workaround for Spotlight availability

ceteri · ceteri · commit 62b2b11e17de · 2024-02-15T09:40:15.000-08:00
diff --git a/NOTES.md b/NOTES.md
@@ -1,5 +1,11 @@
 TODO:
 
+  * can we build a causal graph of the provenance?
+    - https://www.pywhy.org/dowhy/v0.11.1/
+
+  * target publications:
+    - https://drops.dagstuhl.de/entities/issue/TGDK-volume-1-issue-1
+
   * impl a _semantic random walk_ from a source KG
 
   * link entities for lemmas, noun chunks using MediaWiki lookups?
diff --git a/textgraphs/kg.py b/textgraphs/kg.py
@@ -848,7 +848,7 @@ def dbpedia_wikidata_equiv (
     ######################################################################
     ## entity linking
 
-    def _link_spotlight_entities (  # pylint: disable=R0914
+    def _link_spotlight_entities (  # pylint: disable=R0912,R0914
         self,
         pipe: Pipeline,
         *,
@@ -867,7 +867,10 @@ def _link_spotlight_entities (  # pylint: disable=R0914
     yields:
 candidates linked entities
         """
-        ents: typing.List[ spacy.tokens.span.Span ] = list(pipe.aux_doc.ents)
+        ents: typing.List[ spacy.tokens.span.Span ] = []
+
+        if pipe.aux_doc is not None:
+            list(pipe.aux_doc.ents)
 
         if debug:
             ic(ents)
diff --git a/textgraphs/pipe.py b/textgraphs/pipe.py
@@ -251,8 +251,14 @@ def __init__ (  # pylint: disable=R0913
         # `ner_doc` provides the merged-entity spans from NER
         self.ner_doc: spacy.tokens.Doc = ner_pipe(self.text)
 
-        # `aux_doc` e.g.,  span re-indexing for Spotlight entity linking
-        self.aux_doc: spacy.tokens.Doc = aux_pipe(self.text)
+        # `aux_doc` e.g., re-indexing spans for Spotlight entity linking
+        # NB: this is optional, in case the Spotlight service is down
+        self.aux_doc: typing.Optional[ spacy.tokens.Doc ] = None
+
+        try:
+            self.aux_doc = aux_pipe(self.text)
+        except Exception as ex:  # pylint: disable=W0718
+            ic(ex)
 
         self.kg: KnowledgeGraph = kg  # pylint: disable=C0103
         self.infer_rels: typing.List[ InferRel ] = infer_rels
