contrib/coverage-diff: avoid redundant pipelines

Merge multiple sed and "grep | awk" invocations, finally use "sort -u"
instead of "sort | uniq".

Signed-off-by: Beat Bolli <[e-mail]>
Acked-by: Taylor Blau <[e-mail]>
Signed-off-by: Junio C Hamano <[e-mail]>

contrib/coverage-diff.sh

···
 	sort >uncovered_lines.txt
 
 	comm -12 uncovered_lines.txt new_lines.txt |
-	sed -e 's/$/\)/' |
-	sed -e 's/^/ /' >uncovered_new_lines.txt
+	sed -e 's/$/\)/' -e 's/^/ /' >uncovered_new_lines.txt
 
 	grep -q '[^[:space:]]' <uncovered_new_lines.txt &&
 	echo $file >>coverage-data.txt &&
···
 
 echo "Commits introducing uncovered code:"
 
-commit_list=$(cat coverage-data.txt |
-	grep -E '^[0-9a-f]{7,} ' |
-	awk '{print $1;}' |
-	sort |
-	uniq)
+commit_list=$(awk '/^[0-9a-f]{7,}/ { print $1 }' coverage-data.txt | sort -u)
 
 (
 	for commit in $commit_list