From a889f3212586ffb85961ad08af32a53e46e0382d Mon Sep 17 00:00:00 2001
From: Bryan Newbold <bnewbold@robocracy.org>
Date: Tue, 28 Jan 2020 19:34:44 -0800
Subject: additional DOI prefix filters

From martin, thanks.
---
 python/fatcat_tools/workers/changelog.py | 8 ++++++++
 1 file changed, 8 insertions(+)

(limited to 'python/fatcat_tools/workers')

diff --git a/python/fatcat_tools/workers/changelog.py b/python/fatcat_tools/workers/changelog.py
index 353eca8f..81cd0b05 100644
--- a/python/fatcat_tools/workers/changelog.py
+++ b/python/fatcat_tools/workers/changelog.py
@@ -93,6 +93,14 @@ class EntityUpdatesWorker(FatcatWorker):
         self.ingest_pdf_doi_prefix_blocklist = [
             # gbif.org: many DOIs, not PDF fulltext
             "10.15468/",
+            # ssl.fao.org: gene data
+            "10.18730/",
+            # plutof.ut.ee: gene data
+            "10.15156/",
+            # ba.e-pics.ethz.ch: swiss image (photo) archive
+            "10.3932/",
+            # ccdc.cam.ac.uk: crystal structures
+            "10.5517/",
         ]
 
     def want_live_ingest(self, release, ingest_request):
-- 
cgit v1.2.3