From 3f8ead3d07bed78c750b9f6a8b7e95ebffeff089 Mon Sep 17 00:00:00 2001
From: Bryan Newbold <bnewbold@archive.org>
Date: Tue, 18 Feb 2020 19:02:48 -0800
Subject: unpaywall2ingestrequest transform script

---
 python/sandcrawler/pdftrio.py             |   2 +-
 python/scripts/unpaywall2ingestrequest.py | 103 ++++++++++++++++++++++++++++++
 2 files changed, 104 insertions(+), 1 deletion(-)
 create mode 100755 python/scripts/unpaywall2ingestrequest.py

(limited to 'python')

diff --git a/python/sandcrawler/pdftrio.py b/python/sandcrawler/pdftrio.py
index 7a2e53c..12be9eb 100644
--- a/python/sandcrawler/pdftrio.py
+++ b/python/sandcrawler/pdftrio.py
@@ -180,7 +180,7 @@ class PdfTrioBlobWorker(SandcrawlerWorker):
         result = dict()
         result['file_meta'] = gen_file_metadata(blob)
         result['key'] = result['file_meta']['sha1hex']
-        result['pdf_trio'] = self.pdftrio_client.classify_pdf(blob, mode=mode)
+        result['pdf_trio'] = self.pdftrio_client.classify_pdf(blob, mode=self.mode)
         result['timing'] = dict(
             pdftrio_sec=result['pdf_trio'].pop('_total_sec', None),
             total_sec=time.time() - start_process,
diff --git a/python/scripts/unpaywall2ingestrequest.py b/python/scripts/unpaywall2ingestrequest.py
new file mode 100755
index 0000000..c51a152
--- /dev/null
+++ b/python/scripts/unpaywall2ingestrequest.py
@@ -0,0 +1,103 @@
+#!/usr/bin/python3
+
+"""
+Transform an unpaywall dump (JSON) into ingest requests.
+"""
+
+import sys
+import json
+import argparse
+
+DOMAIN_BLOCKLIST = [
+    # large OA publishers (we get via DOI)
+
+    # large repos and aggregators (we crawl directly)
+    "://arxiv.org/",
+    "://europepmc.org/",
+    "ncbi.nlm.nih.gov/",
+    "semanticscholar.org/",
+    "://doi.org/",
+    "zenodo.org/",
+    "figshare.com/",
+    "://archive.org/",
+    ".archive.org/",
+]
+
+RELEASE_STAGE_MAP = {
+    'draftVersion':     'draft',
+    'submittedVersion': 'submitted',
+    'acceptedVersion':  'accepted',
+    'publishedVersion': 'published',
+    'updatedVersion':   'updated',
+}
+
+
+def transform(obj):
+    """
+    Transforms from a single unpaywall object to zero or more ingest requests.
+    Returns a list of dicts.
+    """
+
+    requests = []
+    if not obj['doi'].startswith('10.'):
+        return requests
+    if not obj['oa_locations']:
+        return requests
+
+    for location in obj['oa_locations']:
+        if not location['url_for_pdf']:
+            continue
+        skip = False
+        for domain in DOMAIN_BLOCKLIST:
+            if domain in location['url_for_pdf']:
+                skip = True
+        if skip:
+            continue
+
+        request = {
+            'base_url': location['url_for_pdf'],
+            'ingest_type': 'pdf',
+            'link_source': 'unpaywall',
+            'link_source_id': obj['doi'].lower(),
+            'ingest_request_source': 'unpaywall',
+            'release_stage': RELEASE_STAGE_MAP.get(location['version']),
+            'rel': location['host_type'],
+            'ext_ids': {
+                'doi': obj['doi'].lower(),
+            },
+            'edit_extra': {},
+        }
+        if obj.get('oa_status'):
+            request['edit_extra']['oa_status'] = obj['oa_status']
+        if location.get('evidence'):
+            request['edit_extra']['evidence'] = location['evidence']
+        if location['pmh_id']:
+            request['ext_ids']['pmh_id'] = location['pmh_id']
+        requests.append(request)
+
+    return requests
+
+def run(args):
+    for l in args.json_file:
+        if not l.strip():
+            continue
+        row = json.loads(l)
+
+        requests = transform(row) or []
+        for r in requests:
+            print("{}".format(json.dumps(r, sort_keys=True)))
+
+def main():
+    parser = argparse.ArgumentParser(
+        formatter_class=argparse.ArgumentDefaultsHelpFormatter)
+    parser.add_argument('json_file',
+        help="unpaywall dump file to use",
+        type=argparse.FileType('r'))
+    subparsers = parser.add_subparsers()
+
+    args = parser.parse_args()
+
+    run(args)
+
+if __name__ == '__main__':
+    main()
-- 
cgit v1.2.3