From af1f65b980a4faf01004b617a7c34f80c380132b Mon Sep 17 00:00:00 2001
From: Bryan Newbold <bnewbold@robocracy.org>
Date: Tue, 21 May 2019 19:15:25 -0700
Subject: arxiv importer robustification and CLI impl

---
 python/fatcat_import.py                | 21 +++++++++++++++++++++
 python/fatcat_tools/importers/arxiv.py | 29 ++++++++++++++++++++---------
 2 files changed, 41 insertions(+), 9 deletions(-)

(limited to 'python')

diff --git a/python/fatcat_import.py b/python/fatcat_import.py
index 94c90ea5..e80c5d5b 100755
--- a/python/fatcat_import.py
+++ b/python/fatcat_import.py
@@ -22,6 +22,15 @@ def run_jalc(args):
         extid_map_file=args.extid_map_file)
     Bs4XmlLinesPusher(ji, args.xml_file, "<rdf:Description").run()
 
+def run_arxiv(args):
+    ari = ArxivRawImporter(args.api,
+        edit_batch_size=args.batch_size)
+    if args.kafka_mode:
+        raise NotImplementedError
+        #KafkaBs4XmlPusher(ari, args.kafka_hosts, args.kafka_env, "api-arxiv", "fatcat-import").run()
+    else:
+        Bs4XmlFilePusher(ari, args.xml_file, "record").run()
+
 def run_orcid(args):
     foi = OrcidImporter(args.api,
         edit_batch_size=args.batch_size)
@@ -164,6 +173,18 @@ def main():
         help="DOI-to-other-identifiers sqlite3 database",
         default=None, type=str)
 
+    sub_arxiv = subparsers.add_parser('arxiv')
+    sub_arxiv.set_defaults(
+        func=run_arxiv,
+        auth_var="FATCAT_AUTH_WORKER_ARXIV",
+    )
+    sub_arxiv.add_argument('xml_file',
+        help="arXivRaw XML file to import from",
+        default=sys.stdin, type=argparse.FileType('r'))
+    sub_arxiv.add_argument('--kafka-mode',
+        action='store_true',
+        help="consume from kafka topic (not stdin)")
+
     sub_orcid = subparsers.add_parser('orcid')
     sub_orcid.set_defaults(
         func=run_orcid,
diff --git a/python/fatcat_tools/importers/arxiv.py b/python/fatcat_tools/importers/arxiv.py
index 03ef10d6..5a33bff1 100644
--- a/python/fatcat_tools/importers/arxiv.py
+++ b/python/fatcat_tools/importers/arxiv.py
@@ -12,6 +12,12 @@ from .crossref import lookup_license_slug
 
 latex2text = LatexNodes2Text()
 
+def latex_to_text(raw):
+    try:
+        return latex2text.latex_to_text(raw).strip()
+    except AttributeError:
+        return raw.strip()
+
 def parse_arxiv_authors(raw):
     if not raw:
         return []
@@ -21,7 +27,7 @@ def parse_arxiv_authors(raw):
         if len(last) == 2:
             authors[-1] = last[0]
             authors.append(last[1])
-    authors = [latex2text.latex_to_text(a).strip() for a in authors]
+    authors = [latex_to_text(a).strip() for a in authors]
     return authors
 
 def test_parse_arxiv_authors():
@@ -67,7 +73,11 @@ class ArxivRawImporter(EntityImporter):
 
     def parse_record(self, record):
 
+        if not record:
+            return None
         metadata = record.arXivRaw
+        if not metadata:
+            return None
         extra = dict()
         extra_arxiv = dict()
 
@@ -76,7 +86,7 @@ class ArxivRawImporter(EntityImporter):
         if metadata.doi and metadata.doi.string:
             doi = metadata.doi.string.lower().strip()
             assert doi.startswith('10.')
-        title = latex2text.latex_to_text(metadata.title.string)
+        title = latex_to_text(metadata.title.string)
         authors = parse_arxiv_authors(metadata.authors.string)
         contribs = [fatcat_client.ReleaseContrib(raw_name=a, role='author') for a in authors]
 
@@ -115,7 +125,7 @@ class ArxivRawImporter(EntityImporter):
             number = metadata.find('report-no').string.strip()
             release_type = "report"
         if metadata.find('acm-class') and metadata.find('acm-class').string:
-            extra_arxiv['acm_class'] = metadata.find('acm_class').string.strip()
+            extra_arxiv['acm_class'] = metadata.find('acm-class').string.strip()
         if metadata.categories and metadata.categories.string:
             extra_arxiv['categories'] = metadata.categories.string.split()
         license_slug = None
@@ -133,7 +143,7 @@ class ArxivRawImporter(EntityImporter):
                 orig = both[1].strip()
             if '$' in abst or '{' in abst:
                 mime = "application/x-latex"
-                abst_plain = latex2text.latex_to_text(abst)
+                abst_plain = latex_to_text(abst)
                 abstracts.append(fatcat_client.ReleaseAbstract(content=abst_plain, mimetype="text/plain", lang="en"))
             else:
                 mime = "text/plain"
@@ -250,9 +260,10 @@ class ArxivRawImporter(EntityImporter):
                         # as a flag to not count below
                         v._updated = True
                     existing = existing_doi
-            
-            v._existing_work_id = existing.work_id
-            any_work_id = existing.work_id
+
+            if existing:
+                v._existing_work_id = existing.work_id
+                any_work_id = existing.work_id
 
         last_edit = None
         for v in versions:
@@ -262,11 +273,11 @@ class ArxivRawImporter(EntityImporter):
                 continue
             if not any_work_id and last_edit:
                 # fetch the last inserted release from this group
-                r = self.api.get_release_rev(last_edit.rev)
+                r = self.api.get_release_revision(last_edit.revision)
                 assert r.work_id
                 any_work_id = r.work_id
             v.work_id = any_work_id
-            last_edit = self.api.insert_release(self.get_editgroup_id(), v)
+            last_edit = self.api.create_release(self.get_editgroup_id(), v)
             self.counts['insert'] += 1
 
         return False
-- 
cgit v1.2.3