Allow using system feedparser with READER_USE_SYSTEM_FEEDPARSER. #350

CHANGES.rst

@@ -19,6 +19,13 @@ Unreleased
     :attr:`~.RetrievedFeed.caching_info`).
   * The ``HTTPAcceptParserType`` was renamed to :class:`.AcceptParserType`.
 
+* Allow :ref:`using the system-installed feedparser <use-system-feedparser>`
+  instead of the one vendored by *reader*.
+  This is useful for working around issues in the vendored feedparser,
+  like the libxml2 <=2.13.3 incompatibility reported in :issue:`350`;
+  thanks to `Maks Verver`_  for reporting,
+  root causing, and following up with both dependencies.
+
 * Fix a number of brittle / broken tests.
   Thanks to `Maks Verver`_ for the issues and fixes.
   (:issue:`348`, :issue:`349`, :issue:`355`)
@@ -670,6 +677,8 @@ Released 2022-03-12
   Fix broken error flashing.
 
 
+.. _version 2.9:
+
 Version 2.9
 -----------
 

docs/install.rst

@@ -27,6 +27,20 @@ These packages will be installed automatically when installing *reader*:
 (at least SQLite 3.18 with the `JSON1`_ and `FTS5`_ extensions).
 
 
+.. _use-system-feedparser:
+
+.. note::
+
+  Because `feedparser`_ makes PyPI releases at a lower cadence,
+  *reader* uses a vendored version of feedparser's `develop`_ branch
+  by default since :ref:`version 2.9`.
+  To opt out of this behavior, and make *reader* use
+  the system-installed ``feedparser`` package,
+  set the ``READER_USE_SYSTEM_FEEDPARSER`` environment variable to ``1``.
+
+.. _develop: https://github.com/kurtmckee/feedparser
+
+
 .. _optional dependencies:
 
 Optional dependencies

src/reader/_parser/feedparser.py

@@ -2,6 +2,7 @@
 
 import calendar
 import logging
+import os
 import time
 import warnings
 from datetime import datetime
@@ -12,14 +13,19 @@
 
 from .._types import EntryData
 from .._types import FeedData
-from .._vendor import feedparser
 from ..exceptions import ParseError
 from ..types import Content
 from ..types import Enclosure
 from ._http_utils import parse_accept_header
 from ._http_utils import unparse_accept_header
 
 
+if os.environ.get('READER_USE_SYSTEM_FEEDPARSER', '') not in ('', '0'):
+    import feedparser  # type: ignore
+else:
+    from .._vendor import feedparser
+
+
 if TYPE_CHECKING:  # pragma: no cover
     from . import FeedAndEntries
     from .requests import Headers
@@ -55,7 +61,7 @@ def __call__(
             resource,
             resolve_relative_uris=True,
             sanitize_html=True,
-            response_headers=headers or {},  # type: ignore[arg-type]
+            response_headers=headers or {},
         )
         return _process_feed(url, result)
 

tests/test_parser.py

@@ -14,12 +14,12 @@
 from reader._parser import Parser
 from reader._parser import RetrievedFeed
 from reader._parser import RetrieveError
+from reader._parser.feedparser import feedparser
 from reader._parser.feedparser import FeedparserParser
 from reader._parser.file import FileRetriever
 from reader._parser.jsonfeed import JSONFeedParser
 from reader._parser.requests import SessionWrapper
 from reader._types import FeedData
-from reader._vendor import feedparser
 from reader.exceptions import ParseError
 from utils import make_url_base
 
@@ -1164,3 +1164,24 @@ def retrieve(*_, **__):
     assert exc_info.value.__cause__ is cause
 
     assert parse.last_result.http_info == HTTPInfo(200, {})
+
+
+def test_reader_use_system_feedparser(monkeypatch, reload_module):
+    import feedparser
+
+    import reader._parser.feedparser
+    import reader._vendor.feedparser
+
+    name = 'READER_USE_SYSTEM_FEEDPARSER'
+
+    monkeypatch.delenv(name, raising=False)
+    reload_module(reader._parser.feedparser)
+    assert reader._parser.feedparser.feedparser is reader._vendor.feedparser
+
+    monkeypatch.setenv(name, '0')
+    reload_module(reader._parser.feedparser)
+    assert reader._parser.feedparser.feedparser is reader._vendor.feedparser
+
+    monkeypatch.setenv(name, '1')
+    reload_module(reader._parser.feedparser)
+    assert reader._parser.feedparser.feedparser is feedparser