From 2269340c75402260ca92572371477c80cff4bdac Mon Sep 17 00:00:00 2001 From: Martin Czygan Date: Thu, 15 Jul 2021 18:03:36 +0200 Subject: tasks: tweak CDXURL --- python/refcat/tasks.py | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/python/refcat/tasks.py b/python/refcat/tasks.py index 4a4a357..a7a3ad9 100644 --- a/python/refcat/tasks.py +++ b/python/refcat/tasks.py @@ -1500,15 +1500,17 @@ class CDXURL(Refcat): limit = luigi.IntParameter(default=10000, significant=False) def requires(self): - return URLList() + return RefsURL() def run(self): output = shellout(""" zstdcat -T0 {input} | - head -n {limit} | + LC_ALL=C cut -f 1 | + LC_ALL=C head -n {limit} | skate-cdx-lookup -q -s 50ms -c {cache} -j -B | skate-map -m cdxu | - LC_ALL=C sort -T {tmpdir} -k1,1 -S25% | zstd -c -T0 > {output} + LC_ALL=C sort -T {tmpdir} -k1,1 -S25% | + zstd -c -T0 > {output} """, limit=self.limit, input=self.input().path, -- cgit v1.2.3