awk在第一个字段中替换字符串的一部分

Question

Please, I have a file that contains fields separated by the character | , and the first column contain a date and hour 2014-10-09/10:00:00.00 . here is the file:

2014-10-09/10:01:00.27|  tha|  99|  awfkj|             kiuaj|   oauhhg|
2014-10-09/10:02:49.00|  okh|  69|  azakj|             wklkj|   hjjhhg| 
2014-10-09/10:15:30.06|  hnt|  19|  klkkj|             kjhkj|   okjhhg|

I want to change the format of date and replace the hour with "000000" it must be like this:

09102014000000|  tha|  99|  awfkj|             kiuaj|   oauhhg|
09102014000000|  okh|  69|  azakj|             wklkj|   hjjhhg| 
09102014000000|  hnt|  19|  klkkj|             kjhkj|   okjhhg|

I don't know what I need to use. For example this?

awk '{ gsub("..", "..", $1) ; print }'

thank you

Answer 1

You could simply do this through sed,

GNU sed,

$ sed -r 's~^([0-9]{4})-([0-9]{2})-([0-9]{2})[^\|]*~\3\2\1000000~' file
09102014000000|  tha|  99|  awfkj|             kiuaj|   oauhhg|
09102014000000|  okh|  69|  azakj|             wklkj|   hjjhhg| 
09102014000000|  hnt|  19|  klkkj|             kjhkj|   okjhhg|

Basic sed,

$ sed 's~^\([0-9]\{4\}\)-\([0-9]\{2\}\)-\([0-9]\{2\}\)[^\|]*~\3\2\1000000~' file
09102014000000|  tha|  99|  awfkj|             kiuaj|   oauhhg|
09102014000000|  okh|  69|  azakj|             wklkj|   hjjhhg| 
09102014000000|  hnt|  19|  klkkj|             kjhkj|   okjhhg|

Answer 2

sed 's/\([0-9]*\)-\([0-9]*\)-\([0-9]*\)[^|]*/\3\2\1000000/'

Answer 3

尝试这个：

sed 's/^\(....\)-\(..\)-\(..\)\/\(..:..:.....\)/\3\2\1000000/' input.txt

Answer 4

As this is about handling dates, I would use a little bash script and call the date command. This way, you can tune it as you wish, just using the date format options:

while IFS="|" read -r a b
do
    a=$(date -d"${a%/*}" +"%d%m%Y")
    echo "${a}000000| $b"
done < file

a=$(date -d"${a%/*}" +"%d%m%Y") gets the first block of text, removes from the slash / and converst it into a date on the format DDMMYYYY .

Test

$ while IFS="|" read -r a b; do a=$(date -d"${a%/*}" +"%d%m%Y"); echo "${a}000000| $b"; done < file
09102014000000|   tha|  99|  awfkj|             kiuaj|   oauhhg|
09102014000000|   okh|  69|  azakj|             wklkj|   hjjhhg| 
09102014000000|   hnt|  19|  klkkj|             kjhkj|   okjhhg|

Answer 5

$ sed -r 's/(^[0-9]{4})-([0-9]{2})-([0-9]{2}).{13}/\3\2\1000000|/g'

09102014000000|  tha|  99|  awfkj|             kiuaj|   oauhhg|
09102014000000|  okh|  69|  azakj|             wklkj|   hjjhhg|
09102014000000|  hnt|  19|  klkkj|             kjhkj|   okjhhg|

Answer 6

A gawk specific way

awk '{a=gensub(/(.+)-(.+)-(.+)(\/.+)/,"\\3\\2\\1000000","g",$1);sub($1,a);print}' input.txt

Output:

09102014000000|  tha|  99|  awfkj|             kiuaj|   oauhhg|
09102014000000|  okh|  69|  azakj|             wklkj|   hjjhhg| 
09102014000000|  hnt|  19|  klkkj|             kjhkj|   okjhhg|