From c613297265f101e31fe915ac74bc80b43f6ffe00 Mon Sep 17 00:00:00 2001
From: Bryan Newbold <bnewbold@archive.org>
Date: Sun, 25 Jul 2021 15:09:27 -0700
Subject: refs transform: many fixes

- include year correctly (many cases)
- test coverage for Crossref transform
- pass-through 'edition' as 'version'
- series-title parsed in to title or container as appropriate
- missing release stage
- fix 0-index vs. 1-index ref index field
---
 tests/test_refs_transform.py | 50 +++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 49 insertions(+), 1 deletion(-)

(limited to 'tests/test_refs_transform.py')

diff --git a/tests/test_refs_transform.py b/tests/test_refs_transform.py
index 5b48396..7cd46e8 100644
--- a/tests/test_refs_transform.py
+++ b/tests/test_refs_transform.py
@@ -1,7 +1,8 @@
+import json
 from fatcat_openapi_client import ReleaseEntity
 
 from fatcat_scholar.grobid2json import teixml2json
-from fatcat_scholar.transform import refs_from_grobid
+from fatcat_scholar.transform import refs_from_grobid, refs_from_crossref
 
 
 def test_transform_refs_grobid() -> None:
@@ -40,3 +41,50 @@ def test_transform_refs_grobid() -> None:
         ref.biblio.unstructured
         == "Tasa K, Baker R, Murray M. Using patient feedback for qua- lity improvement. Quality Management in Health Care 1996;8:206-19."
     )
+
+def test_transform_refs_crossref() -> None:
+
+    with open("tests/files/example_crossref_record.json", "r") as f:
+        record = json.loads(f.read())
+
+    dummy_release = ReleaseEntity(
+        ident="releasedummy22222222222222",
+        work_id="workdummy22222222222222222",
+        release_year=1234,
+        release_stage="accepted",
+        ext_ids={},
+    )
+
+    refs = refs_from_crossref(dummy_release, record)
+
+    assert refs[0].release_ident == "releasedummy22222222222222"
+    assert refs[0].work_ident == "workdummy22222222222222222"
+    assert refs[0].release_stage == "accepted"
+    assert refs[0].release_year == 1234
+    assert refs[0].ref_source == "crossref"
+    assert refs[0].key == "BIB0001|his12200-cit-0001"
+    assert refs[0].index == 1
+    assert refs[0].locator is None
+    assert refs[0].biblio.contrib_raw_names is not None
+    assert refs[0].biblio.contrib_raw_names[0] == "Churg"
+    assert refs[0].biblio.container_name == "Arch. Pathol. Lab. Med."
+    assert refs[0].biblio.title == "The separation of benign and malignant mesothelial proliferations"
+    assert refs[0].biblio.year == 2012
+    assert refs[0].biblio.pages == "1217"
+    assert refs[0].biblio.volume == "136"
+    assert refs[0].biblio.doi == "10.5858/arpa.2012-0112-RA"
+    assert refs[0].biblio.unstructured is None
+
+    assert refs[6].biblio.title == "Advances in Laser Remote Sensing – Selected Papers Presented at the 20th International Laser Radar Conference"
+    assert refs[6].biblio.year == 2001
+
+    assert refs[7].key == "CIT0041"
+    assert refs[7].biblio.unstructured == "Linda Weiss,Creating Capitalism. Oxford: Blackwell, 1988. 272 pp. £29.95. ISBN 0 631 15733 6."
+
+
+    assert refs[8].key == "576_CR3"
+    assert refs[8].biblio.unstructured is not None
+    assert refs[8].biblio.title == "The NURBS Book, Monographs in Visual Communication"
+    assert refs[8].biblio.year == 1997
+    assert refs[8].biblio.version == "2"
+
-- 
cgit v1.2.3