优化多个sed语句

Question

I want to optimize my processing of a file with this structure:

2014-01-21 14:26:05.900,2014-01-21 14:26:05.740,    0.000,    192.168.40.2,   192.168.40.26,6    ,  8000, 33311,  172000,    2000,.A..S.,  0
2014-01-21 14:29:23.900,2014-01-21 14:29:23.340,    0.000,   192.168.40.26,    192.168.40.2,6    , 33317,  8000, 3052000,    2000,.A....,  0
2014-01-21 14:30:25.900,2014-01-21 14:30:25.330,    0.000,   192.168.40.26,    192.168.40.2,17   , 36193,   514,  558000,    2000,......,  0
2014-01-21 14:31:04.901,2014-01-21 14:31:04.222,    0.000,  192.168.40.242,    192.168.40.2,17   , 57516,   514,  422000,    2000,......,  0
2014-01-21 14:31:13.900,2014-01-21 14:31:13.143,    0.000,   192.168.40.16,    192.168.40.2,17   , 53313,   514,  540000,    2000,......,  0

to a file with this structure:

2014-01-21 14:26:05.900,900,0.000,192.168.40.2,192.168.40.26,6,8000,33311,172000,2000,.A..S.,0
2014-01-21 14:29:23.900,900,0.000,192.168.40.26,192.168.40.2,6,33317,8000,3052000,2000,.A....,0
2014-01-21 14:30:25.900,900,0.000,192.168.40.26,192.168.40.2,17,36193,514,558000,2000,......,0
2014-01-21 14:31:04.901,901,0.000,192.168.40.242,192.168.40.2,17,57516,514,422000,2000,......,0
2014-01-21 14:31:13.900,900,0.000,192.168.40.16,192.168.40.2,17,53313,514,540000,2000,......,0

Commands to optimize:

sed -e 's/,\s\+/,/g' -i /tmp/to_filter
sed -e 's/\s\+,/,/g' -i /tmp/to_filter
while IFS=, read -r f1 f2 f3 f4 f5 f6 f7 f8 f9 f10; do
    echo "$f1,${f1##*.},$f3,$f4,$f5,$f6,$f7,$f8,$f9,$f10"
done < /tmp/to_filter

Answer 1

awk 'BEGIN{FS=OFS=","} {t=$2=$1; sub(/.*\./,"",$2); gsub(/ /,""); $1=t} 1' file      
2014-01-21 14:26:05.900,900,0.000,192.168.40.2,192.168.40.26,6,8000,33311,172000,2000,.A..S.,0
2014-01-21 14:29:23.900,900,0.000,192.168.40.26,192.168.40.2,6,33317,8000,3052000,2000,.A....,0
2014-01-21 14:30:25.900,900,0.000,192.168.40.26,192.168.40.2,17,36193,514,558000,2000,......,0
2014-01-21 14:31:04.901,901,0.000,192.168.40.242,192.168.40.2,17,57516,514,422000,2000,......,0
2014-01-21 14:31:13.900,900,0.000,192.168.40.16,192.168.40.2,17,53313,514,540000,2000,......,0

Answer 2

This might work for you (GNU sed):

sed -r 's/^([^,.]*\.([^,]*)),[^,]*/\1,\2/;s/\s*,\s*/,/g' file

EDIT:

sed -r 's/\.([^,]*),[^,]*/.\1,\1/;s/\s*,\s*/,/g' file

Answer 3

I would use a perl one-liner. It maps over every field to remove leading and trailing spaces, then removes all characters from second fields until last . , and then print with all fields joined by a comma:

perl -F, -ane '
    @F = map { s/\A\s+//; s/\s+\Z//; $_ } @F; 
    $F[1] =~ s/\A.*\.//; 
    printf qq|%s\n|, join q|,|, @F
' infile

It yields:

2014-01-21 14:26:05.900,900,0.000,192.168.40.2,192.168.40.26,6,8000,33311,172000,2000,.A..S.,0
2014-01-21 14:29:23.900,900,0.000,192.168.40.26,192.168.40.2,6,33317,8000,3052000,2000,.A....,0
2014-01-21 14:30:25.900,900,0.000,192.168.40.26,192.168.40.2,17,36193,514,558000,2000,......,0
2014-01-21 14:31:04.901,901,0.000,192.168.40.242,192.168.40.2,17,57516,514,422000,2000,......,0
2014-01-21 14:31:13.900,900,0.000,192.168.40.16,192.168.40.2,17,53313,514,540000,2000,......,0

Answer 4

使用awk

awk '{t=$1;gsub(/ /,"");split($1,a,".");$1=t;$2=a[2]}1' FS=, OFS=, file