diff options
author | Bryan Newbold <bnewbold@archive.org> | 2021-06-30 18:58:20 -0700 |
---|---|---|
committer | Bryan Newbold <bnewbold@archive.org> | 2021-06-30 18:58:22 -0700 |
commit | 5a1d53e8705d5ea59ea0c007f8a53940a353000b (patch) | |
tree | f2dd1f9997e8569ecda641b43466f73fad4aaedf | |
parent | 8a34f3747a6881eab66775388a9b6d7878c00a6c (diff) | |
download | fatcat-scholar-5a1d53e8705d5ea59ea0c007f8a53940a353000b.tar.gz fatcat-scholar-5a1d53e8705d5ea59ea0c007f8a53940a353000b.zip |
HACK: don't parse TEI-XML for a specific paper/file
GROBID v0.5.5 returns TEI-XML for this one PDF which is not valid XML,
due to a text encoding issue.
-rw-r--r-- | fatcat_scholar/transform.py | 6 |
1 files changed, 4 insertions, 2 deletions
diff --git a/fatcat_scholar/transform.py b/fatcat_scholar/transform.py index 6904770..388f2f5 100644 --- a/fatcat_scholar/transform.py +++ b/fatcat_scholar/transform.py @@ -481,7 +481,8 @@ def transform_heavy(heavy: IntermediateBundle) -> Optional[ScholarDoc]: else: raise NotImplementedError(f"doc_type: {heavy.doc_type}") - if heavy.grobid_fulltext: + # TODO: this crude filter should not be necessary once we upgrade to GROBID v0.6+ + if heavy.grobid_fulltext and heavy.grobid_fulltext.get('file_ident') != 'gbbvrg2tpzan5hl3qcsfzh4vfq': fulltext_release = [ r for r in heavy.releases @@ -792,7 +793,8 @@ def refs_from_heavy(heavy: IntermediateBundle) -> Sequence[RefStructured]: break fulltext_refs: List[RefStructured] = [] - if heavy.grobid_fulltext: + # TODO: this crude filter should not be necessary once we upgrade to GROBID v0.6+ + if heavy.grobid_fulltext and heavy.grobid_fulltext.get('file_ident') != 'gbbvrg2tpzan5hl3qcsfzh4vfq': fulltext_release = [ r for r in heavy.releases |