diff options
Diffstat (limited to 'python/scripts/oai2ingestrequest.py')
| -rwxr-xr-x | python/scripts/oai2ingestrequest.py | 19 |
1 files changed, 11 insertions, 8 deletions
diff --git a/python/scripts/oai2ingestrequest.py b/python/scripts/oai2ingestrequest.py index 315b8d2..1f4a19f 100755 --- a/python/scripts/oai2ingestrequest.py +++ b/python/scripts/oai2ingestrequest.py @@ -1,5 +1,4 @@ #!/usr/bin/env python3 - """ Transform an OAI-PMH bulk dump (JSON) into ingest requests. @@ -33,17 +32,19 @@ DOMAIN_BLOCKLIST = [ ] RELEASE_STAGE_MAP = { - 'info:eu-repo/semantics/draftVersion': 'draft', + 'info:eu-repo/semantics/draftVersion': 'draft', 'info:eu-repo/semantics/submittedVersion': 'submitted', - 'info:eu-repo/semantics/acceptedVersion': 'accepted', + 'info:eu-repo/semantics/acceptedVersion': 'accepted', 'info:eu-repo/semantics/publishedVersion': 'published', - 'info:eu-repo/semantics/updatedVersion': 'updated', + 'info:eu-repo/semantics/updatedVersion': 'updated', } + def canon(s): parsed = urlcanon.parse_url(s) return str(urlcanon.whatwg(parsed)) + def transform(obj): """ Transforms from a single OAI-PMH object to zero or more ingest requests. @@ -112,6 +113,7 @@ def transform(obj): return requests + def run(args): for l in args.json_file: if not l.strip(): @@ -122,17 +124,18 @@ def run(args): for r in requests: print("{}".format(json.dumps(r, sort_keys=True))) + def main(): - parser = argparse.ArgumentParser( - formatter_class=argparse.ArgumentDefaultsHelpFormatter) + parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter) parser.add_argument('json_file', - help="OAI-PMH dump file to use (usually stdin)", - type=argparse.FileType('r')) + help="OAI-PMH dump file to use (usually stdin)", + type=argparse.FileType('r')) subparsers = parser.add_subparsers() args = parser.parse_args() run(args) + if __name__ == '__main__': main() |