Enable XML metadata parsing from documents for GeoNode. See https://github.com/GeoNode/geonode/issues/6876

document_xml_metadata_ingest.patch

Index: geonode/documents/views.py
IDEA additional info:
Subsystem: com.intellij.openapi.diff.impl.patch.CharsetEP
<+>UTF-8
===================================================================
diff --git a/geonode/documents/views.py b/geonode/documents/views.py
--- a/geonode/documents/views.py	(revision a5a217bd49e7f91b69ace14a54cdb7219fa5d0dc)
+++ b/geonode/documents/views.py	(revision 9e436c39e470b5a8aae6304f4a62ccbe00f4cf1c)
@@ -49,7 +49,8 @@
 from geonode.base.forms import CategoryForm, TKeywordForm
 from geonode.base.models import (
     Thesaurus,
-    TopicCategory)
+    TopicCategory,
+    ResourceBase)
 from geonode.documents.enumerations import DOCUMENT_TYPE_MAP, DOCUMENT_MIMETYPE_MAP
 from geonode.documents.models import Document, get_related_resources
 from geonode.documents.forms import DocumentForm, DocumentCreateForm, DocumentReplaceForm
@@ -62,6 +63,12 @@
 
 from dal import autocomplete
 
+# these two should get refactored into the base model
+from geonode.layers.metadata import set_metadata
+from geonode.layers.utils import resolve_regions
+
+from geonode.base.models import SpatialRepresentationType
+
 logger = logging.getLogger("geonode.documents.views")
 
 ALLOWED_DOC_TYPES = settings.ALLOWED_DOCUMENT_TYPES
@@ -238,6 +245,8 @@
         form.save_many2many()
         self.object.set_permissions(form.cleaned_data['permissions'])
 
+        identifier = None
+        title = None
         abstract = None
         date = None
         regions = []
@@ -258,6 +267,52 @@
             except Exception:
                 logger.error("Exif extraction failed.")
 
+        # extract XML metadata
+        if self.object.mime_type == 'application/xml' \
+                and getattr(settings, 'EXTRACT_METADATA_FROM_XML_DOCS', True):
+            try:
+                # set metadata
+                with open(self.object.doc_file.path) as f:
+                    xml_file = f.read()
+
+                # get model properties from XML
+                identifier, vals, regions, keywords = set_metadata(xml_file)
+
+                metadata = {}
+
+                metadata['metadata_xml'] = xml_file
+                metadata['uuid'] = identifier
+
+                for key, value in vals.items():
+                    if key == 'spatial_representation_type':
+                        value = SpatialRepresentationType(identifier=value)
+                    elif key == 'topic_category':
+                        value, created = TopicCategory.objects.get_or_create(
+                            identifier=value.lower(),
+                            defaults={'description': '', 'gn_description': value})
+                        key = 'category'
+                    metadata[key] = value
+            except Exception:
+                logger.error("XML metadata extraction failed.")
+
+            # use values to populate Document fields
+            title = metadata['title']
+            abstract = metadata['abstract']
+            date = metadata['date']
+
+        if identifier:
+            if not (ResourceBase.objects.filter(uuid=identifier)):
+                self.object.uuid = identifier
+            else:
+                logger.error("The UUID identifier from the XML Metadata is already in use in this system.")
+                # uuid is used but self.object is already created in database
+                # delete it and raise error message
+                self.object.delete()
+                raise AttributeError("The UUID identifier from the XML Metadata is already in use in this system.")
+
+        if title:
+            self.object.title = title
+
         if abstract:
             self.object.abstract = abstract
 
@@ -265,8 +320,19 @@
             self.object.date = date
             self.object.date_type = "Creation"
 
+        # resolving all regions found by the Region model see geonode/geoserver/tasks.py:327
+        # add the unresolved ones to the keywords
         if len(regions) > 0:
-            self.object.regions.add(*regions)
+            regions_resolved, regions_unresolved = resolve_regions(regions)
+            regions_resolved = list(set(regions_resolved))
+            keywords.extend(regions_unresolved)
+            if regions_resolved:
+                if len(regions_resolved) > 0:
+                    if not self.object.regions:
+                        self.object.regions = regions_resolved
+                    else:
+                        self.object.regions.clear()
+                        self.object.regions.add(*regions_resolved)
 
         if len(keywords) > 0:
             self.object.keywords.add(*keywords)