fastfeedparser 0.5.8__tar.gz → 0.5.10__tar.gz

{fastfeedparser-0.5.8/src/fastfeedparser.egg-info → fastfeedparser-0.5.10}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: fastfeedparser
-Version: 0.5.8
+Version: 0.5.10
 Summary: High performance RSS, Atom, JSON and RDF feed parser in Python
 Home-page: https://github.com/kagisearch/fastfeedparser
 Author: Vladimir Prelovac

{fastfeedparser-0.5.8 → fastfeedparser-0.5.10}/setup.cfg

@@ -1,6 +1,6 @@
 [metadata]
 name = fastfeedparser
-version = 0.5.8
+version = 0.5.10
 author = Vladimir Prelovac
 author_email = vlad@kagi.com
 description = High performance RSS, Atom, JSON and RDF feed parser in Python

{fastfeedparser-0.5.8 → fastfeedparser-0.5.10}/src/fastfeedparser/main.py

@@ -253,12 +253,15 @@ def _fix_malformed_xml_bytes(content: bytes, actual_encoding: str = "utf-8") ->
 def _prepare_xml_bytes(xml_content: str | bytes) -> bytes:
     if isinstance(xml_content, bytes):
         cleaned = _clean_feed_bytes(xml_content)
-        if not cleaned.strip():
+        if not cleaned:
             raise ValueError("Empty content")
 
         # Replace Unicode LINE SEPARATOR (U+2028) and PARAGRAPH SEPARATOR (U+2029)
         # with regular newlines — these are invalid in XML 1.0 and cause lxml to fail.
-        if b"\xe2\x80\xa8" in cleaned or b"\xe2\x80\xa9" in cleaned:
+        # These are extremely rare; probe a small prefix to avoid full O(n) scan on
+        # multi-MB feeds.  If neither appears in the first 64 KB, skip the scan.
+        _PROBE = cleaned[:65536]
+        if b"\xe2\x80\xa8" in _PROBE or b"\xe2\x80\xa9" in _PROBE:
             cleaned = cleaned.replace(b"\xe2\x80\xa8", b"\n").replace(
                 b"\xe2\x80\xa9", b"\n"
             )
@@ -519,12 +522,14 @@ _STRICT_XML_PARSER = etree.XMLParser(
     recover=False,
     collect_ids=False,
     resolve_entities=False,
+    huge_tree=True,
 )
 _RECOVER_XML_PARSER = etree.XMLParser(
     ns_clean=True,
     recover=True,
     collect_ids=False,
     resolve_entities=False,
+    huge_tree=True,
 )
 
 
@@ -630,6 +635,7 @@ def _raise_for_non_feed_root(
 
 
 _RE_META_REFRESH_URL = re.compile(r'url\s*=\s*["\']?\s*([^"\'>\s]+)', re.IGNORECASE)
+_MAX_META_REDIRECTS = 3
 
 
 def _extract_meta_refresh_url(content: str | bytes, base_url: str) -> str | None:
@@ -861,31 +867,32 @@ def parse(
     else:
         content = source
 
-    try:
-        return _parse_content(
-            content,
-            include_content=include_content,
-            include_tags=include_tags,
-            include_media=include_media,
-            include_enclosures=include_enclosures,
-        )
-    except ValueError as e:
-        if not is_url:
-            raise
-        assert isinstance(source, str)
-        err_msg = str(e)
-        if "HTML" not in err_msg and "not a valid RSS/Atom feed" not in err_msg:
-            raise
-        redirect_url = _extract_meta_refresh_url(content, source)
-        if redirect_url is None:
-            raise
-        return parse(
-            redirect_url,
-            include_content=include_content,
-            include_tags=include_tags,
-            include_media=include_media,
-            include_enclosures=include_enclosures,
-        )
+    parse_kwargs = dict(
+        include_content=include_content,
+        include_tags=include_tags,
+        include_media=include_media,
+        include_enclosures=include_enclosures,
+    )
+
+    redirects_left = _MAX_META_REDIRECTS
+    while True:
+        try:
+            return _parse_content(content, **parse_kwargs)
+        except ValueError as e:
+            if not is_url:
+                raise
+            assert isinstance(source, str)
+            err_msg = str(e)
+            if "HTML" not in err_msg and "not a valid RSS/Atom feed" not in err_msg:
+                raise
+            if redirects_left <= 0:
+                raise ValueError("too many meta-refresh redirects") from e
+            redirect_url = _extract_meta_refresh_url(content, source)
+            if redirect_url is None:
+                raise
+            content = _fetch_url_content(redirect_url)
+            source = redirect_url
+            redirects_left -= 1
 
 
 def _parse_feed_info(

{fastfeedparser-0.5.8 → fastfeedparser-0.5.10/src/fastfeedparser.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: fastfeedparser
-Version: 0.5.8
+Version: 0.5.10
 Summary: High performance RSS, Atom, JSON and RDF feed parser in Python
 Home-page: https://github.com/kagisearch/fastfeedparser
 Author: Vladimir Prelovac