fallback in extractor if URN not valid

src/extractor.py

@@ -402,6 +402,23 @@ def get_pdf_page_count(pdf_bytes: bytes) -> int:
         return 0
 
 
+def _profile_from_urn(urn: str) -> str:
+    """Extract a short profile name from a Factur-X/ZUGFeRD URN.
+
+    Falls back to the last segment of the URN after '#', or 'unknown'.
+
+    Args:
+        urn: The full profile URN (e.g.
+            'urn:cen.eu:en16931:2017#compliant#urn:xeinkauf.de:kosit:xrechnung_3.0')
+
+    Returns:
+        Short profile name (e.g. 'xrechnung_3.0')
+    """
+    if not urn:
+        return "unknown"
+    return urn.rsplit("#", maxsplit=1)[-1].rsplit(":", maxsplit=1)[-1]
+
+
 def extract_zugferd(pdf_bytes: bytes) -> ExtractResponse:
     """Extract ZUGFeRD data from PDF bytes.
 
@@ -460,7 +477,17 @@ def extract_zugferd(pdf_bytes: bytes) -> ExtractResponse:
 
     xml_root = etree.fromstring(xml_bytes)
     flavor = get_flavor(xml_root)
-    level = get_level(xml_root, flavor)
+
+    try:
+        level = get_level(xml_root, flavor)
+    except ValueError:
+        urn_nodes = xml_root.xpath(
+            "//rsm:ExchangedDocumentContext/"
+            "rsm:GuidelineSpecifiedDocumentContextParameter/ram:ID/text()",
+            namespaces=NAMESPACES,
+        )
+        urn = urn_nodes[0] if urn_nodes else ""
+        level = _profile_from_urn(urn)
 
     xml_data = parse_zugferd_xml(xml_bytes)
     pdf_text = extract_text_from_pdf(pdf_bytes)