diff options
author | Martin Czygan <martin.czygan@gmail.com> | 2021-06-01 16:16:59 +0200 |
---|---|---|
committer | Martin Czygan <martin.czygan@gmail.com> | 2021-06-01 16:16:59 +0200 |
commit | 61105115ce24706a0e0c16c0f29b0f49e07ee2f7 (patch) | |
tree | c837952ebcc79834a53860835b2c01dea4171d97 /python | |
parent | 6833ddc7ac15a17961264ccb8df433e8d4fa1f07 (diff) | |
download | refcat-61105115ce24706a0e0c16c0f29b0f49e07ee2f7.tar.gz refcat-61105115ce24706a0e0c16c0f29b0f49e07ee2f7.zip |
add UnmatchedResolveJournalNames
Diffstat (limited to 'python')
-rw-r--r-- | python/refcat/tasks.py | 22 |
1 files changed, 22 insertions, 0 deletions
diff --git a/python/refcat/tasks.py b/python/refcat/tasks.py index 17caf96..4193867 100644 --- a/python/refcat/tasks.py +++ b/python/refcat/tasks.py @@ -864,6 +864,28 @@ class UnmatchedMapped(Refcat): return luigi.LocalTarget(path=self.path(ext="tsv.zst"), format=Zstd) +class UnmatchedResolveJournalNames(Refcat): + """ + Try to resolve journal names so we can match against both abbreviations and + full names. + """ + def requires(self): + return UnmatchedMapped() + + def run(self): + output = shellout(""" + zstdcat -T0 {input} | + skate-resolve-journal-name -f 2 -B -A {abbrev} + zstd -T0 -c > {output} + """, + abbrev=settings.JOURNAL_ABBREVIATIONS, + input=self.input().path) + luigi.LocalTarget(output).move(self.output().path) + + def output(self): + return luigi.LocalTarget(path=self.path(ext="tsv.zst"), format=Zstd) + + class UnmatchedOpenLibraryMatchTable(Refcat): """ Run matching and write tabular results to file. About 50M rows. |