diff options
| -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 63b12f4..698ef24 100644 --- a/python/refcat/tasks.py +++ b/python/refcat/tasks.py @@ -460,6 +460,28 @@ class URLTabs(Refcat): def output(self): return luigi.LocalTarget(path=self.path(ext="tsv.zst"), format=Zstd) +class URLTabsCleaned(Refcat): + """ + URLTabs, cleaned. + """ + def requires(self): + return URLTabs() + + def run(self): + output = shellout(""" + zstdcat -T0 {input} | + skate-cleanup -X -c url -B -S -f 3 | + LC_ALL=C sort -u -T {tmpdir} -k3,3 -S25% --parallel 4 | + LC_ALL=C grep -E '^https?://' | + zstd -T0 -c > {output} + """, + n=self.n, + 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) class URLList(Refcat): """ |