
+            '(https://git.scottworley.com/paperdoorknob/)'}
+

 
 class Fetcher(ABC):
     @abstractmethod
 
 class Fetcher(ABC):
     @abstractmethod


@@ -26,11 +33,38 @@ class _SessionFetcher(Fetcher):
         self._timeout = timeout
 
     def fetch(self, url: str) -> bytes:
         self._timeout = timeout
 
     def fetch(self, url: str) -> bytes:

-        with self._session.get(url, timeout=self._timeout) as r:
+        with self._session.get(url, timeout=self._timeout, headers=_headers) as r:

             r.raise_for_status()
             return r.content
 
 
             r.raise_for_status()
             return r.content
 
 

+class _CachingFetcher(Fetcher):
+
+    def __init__(
+            self,
+            session: requests_cache.CachedSession,
+            timeout: int) -> None:
+        self._session = session
+        self._timeout = timeout
+        self._request_count = 0
+        self._cache_hit_count = 0
+
+    def fetch(self, url: str) -> bytes:
+        with self._session.get(url, timeout=self._timeout, headers=_headers) as r:
+            r.raise_for_status()
+            self._request_count += 1
+            self._cache_hit_count += int(r.from_cache)
+            return r.content
+
+    @property
+    def request_count(self) -> int:
+        return self._request_count
+
+    @property
+    def cache_hit_count(self) -> int:
+        return self._cache_hit_count
+
+

 @contextmanager
 def DirectFetcher(timeout: int) -> Iterator[_SessionFetcher]:
     with requests.session() as session:
 @contextmanager
 def DirectFetcher(timeout: int) -> Iterator[_SessionFetcher]:
     with requests.session() as session:


@@ -38,6 +72,30 @@ def DirectFetcher(timeout: int) -> Iterator[_SessionFetcher]:
 
 
 @contextmanager
 
 
 @contextmanager

-def CachingFetcher(cache_path: str, timeout: int) -> Iterator[_SessionFetcher]:
+def CachingFetcher(
+    cache_path: str,
+    timeout: int,
+    log: Callable[[str], None] = lambda x: print(x, file=stderr),
+) -> Iterator[_CachingFetcher]:

     with requests_cache.CachedSession(cache_path, cache_control=True) as session:
     with requests_cache.CachedSession(cache_path, cache_control=True) as session:

-        yield _SessionFetcher(session, timeout)
+        fetcher = _CachingFetcher(session, timeout)
+        yield fetcher
+        if fetcher.request_count > 0:
+            percent = 100.0 * fetcher.cache_hit_count / fetcher.request_count
+            log(f"Fetch cache hits: {fetcher.cache_hit_count} ({percent:.1f}%)")
+
+
+class FakeFetcher(Fetcher):
+
+    def __init__(self, resources: dict[str, bytes]) -> None:
+        self._resources = resources
+        self._fetch_count = 0
+
+    def fetch(self, url: str) -> bytes:
+        self._fetch_count += 1
+        if url not in self._resources:
+            raise requests.HTTPError("URL not found", url)
+        return self._resources[url]
+
+    def request_count(self) -> int:
+        return self._fetch_count