diff options
| author | Bryan Newbold <bnewbold@archive.org> | 2021-07-01 16:36:27 -0700 |
|---|---|---|
| committer | Bryan Newbold <bnewbold@archive.org> | 2021-07-01 16:36:27 -0700 |
| commit | 3b4d87826eea9b827fdd8b824e6cd738d89db5bd (patch) | |
| tree | 985f29c7c21bb732209ca66fa58123125248940f /fatcat_scholar/transform.py | |
| parent | 5a1d53e8705d5ea59ea0c007f8a53940a353000b (diff) | |
| download | fatcat-scholar-3b4d87826eea9b827fdd8b824e6cd738d89db5bd.tar.gz fatcat-scholar-3b4d87826eea9b827fdd8b824e6cd738d89db5bd.zip | |
refs: clean up GROBID DOIs and PMCIDs
Diffstat (limited to 'fatcat_scholar/transform.py')
| -rw-r--r-- | fatcat_scholar/transform.py | 5 |
1 files changed, 3 insertions, 2 deletions
diff --git a/fatcat_scholar/transform.py b/fatcat_scholar/transform.py index 388f2f5..f9616c4 100644 --- a/fatcat_scholar/transform.py +++ b/fatcat_scholar/transform.py @@ -12,6 +12,7 @@ from fatcat_scholar.api_entities import * from fatcat_scholar.schema import * from fatcat_scholar.config import settings, GIT_REVISION from fatcat_scholar.grobid2json import teixml2json +from fatcat_scholar.identifiers import clean_doi, clean_pmcid MAX_BODY_CHARS = 512 * 1024 @@ -631,9 +632,9 @@ def refs_from_grobid(release: ReleaseEntity, tei_dict: dict) -> List[RefStructur volume=ref.get("volume"), issue=ref.get("issue"), pages=ref.get("pages"), - doi=ref.get("doi"), + doi=clean_doi(ref.get("doi")), pmid=ref.get("pmid"), - pmcid=ref.get("pmcid"), + pmcid=clean_pmcid(ref.get("pmcid")), arxiv_id=ref.get("arxiv_id"), # isbn13: Optional[str] url=clean_url_conservative(ref.get("url")), |