From bd8b0b4d0d1fb04e34002dd3da91d83b922d0991 Mon Sep 17 00:00:00 2001
From: Bryan Newbold <bnewbold@archive.org>
Date: Fri, 12 Nov 2021 14:55:53 -0800
Subject: grobid: handle XML parsing errors, and have them recorded in
 sandcrawler-db

---
 python/sandcrawler/grobid.py | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/python/sandcrawler/grobid.py b/python/sandcrawler/grobid.py
index 791e0fe..8c575bc 100644
--- a/python/sandcrawler/grobid.py
+++ b/python/sandcrawler/grobid.py
@@ -157,7 +157,11 @@ class GrobidClient(object):
     def metadata(self, result: Dict[str, Any]) -> Optional[Dict[str, Any]]:
         if result["status"] != "success":
             return None
-        tei_doc = parse_document_xml(result["tei_xml"])
+        try:
+            tei_doc = parse_document_xml(result["tei_xml"])
+        except xml.etree.ElementTree.ParseError as pe:
+            result['status'] = 'bad-grobid-xml'
+            return dict(error_msg=str(pe)[:1000])
         tei_doc.remove_encumbered()
         tei_json = tei_doc.to_legacy_dict()
         meta = dict()
-- 
cgit v1.2.3