From 5e79faaa4e741ca15f9f12306a81496aaccc6705 Mon Sep 17 00:00:00 2001
From: Martin Czygan <martin.czygan@gmail.com>
Date: Wed, 14 Apr 2021 22:33:34 +0200
Subject: tests: run es tests against public search endpoint

---
 tests/test_matching.py | 39 +++++++++++++++++++++++++++++++--------
 1 file changed, 31 insertions(+), 8 deletions(-)

(limited to 'tests')

diff --git a/tests/test_matching.py b/tests/test_matching.py
index 7d8b7af..48f7b3a 100644
--- a/tests/test_matching.py
+++ b/tests/test_matching.py
@@ -1,4 +1,5 @@
 from fuzzycat.matching import anything_to_entity, match_release_fuzzy
+from fuzzycat.entities import entity_from_dict
 from fatcat_openapi_client import ReleaseEntity
 import pytest
 import elasticsearch
@@ -17,14 +18,36 @@ def es_client():
     return elasticsearch.Elasticsearch([FATCAT_SEARCH_URL])
 
 
-@pytest.mark.skip(reason="we cannot use POST on es, which client uses: https://git.io/JLssk")
-def test_match_release_fuzzy(es_client):
-    cases = (("wtv64ahbdzgwnan7rllwr3nurm", 2), )
-    for case, count in cases:
-        entity = anything_to_entity(case, ReleaseEntity)
-        logger.info(entity.title)
+def test_match_release_fuzzy(es_client, caplog):
+    cases = (
+        ("wtv64ahbdzgwnan7rllwr3nurm", 1),
+        ("eqcgtpav3na5jh56o5vjsvb4ei", 1),
+    )
+    for i, (ident, count) in enumerate(cases):
+        entity = anything_to_entity(ident, ReleaseEntity)
 
         result = match_release_fuzzy(entity, es=es_client)
-        logger.info("given: {}".format(entity.title))
-        logger.info("found: {}".format(len(result)))
+        logger.info("[{}] given {}, found {}".format(i, entity.title, len(result)))
+        assert len(result) == count
+
+    # Partial data.
+    cases = (
+        ({
+            "title": "digital libraries",
+            "ext_ids": {}
+        }, 5),
+        ({
+            "title": "The Future of Digital Scholarship",
+            "contribs": [{
+                "raw_name": "Costantino Thanos"
+            }],
+            "ext_ids": {}
+        }, 5),
+    )
+    for i, (doc, count) in enumerate(cases):
+        entity = entity_from_dict(doc, ReleaseEntity)
+        result = match_release_fuzzy(entity, es=es_client)
+        with caplog.at_level(logging.INFO):
+            logging.info("[{}] given {}, found {}, {}".format(i, entity.title, len(result),
+                                                              [v.title for v in result]))
         assert len(result) == count
-- 
cgit v1.2.3