Escape angle brackets appearing in the abstract

src/pdf_extract.py

@@ -40,6 +40,10 @@ def write_record(self, name, model_endpath, pdf_file, pdf_url, organisation, tit
         keywords = list(kwset)
         bbox_list = [str(bbox['west']), str(bbox['east']), str(bbox['south']), str(bbox['north'])]
 
+        # Remove encodings that can upset XML
+        summary = summary.replace('<', '&lt;')
+        summary = summary.replace('>', '&gt;')
+
         # Assemble dict for jinja template
         mcf_dict = {
             "mcf": {
@@ -135,11 +139,12 @@ def write_record(self, name, model_endpath, pdf_file, pdf_url, organisation, tit
                     "level": "dataset"
                 },
                 "lineage": {
-                    "statement": f"This metadata record was reproduced from the PDF report retrieved from {pdf_url} on {datetime.datetime.now():%d %b %Y}. The abstract was generated by Antrhopic Claude V2.0 (https://www.anthropic.com/). Keywords were taken from USGS Thesaurus (https://apps.usgs.gov/thesaurus/) and extracted by yake (https://pypi.org/project/yake)"
+                    "statement": f"This metadata record was reproduced from the PDF report retrieved from {pdf_url} on {datetime.datetime.now():%d %b %Y}. The abstract was generated by Anthropic Claude V2.0 (https://www.anthropic.com/). Keywords were taken from USGS Thesaurus (https://apps.usgs.gov/thesaurus/) and extracted by yake (https://pypi.org/project/yake)"
                 }
             }
         }
 
+        # Create ISO 19115-3 XML with a modified version pygeometa's jinja template
         template_dir = os.path.join(os.path.dirname(__file__), '../data/templates/ISO19115-3')
         xml_string = render_j2_template(mcf_dict, template_dir=template_dir)