fastfeedparser 0.5.7__tar.gz → 0.5.9__tar.gz

{fastfeedparser-0.5.7/src/fastfeedparser.egg-info → fastfeedparser-0.5.9}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: fastfeedparser
-Version: 0.5.7
+Version: 0.5.9
 Summary: High performance RSS, Atom, JSON and RDF feed parser in Python
 Home-page: https://github.com/kagisearch/fastfeedparser
 Author: Vladimir Prelovac

{fastfeedparser-0.5.7 → fastfeedparser-0.5.9}/setup.cfg

@@ -1,6 +1,6 @@
 [metadata]
 name = fastfeedparser
-version = 0.5.7
+version = 0.5.9
 author = Vladimir Prelovac
 author_email = vlad@kagi.com
 description = High performance RSS, Atom, JSON and RDF feed parser in Python

{fastfeedparser-0.5.7 → fastfeedparser-0.5.9}/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,
 )
 
 
@@ -991,8 +996,19 @@ def _parse_feed_info(
     for link in channel.findall(f"{{{atom_ns}}}link"):
         rel = link.get("rel")
         href = link.get("href") or link.get("link")
-        if rel is None and href:
+        if rel == "alternate" and href and not feed_link:
             feed_link = href
+            feed_links.append(
+                {
+                    "rel": rel,
+                    "type": link.get("type"),
+                    "href": href,
+                    "title": link.get("title"),
+                }
+            )
+        elif rel is None and href:
+            if not feed_link:
+                feed_link = href
         elif rel not in {"hub", "self", "replies", "edit"}:
             feed_links.append(
                 {
@@ -1134,7 +1150,10 @@ def _populate_entry_links_from_elements(
             "title": link.get("title"),
         }
         if rel == "alternate":
-            alternate_link = link_dict
+            if alternate_link is None:
+                alternate_link = link_dict
+            else:
+                entry_links.append(link_dict)
         elif rel not in {"edit", "self"}:
             entry_links.append(link_dict)
 

{fastfeedparser-0.5.7 → fastfeedparser-0.5.9/src/fastfeedparser.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: fastfeedparser
-Version: 0.5.7
+Version: 0.5.9
 Summary: High performance RSS, Atom, JSON and RDF feed parser in Python
 Home-page: https://github.com/kagisearch/fastfeedparser
 Author: Vladimir Prelovac