diff options
| author | Martin Czygan <martin.czygan@gmail.com> | 2021-05-24 23:57:12 +0200 |
|---|---|---|
| committer | Martin Czygan <martin.czygan@gmail.com> | 2021-05-24 23:57:12 +0200 |
| commit | 36b08fbf951e75eb034f06a30938d5ebaedba354 (patch) | |
| tree | 98cfce23087f0cab9f498ff6b7561cabc2242375 | |
| parent | 1f031bed6e0014bfe1f16eadd9be08f02a99ede4 (diff) | |
| download | refcat-36b08fbf951e75eb034f06a30938d5ebaedba354.tar.gz refcat-36b08fbf951e75eb034f06a30938d5ebaedba354.zip | |
UnmatchedMapped: move target
| -rw-r--r-- | python/refcat/tasks.py | 5 |
1 files changed, 3 insertions, 2 deletions
diff --git a/python/refcat/tasks.py b/python/refcat/tasks.py index cf23a55..4880f43 100644 --- a/python/refcat/tasks.py +++ b/python/refcat/tasks.py @@ -751,7 +751,7 @@ class OpenLibraryWorksSorted(Refcat): class UnmatchedMapped(Refcat): """ Map unmatched refs (in release schema) to titles to do approximate title - matches with OL; 29m16.100s. + matches with OL; 35m14.801s. """ def requires(self): @@ -760,12 +760,13 @@ class UnmatchedMapped(Refcat): def run(self): output = shellout(""" zstdcat -T0 {input} | - skate-map -m cns | + skate-map -m cns -skip-on-empty 1 | LC_ALL=C sort -T {tmpdir} -S25% -k1,1 --parallel 4 | zstd -T0 -c > {output} """, tmpdir=self.tmpdir, 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) |