Link citations cannot be disambiguated

cl/citations/annotate_citations.py

@@ -2,9 +2,13 @@
 import re
 from typing import Dict, List
 
+from django.urls import reverse
 from eyecite import annotate_citations, clean_text
 
-from cl.citations.match_citations import NO_MATCH_RESOURCE
+from cl.citations.match_citations import (
+    MULTIPLE_MATCHES_RESOURCE,
+    NO_MATCH_RESOURCE,
+)
 from cl.citations.types import MatchedResourceType, SupportedCitationType
 from cl.custom_filters.templatetags.text_filters import best_case_name
 from cl.lib.string_utils import trunc
@@ -62,6 +66,13 @@ def generate_annotations(
                 '<span class="citation no-link">',
                 "</span>",
             ]
+        elif opinion is MULTIPLE_MATCHES_RESOURCE:  # If multiple matches,
+            # can't disambiguate...
+            annotation = [
+                '<span class="citation multiple-matches">'
+                f'<a href="{reverse("citation_redirector", args=[citations[0].groups["reporter"], citations[0].groups["volume"], citations[0].groups["page"]])}">',
+                "</a></span>",
+            ]
         else:  # If successfully matched...
             case_name = trunc(best_case_name(opinion.cluster), 60, "...")
             safe_case_name = html.escape(case_name)

cl/citations/match_citations.py

@@ -28,6 +28,11 @@
 
 
 NO_MATCH_RESOURCE = Resource(case_citation(source_text="UNMATCHED_CITATION"))
+MULTIPLE_MATCHES_RESOURCE = Resource(
+    case_citation(
+        source_text="MULTIPLE_MATCHES", page="999999", volume="999999"
+    )
+)
 
 
 def filter_by_matching_antecedent(
@@ -56,6 +61,11 @@ def resolve_fullcase_citation(
     if type(full_citation) is FullCaseCitation:
         db_search_results: list[Hit]
         db_search_results, _ = es_search_db_for_full_citation(full_citation)
+        # If there is more than one result, return a placeholder with the
+        # citation with multiple results
+        if len(db_search_results) > 1:
+            return MULTIPLE_MATCHES_RESOURCE
+
         # If there is one search result, try to return it
         if len(db_search_results) == 1:
             result_id = db_search_results[0]["id"]

cl/citations/match_citations_queries.py

@@ -214,9 +214,9 @@ def es_search_db_for_full_citation(
                 full_citation,
                 full_citation.citing_opinion,
             )
-            return results, citation_found
-    # Give up.
-    return [], citation_found
+
+    # Return all possible results
+    return results, citation_found
 
 
 def es_get_query_citation(
@@ -244,6 +244,7 @@ def es_get_query_citation(
             missing_citations.append(citation)
 
     if len(citations) == 1 and matches and len(matches) == 1:
-        # If more than one match, don't show the tip
+        # If only one match, show the tip
         return matches[0], missing_citations
-    return matches, missing_citations
+    # No exact match, don't show the tip
+    return None, missing_citations

cl/citations/tasks.py

@@ -18,6 +18,7 @@
     is_parenthetical_descriptive,
 )
 from cl.citations.match_citations import (
+    MULTIPLE_MATCHES_RESOURCE,
     NO_MATCH_RESOURCE,
     do_resolve_citations,
 )
@@ -163,6 +164,9 @@ def store_opinion_citations_and_update_parentheticals(
     # Delete the unmatched citations
     citation_resolutions.pop(NO_MATCH_RESOURCE, None)
 
+    # Delete citations with multiple matches
+    citation_resolutions.pop(MULTIPLE_MATCHES_RESOURCE, None)
+
     # Increase the citation count for the cluster of each matched opinion
     # if that cluster has not already been cited by this opinion. First,
     # calculate a list of the IDs of every opinion whose cluster will need

cl/citations/tests.py

@@ -42,6 +42,7 @@
     get_representative_parenthetical,
 )
 from cl.citations.match_citations import (
+    MULTIPLE_MATCHES_RESOURCE,
     NO_MATCH_RESOURCE,
     do_resolve_citations,
     resolve_fullcase_citation,
@@ -503,6 +504,7 @@ def setUpTestData(cls) -> None:
         # Courts
         cls.court_scotus = CourtFactory(id="scotus")
         court_ca1 = CourtFactory(id="ca1")
+        cls.court_ca5 = CourtFactory(id="ca5")
 
         # Citation 1
         cls.citation1 = CitationWithParentsFactory.create(
@@ -579,6 +581,29 @@ def setUpTestData(cls) -> None:
                 ),
             ),
         )
+
+        cls.citation6 = CitationWithParentsFactory.create(
+            volume="114",
+            reporter="F.3d",
+            page="1182",
+            cluster=OpinionClusterFactoryWithChildrenAndParents(
+                docket=DocketFactory(court=cls.court_ca5),
+                case_name="Foo v. Bar",
+                date_filed=date(1997, 4, 10),
+            ),
+        )
+
+        cls.citation7 = CitationWithParentsFactory.create(
+            volume="114",
+            reporter="F.3d",
+            page="1182",
+            cluster=OpinionClusterFactoryWithChildrenAndParents(
+                docket=DocketFactory(court=cls.court_ca5),
+                case_name="Lorem v. Ipsum",
+                date_filed=date(1997, 4, 8),
+            ),
+        )
+
         call_command(
             "cl_index_parent_and_child_docs",
             search_type=SEARCH_TYPES.OPINION,
@@ -897,6 +922,15 @@ def test_citation_matching_issue621(self) -> None:
         results = resolve_fullcase_citation(citation)
         self.assertEqual(NO_MATCH_RESOURCE, results)
 
+    def test_citation_multiple_matches(self) -> None:
+        """Make sure that we can identify multiple matches for a single citation"""
+        citation_str = "114 F.3d 1182"
+        citation = get_citations(citation_str, tokenizer=HYPERSCAN_TOKENIZER)[
+            0
+        ]
+        results = resolve_fullcase_citation(citation)
+        self.assertEqual(MULTIPLE_MATCHES_RESOURCE, results)
+
     def test_citation_increment(self) -> None:
         """Make sure that found citations update the increment on the cited
         opinion's citation count"""