From 1305a503b116e8f8081cbc48dc14c92b4b64b17d Mon Sep 17 00:00:00 2001 From: Bryan Newbold Date: Tue, 29 Jan 2019 16:58:04 -0800 Subject: crossref import tweaks/fixes - refs: article-title not title; save unstructured; authors not author - save 'language' field (already an ISO code) --- python/fatcat_tools/importers/crossref.py | 16 +++++++++------- python/tests/files/crossref-works.single.json | 1 + python/tests/import_crossref.py | 3 ++- 3 files changed, 12 insertions(+), 8 deletions(-) diff --git a/python/fatcat_tools/importers/crossref.py b/python/fatcat_tools/importers/crossref.py index c4e55962..f9a4fe46 100644 --- a/python/fatcat_tools/importers/crossref.py +++ b/python/fatcat_tools/importers/crossref.py @@ -239,11 +239,13 @@ class CrossrefImporter(EntityImporter): ref_extra['journal-title'] = rm['journal-title'] if rm.get('DOI'): ref_extra['doi'] = rm.get('DOI').lower() - # TODO: what fields here? CSL citation stuff - for k in ('author', 'editor', 'edition', 'authority', 'version', - 'genre', 'url', 'event', 'issue', 'volume', 'date', - 'accessed_date', 'issued', 'page', 'medium', - 'collection_title', 'chapter_number'): + author = clean(rm.get('author')) + if author: + ref_extra['authors'] = [author] + for k in ('editor', 'edition', 'authority', 'version', 'genre', + 'url', 'event', 'issue', 'volume', 'date', 'accessed_date', + 'issued', 'page', 'medium', 'collection_title', 'chapter_number', + 'unstructured', 'series-title', 'volume-title'): if clean(rm.get(k)): ref_extra[k] = clean(rm[k]) if not ref_extra: @@ -255,7 +257,7 @@ class CrossrefImporter(EntityImporter): key=key, year=year, container_name=clean(container_name), - title=clean(rm.get('title')), + title=clean(rm.get('article-title')), locator=clean(rm.get('first-page')), # TODO: just dump JSON somewhere here? extra=ref_extra)) @@ -378,7 +380,7 @@ class CrossrefImporter(EntityImporter): volume=clean(obj.get('volume')), issue=clean(obj.get('issue')), pages=clean(obj.get('page')), - language=None, # crossref doesn't supply language info + language=clean(obj.get('language')), license_slug=license_slug, extra=extra, abstracts=abstracts, diff --git a/python/tests/files/crossref-works.single.json b/python/tests/files/crossref-works.single.json index f00c2142..2b775647 100644 --- a/python/tests/files/crossref-works.single.json +++ b/python/tests/files/crossref-works.single.json @@ -15,6 +15,7 @@ "$numberLong": "1508768117199" } }, + "language": "fr", "reference-count": 25, "publisher": "Wiley-Blackwell", "issue": "4", diff --git a/python/tests/import_crossref.py b/python/tests/import_crossref.py index cb2143b3..d439137d 100644 --- a/python/tests/import_crossref.py +++ b/python/tests/import_crossref.py @@ -82,6 +82,7 @@ def test_crossref_dict_parse(crossref_importer): assert r.license_slug == "CC-BY-NC-ND" assert r.original_title == "Renormalized perturbation theory auf deutsch" assert r.isbn13 == "978-3-16-148410-0" + assert r.language == "fr" assert 'subtitle' not in r.extra assert 'subtitle' not in r.extra['crossref'] assert 'funder' not in r.extra @@ -107,7 +108,7 @@ def test_crossref_dict_parse(crossref_importer): assert r.refs[0].year == 1972 assert r.refs[0].locator == "1734" assert r.refs[0].container_name == "J. Chem. Phys." - assert r.refs[0].extra == {"volume": "57", "author": "Swenson", "doi": "10.1063/1.1678462", "medium": "DVD"} + assert r.refs[0].extra == {"volume": "57", "authors": ["Swenson"], "doi": "10.1063/1.1678462", "medium": "DVD"} assert r.refs[2].key == 'BIB3' assert r.refs[2].extra.get('author') is None assert r.refs[2].container_name == "Hypervirial Theorem's, Lecture Notes in Chemistry <3" -- cgit v1.2.3