1 files changed, 35 insertions, 34 deletions
diff --git a/python/scripts/import_grobid_metadata.py b/python/scripts/import_grobid_metadata.py
index 3d2e14c..f941881 100755
--- a/python/scripts/import_grobid_metadata.py
+++ b/python/scripts/import_grobid_metadata.py
@@ -1,69 +1,67 @@
 #!/usr/bin/env python3
 
-import sys
-import json
 import datetime
+import json
+import sys
+
+MAX_ABSTRACT_BYTES = 4096
 
-MAX_ABSTRACT_BYTES=4096
 
 def parse_grobid_json(obj):
 
-    if not obj.get('title'):
+    if not obj.get("title"):
         return None
 
     extra = dict()
 
-    if obj.get('abstract') and len(obj.get('abstract')) < MAX_ABSTRACT_BYTES:
-        abobj = dict(
-            mimetype="text/plain",
-            language=None,
-            content=obj.get('abstract').strip())
+    if obj.get("abstract") and len(obj.get("abstract")) < MAX_ABSTRACT_BYTES:
+        abobj = dict(mimetype="text/plain", language=None, content=obj.get("abstract").strip())
         abstracts = [abobj]
     else:
         abstracts = None
 
     contribs = []
-    for a in obj.get('authors', []):
+    for a in obj.get("authors", []):
         c = dict(raw_name=a, role="author")
         contribs.append(c)
 
     refs = []
-    for raw in obj.get('citations', []):
+    for raw in obj.get("citations", []):
         extra = dict()
         ref = dict()
-        ref['key'] = raw.get('id')
-        if raw.get('title'):
-            ref['title'] = raw['title'].strip()
-        if raw.get('date'):
+        ref["key"] = raw.get("id")
+        if raw.get("title"):
+            ref["title"] = raw["title"].strip()
+        if raw.get("date"):
             try:
-                year = int(raw['date'].strip()[:4])
-                ref['year'] = year
+                year = int(raw["date"].strip()[:4])
+                ref["year"] = year
             except:
                 pass
-        for key in ('volume', 'url', 'issue', 'publisher'):
+        for key in ("volume", "url", "issue", "publisher"):
             if raw.get(key):
                 extra[key] = raw[key].strip()
-        if raw.get('authors'):
-            extra['authors'] = [a['name'] for a in raw['authors']]
+        if raw.get("authors"):
+            extra["authors"] = [a["name"] for a in raw["authors"]]
         if extra:
             extra = dict(grobid=extra)
         else:
             extra = None
-        ref['extra'] = extra
+        ref["extra"] = extra
         refs.append(ref)
 
     release_type = "journal-article"
     release_date = None
-    if obj.get('date'):
+    if obj.get("date"):
         # TODO: only returns year, ever? how to handle?
-        release_date = datetime.datetime(year=obj['date'], month=1, day=1)
+        release_date = datetime.datetime(year=obj["date"], month=1, day=1)
 
-    if obj.get('doi'):
-        extra['doi'] = obj['doi']
-    if obj['journal'].get('name'):
-        extra['container_name'] = obj['journal']['name']
+    if obj.get("doi"):
+        extra["doi"] = obj["doi"].lower()
+    if obj["journal"].get("name"):
+        extra["container_name"] = obj["journal"]["name"]
 
-    extra['is_longtail_oa'] = True
+    extra["is_longtail_oa"] = True
 
     # TODO: ISSN/eISSN handling? or just journal name lookup?
 
@@ -73,15 +71,17 @@ def parse_grobid_json(obj):
         extra = None
 
     return dict(
-        title=obj['title'].strip(),
+        title=obj["title"].strip(),
         contribs=contribs,
-        publisher=obj['journal'].get('publisher'),
-        volume=obj['journal'].get('volume'),
-        issue=obj['journal'].get('issue'),
+        publisher=obj["journal"].get("publisher"),
+        volume=obj["journal"].get("volume"),
+        issue=obj["journal"].get("issue"),
         abstracts=abstracts,
         release_type=release_type,
         release_date=release_date,
-        extra=extra)
+        extra=extra,
+    )
+
 
 def run():
     for line in sys.stdin:
@@ -90,5 +90,6 @@ def run():
         if out:
             print(out)
 
-if __name__=="__main__":
+
+if __name__ == "__main__":
     run()