datacite: skip empty abstracts

Do not add abstracts where `clean` results in the empty string - this
violates a constraint: `either abstract_sha1 or content is required`

4 files changed, 95 insertions, 2 deletions

diff --git a/python/fatcat_tools/importers/datacite.py b/python/fatcat_tools/importers/datacite.py
index 703dbc27..eb49596f 100644
--- a/python/fatcat_tools/importers/datacite.py
+++ b/python/fatcat_tools/importers/datacite.py
@@ -549,10 +549,13 @@ class DataciteImporter(EntityImporter):
                 lang = langdetect.detect(text)
             except (langdetect.lang_detect_exception.LangDetectException, TypeError) as err:
                 print('[{}] language detection failed with {} on {}'.format(doi, err, text), file=sys.stderr)
+            abstract_text = clean(text)
+            if not abstract_text:
+                continue
             abstracts.append(
                 fatcat_openapi_client.ReleaseAbstract(
                     mimetype="text/plain",
-                    content=clean(text),
+                    content=abstract_text,
                     lang=lang,
                 ))
 
diff --git a/python/tests/files/datacite/datacite_doc_36.json b/python/tests/files/datacite/datacite_doc_36.json
new file mode 100644
index 00000000..66aba00c
--- /dev/null
+++ b/python/tests/files/datacite/datacite_doc_36.json
@@ -0,0 +1,65 @@
+{
+  "id": "10.17912/micropub.biology.000143",
+  "type": "dois",
+  "attributes": {
+    "doi": "10.17912/micropub.biology.000143",
+    "identifiers": null,
+    "creators": [
+      {
+        "name": "Paul Katz",
+        "givenName": "",
+        "familyName": "",
+        "affiliation": [],
+        "role": "author"
+      }
+    ],
+    "titles": [
+      {
+        "lang": "da"
+      },
+      {
+        "title": "Sample"
+      }
+    ],
+    "publisher": "microPublication Biology",
+    "publicationYear": 2019,
+    "types": {
+      "resourceTypeGeneral": "DataPaper"
+    },
+    "relatedIdentifiers": [],
+    "sizes": [],
+    "formats": [],
+    "version": null,
+    "rightsList": [],
+    "descriptions": [
+      {
+        "description": "                 ",
+        "descriptionType": "Abstract"
+      }
+    ],
+    "geoLocations": [],
+    "fundingReferences": [],
+    "url": "https://www.micropublication.org/journals/biology/micropub.biology.000143",
+    "created": "2019-08-19T14:43:08.000Z",
+    "registered": "2019-08-19T14:43:09.000Z",
+    "published": "2019",
+    "updated": "2019-11-09T12:32:02.000Z",
+    "contributors": [
+      {
+        "name": "Paul Katz",
+        "givenName": "",
+        "familyName": "",
+        "affiliation": [],
+        "role": "illustrator"
+      }
+    ]
+  },
+  "relationships": {
+    "client": {
+      "data": {
+        "id": "caltech.micropub",
+        "type": "clients"
+      }
+    }
+  }
+}
diff --git a/python/tests/files/datacite/datacite_result_36.json b/python/tests/files/datacite/datacite_result_36.json
new file mode 100644
index 00000000..8c958848
--- /dev/null
+++ b/python/tests/files/datacite/datacite_result_36.json
@@ -0,0 +1,25 @@
+{
+  "abstracts": [],
+  "contribs": [
+    {
+      "given_name": "",
+      "index": 0,
+      "raw_name": "Paul Katz",
+      "role": "author"
+    }
+  ],
+  "ext_ids": {
+    "doi": "10.17912/micropub.biology.000143"
+  },
+  "extra": {
+    "datacite": {
+      "resourceTypeGeneral": "DataPaper"
+    },
+    "container_name": "microPublication Biology"
+  },
+  "refs": [],
+  "release_stage": "published",
+  "release_year": 2019,
+  "publisher": "microPublication Biology",
+  "title": "Sample"
+}
diff --git a/python/tests/import_datacite.py b/python/tests/import_datacite.py
index 8b6797ef..edbb6617 100644
--- a/python/tests/import_datacite.py
+++ b/python/tests/import_datacite.py
@@ -400,7 +400,7 @@ def test_datacite_conversions(datacite_importer):
     for now.
     """
     datacite_importer.debug = True
-    for i in range(36):
+    for i in range(37):
         src = "tests/files/datacite/datacite_doc_{0:02d}.json".format(i)
         dst = "tests/files/datacite/datacite_result_{0:02d}.json".format(i)
         with open(src, "r") as f: