Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,10 +1,36 @@
 import datetime
 from urllib.parse import quote
-
 import requests
+from requests.adapters import HTTPAdapter
+from urllib3.util.retry import Retry
 from lxml import etree
 import streamlit as st
 
+# =====================================================================
+# Networking Configuration
+# =====================================================================
+
+HEADERS = {
+    "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/123.0.0.0 Safari/537.36",
+    "Accept": "application/xml, text/xml, */*"
+}
+
+def get_robust_session():
+    """Returns a requests session with retry logic for transient errors (500, 502, 503, 504)."""
+    session = requests.Session()
+    retry_strategy = Retry(
+        total=5,
+        backoff_factor=1,
+        status_forcelist=[429, 500, 502, 503, 504],
+        allowed_methods=["GET"]
+    )
+    adapter = HTTPAdapter(max_retries=retry_strategy)
+    session.mount("http://", adapter)
+    session.mount("https://", adapter)
+    return session
+
+ROBUST_SESSION = get_robust_session()
+
 # =====================================================================
 # Namespaces
 # =====================================================================
@@ -64,7 +90,7 @@ def build_persistent_url(repo_code: str, object_id: str) -> str:
 
 def fetch_mods_xml(mods_url: str) -> etree._Element:
     """Lädt eine MODS-Datei oder OAI-PMH von einer URL und gibt das MODS Root-Element zurück."""
-    resp = requests.get(mods_url)
+    resp = ROBUST_SESSION.get(mods_url, headers=HEADERS, timeout=60)
     resp.raise_for_status()
     # Use recover=True to handle malformed XML (e.g. unescaped HTML in notes)
     parser = etree.XMLParser(recover=True, remove_blank_text=True)
@@ -443,7 +469,7 @@ class CrossrefSchemaResolver(etree.Resolver):
             schema_url = f"https://www.crossref.org/schemas/{url}"
 
         try:
-            response = requests.get(schema_url, timeout=15)
+            response = ROBUST_SESSION.get(schema_url, headers=HEADERS, timeout=30)
             response.raise_for_status()
             return self.resolve_string(response.content, context)
         except Exception:
@@ -475,7 +501,7 @@ def validate_crossref_xml(xml_bytes: bytes) -> tuple[bool, list[str]]:
                 parser.resolvers.add(CrossrefSchemaResolver())
 
                 # Download main schema
-                schema_resp = requests.get(schema_url, timeout=30)
+                schema_resp = ROBUST_SESSION.get(schema_url, headers=HEADERS, timeout=60)
                 schema_resp.raise_for_status()
 
                 # Parse schema with resolver