aboutsummaryrefslogtreecommitdiffstats
path: root/python/fatcat_tools
diff options
context:
space:
mode:
authorBryan Newbold <bnewbold@robocracy.org>2020-08-10 17:27:26 -0700
committerBryan Newbold <bnewbold@robocracy.org>2021-11-10 13:04:44 -0800
commitc133f3077aa975aa4706a8e5ca894fc1b71fbc67 (patch)
treee0d9d0e31482c2ec92a025bdb871c34447a686b7 /python/fatcat_tools
parent23fd36a3e8505c1ed6d13367a3fb62a8bf2242d7 (diff)
downloadfatcat-c133f3077aa975aa4706a8e5ca894fc1b71fbc67.tar.gz
fatcat-c133f3077aa975aa4706a8e5ca894fc1b71fbc67.zip
datacite import: store less subject metadata
Many of these 'subject' objects have the equivalent of several lines of text, with complex URLs that don't compress well. I think it is fine we have included these thus far instead of parsing more deeply, but going forward I don't think this nested 'extra' metadata is worth the database space.
Diffstat (limited to 'python/fatcat_tools')
-rw-r--r--python/fatcat_tools/importers/datacite.py8
1 files changed, 7 insertions, 1 deletions
diff --git a/python/fatcat_tools/importers/datacite.py b/python/fatcat_tools/importers/datacite.py
index d4d7a9f5..fe02cac4 100644
--- a/python/fatcat_tools/importers/datacite.py
+++ b/python/fatcat_tools/importers/datacite.py
@@ -597,7 +597,13 @@ class DataciteImporter(EntityImporter):
if license_extra:
extra_datacite["license"] = license_extra
if attributes.get("subjects"):
- extra_datacite["subjects"] = attributes["subjects"]
+ # these subjects with schemeUri are too much metadata, which
+ # doesn't compress. filter them out.
+ extra_subjects = [
+ subj for subj in attributes["subjects"] if not subj.get("schemeUri")
+ ]
+ if extra_subjects:
+ extra_datacite["subjects"] = extra_subjects
# Include version information.
metadata_version = attributes.get("metadataVersion") or ""