bugfix: uniq needs LC_ALL=C too

scripts/src/main/bash/oneliners.txt

@@ -115,13 +115,12 @@ wget --background --spider --recursive --level 1 --domains=downloads.dbpedia.org
 # - may fail for other reasons, depending on file content, shell settings, etc.
 sed -r 's/%([89ABCDEF])/\\x\1/g' abc.nt | while read -r line ; do echo -e $line ; done > abc.ttl
 
+# list all page-links files
+ls -1 */*/*-page-links*.{nt,ttl}.gz | grep -v -E 'en-uris|unredirected' > page-links.txt
+
 # generate in-link counts
-# (include unredirected, may be interesting)
-ls -1 */*/*-page-links*.{nt,ttl}.gz | grep -v -E 'en-uris' > ../page-links-in.txt
-cat ../page-links-in.txt | while read i ; do echo $i; gzip -d < $i | grep -v '^#' | cut -d ' ' -f 3 | env LC_ALL=C sort | uniq -c | sort -k 1,1 -n -s -r | awk '{print $2" <http://dbpedia.org/ontology/wikiPageInLinkCount> \""$1"\"^^<http://www.w3.org/2001/XMLSchema#integer> ." }' | gzip -c > ${i/page-links/page-in-link-counts} ; done &> ../page-in-link-counts.log &
+( export LC_ALL=C ; cat page-links.txt | while read i ; do echo $i; gzip -d < $i | grep -v '^#' | cut -d ' ' -f 3 | sort | uniq -c | sort -k 1,1 -n -s -r | awk '{print $2" <http://dbpedia.org/ontology/wikiPageInLinkCount> \""$1"\"^^<http://www.w3.org/2001/XMLSchema#integer> ." }' | gzip -c > ${i/page-links/page-in-link-counts} ; done &> page-in-link-counts.log & )
 
 # generate out-link count
-# (exclude unredirected, yields identical results)
-ls -1 */*/*-page-links*.{nt,ttl}.gz | grep -v -E 'en-uris|unredirected' > ../page-links-out.txt
 # (the first sort is not really necessary - our triples are already grouped by subject, and grouping is enough for uniq - but it's nicer that the in-link and out-link counts are sorted similarly)
-cat ../page-links-out.txt | while read i ; do echo $i; gzip -d < $i | grep -v '^#' | cut -d ' ' -f 1 | env LC_ALL=C sort | uniq -c | sort -k 1,1 -n -s -r | awk '{print $2" <http://dbpedia.org/ontology/wikiPageOutLinkCount> \""$1"\"^^<http://www.w3.org/2001/XMLSchema#integer> ." }' | gzip -c > ${i/page-links/page-out-link-counts} ; done &> ../page-out-link-counts.log &
+( export LC_ALL=C ; cat page-links.txt | while read i ; do echo $i; gzip -d < $i | grep -v '^#' | cut -d ' ' -f 1 | sort | uniq -c | sort -k 1,1 -n -s -r | awk '{print $2" <http://dbpedia.org/ontology/wikiPageOutLinkCount> \""$1"\"^^<http://www.w3.org/2001/XMLSchema#integer> ." }' | gzip -c > ${i/page-links/page-out-link-counts} ; done &> page-out-link-counts.log & )