fix(annotate_citations): Tweaks to PR

Add comment on order and tweak test file to reflect xml ingestiong
freelawproject · Dec 12, 2024 · b27a8bd · b27a8bd
1 parent 8423c77
commit b27a8bd
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 1 deletion.
diff --git a/cl/citations/annotate_citations.py b/cl/citations/annotate_citations.py
@@ -16,6 +16,8 @@ def get_and_clean_opinion_text(document: Opinion | RECAPDocument) -> None:
 
     :param document: The Opinion or RECAPDocument whose text should be parsed
     """
+
+    # We prefer CAP data (xml_harvard) first.
     for attr in [
         "xml_harvard",
         "html_anon_2020",

diff --git a/cl/citations/tests.py b/cl/citations/tests.py
@@ -290,7 +290,7 @@ def test_make_html_from_harvard_xml(self) -> None:
                 s=s,
                 expected_html=expected_html,
             ):
-                opinion = Opinion(html=s)
+                opinion = Opinion(xml_harvard=s)
                 get_and_clean_opinion_text(opinion)
                 citations = get_citations(
                     opinion.cleaned_text, tokenizer=HYPERSCAN_TOKENIZER