diff options
| -rw-r--r-- | python/fatcat_tools/normal.py | 7 | ||||
| -rw-r--r-- | python/tests/files/datacite/datacite_doc_23.json | 2 |
2 files changed, 6 insertions, 3 deletions
diff --git a/python/fatcat_tools/normal.py b/python/fatcat_tools/normal.py index a77c5eb0..7b4bd19c 100644 --- a/python/fatcat_tools/normal.py +++ b/python/fatcat_tools/normal.py @@ -22,7 +22,10 @@ def clean_doi(raw): if not raw: return None raw = raw.strip() - raw = raw.replace('\u2013', '-') # emdash + if '\u2013' in raw: + # Do not attempt to normalize "en dash" and since FC does not allow + # unicode in DOI, treat this as invalid. + return None if len(raw.split()) != 1: return None if raw.startswith("doi:"): @@ -48,7 +51,7 @@ def test_clean_doi(): assert clean_doi("10.1037//0002-9432.72.1.50") == "10.1037/0002-9432.72.1.50" assert clean_doi("10.1037/0002-9432.72.1.50") == "10.1037/0002-9432.72.1.50" assert clean_doi("10.23750/abm.v88i2 -s.6506") == None - assert clean_doi("10.17167/mksz.2017.2.129–155") == "10.17167/mksz.2017.2.129-155" + assert clean_doi("10.17167/mksz.2017.2.129–155") == None assert clean_doi("http://doi.org/10.1234/asdf ") == "10.1234/asdf" assert clean_doi("https://dx.doi.org/10.1234/asdf ") == "10.1234/asdf" assert clean_doi("doi:10.1234/asdf ") == "10.1234/asdf" diff --git a/python/tests/files/datacite/datacite_doc_23.json b/python/tests/files/datacite/datacite_doc_23.json index b755f1a5..8eaa8c21 100644 --- a/python/tests/files/datacite/datacite_doc_23.json +++ b/python/tests/files/datacite/datacite_doc_23.json @@ -23,7 +23,7 @@ "dateType": "Issued" } ], - "doi": "10.7916/d86x0cg1–xxx", + "doi": "10.7916/d86x0cg1-xxx", "isActive": true, "language": "GERMAN", "publicationYear": 2017, |