Closed jmartinm closed 7 years ago
@jmartinm Do we have tests for this kind of extraction?
I don't think so, the only thing I see is this mock in the integration test for workflows.
Unit tests are needed for https://github.com/inspirehep/inspire-next/blob/d0cb6ba6d761279a76b849990dadd8160156eecb/inspirehep/modules/refextract/tasks.py#L86
But, aren't we anyway discarding references before sending to legacy?
But, aren't we anyway discarding references before sending to legacy?
I guess they were just extracted to show something in the Holding Pen that helps decide if the article is core/non-core.
Question on this PR was to know if we want to keep doing that or we can extract the references on Labs (and possibly as well the ones given as free text by the user) and send them to legacy.
Nope we can't yet. We need to port refextract to use journal from Labs, but for the time being this has not yet been done: https://github.com/inspirehep/refextract/issues/3
@jmartinm if nobody is doing this, move to ready and unassign so anyone can pick it up :+1:
This has happened in https://github.com/inspirehep/inspire-next/pull/2558.
Current article workflow extracts references for arXiv papers:
TODO:
arxiv_refextract
task needs to be reviewed to see if references are being added into the correct reference fields using latesthep
schema.BONUS POINTS:
refextract
to extract references when the user inputs them in free form (in the references textarea)