How to change date format using sed or awk

[Eric Leung_2]

Hi,

I've got a text file in the following format:

abcdefg|MM/DD/YY|00:30|xyzxyz
where MM=month, DD=day and YY=year

How can I change the format into:

abcdefg|YY/MM/DD|00:30|xyzxyz

using sed or awk?

Thanks in advance

[Jose Mosquera]

Hi,

Try this script procedure:

TEXT=`cat text_file`
FIELD_1=`echo $TEXT|awk -F"|" '{ print $1 }'`
FIELD_2=`echo $TEXT|awk -F"|" '{ print $2 }'`
FIELD_3=`echo $TEXT|awk -F"|" '{ print $3 }'`
FIELD_4=`echo $TEXT|awk -F"|" '{ print $4 }'`
DATE=`echo $FIELD_2|awk -F"/" '{ print $3"/"$1"/"$2 }'`
echo "$FIELD_1|$DATE|$FIELD_3|$FIELD_4" > text_file

Rgds.

[Steve Steel]

Here ia another way. file is parameter

filein=$1
cat $filein|while read linein
do
OLDIFS=$IFS
export IFS="|"
echo "$linein"|read a b c d
export IFS="/"
echo "$b"|read e f g
export IFS=$OLDIFS
echo "$a""|""$g""/""$e""/""$f""|""$c""|""$d"
done


Steve Steel

[Eric Leung_2]

Thanks for your replies and maybe I need to define my problem more clearly.

Actually, I need to change five different files and the number of fields varies from file to file. Unfortunately, there are more than 20 fields in some files.

Hence, it would be better if there is a function to handle all these files.

Thanks.

[Muthukumar_5]

It is good to operate it with the awk. There are | and / can be used as field separators.

FLD="abcdefg|MM/DD/YY|00:30|xyzxyz"
echo "$(echo $FLD | awk -F "|" '{ print $1 }')|$(echo $FLD | awk -F "|" '{ print $2 }' | awk -F "/" '{ print $3"/"$2"/"$1 }')|$(echo $FLD | awk -F "|" '{ print $3"|"$4 }')"

It will give you the desired output.

[Mark Grant]

or

perl -pne 's#(..)/(..)/(..)#$3/$1/$2#' textfile > newfile

[Muthukumar_5]

Get line by line from a file as,

fun1()
{
filename=$1
newfile=/tmp/test.log
rm -f $newfile
touch $newfile

while read line; do
echo line | grep -q "[[:alpha:]]*/[[:alpha:]]*/[[:alpha:]]"
if [[ $? -eq 0 ]]; then
operation of awk to change the format >> newfile
else
echo $file >>newfile
done < filename

mv $newfile $filename

}

fun1
...
fun1

It is good to have your file examples to make a scipt appropriate to ur requirement.

[Fred Ruffet]

Without sed, without awk ? just cut it :)

$>cat file
abcdefg|MM/DD/YY|00:30|xyzxyz
jdfds|07/20/04|HH:MM|jsfklm
fdslmkqjfq|06/15/03|23:14|jfds
$>for i in `cat file`
> do
> echo "`echo $i | cut -d '|' -f 1`|\c"
> echo "`echo $i | cut -d '|' -f 2 | cut -d '/' -f 2`/\c"
> echo "`echo $i | cut -d '|' -f 2 | cut -d '/' -f 1`/\c"
> echo "`echo $i | cut -d '|' -f 2 | cut -d '/' -f 3`|\c"
> echo "`echo $i | cut -d '|' -f 3,4`"
> done
abcdefg|DD/MM/YY|00:30|xyzxyz
jdfds|20/07/04|HH:MM|jsfklm
fdslmkqjfq|15/06/03|23:14|jfds

Regards,

Fred

[curt larson_1 · Accepted solution]

cat yourFile |
awk -F"|" '{
split($2,a,"/");
$2=sprintf("%s/%s/%s",a[3],a[1],a[2]);
print $0;
}'

i don't have my computer running to test that, but it should be pretty close to what you need.

[curt larson_1]

sed would be something like this:

sed 's!|\(..\)/\(..\)/\(..\)|!|\3/\1/\2|!' yourFile

[Rodney Hills]

If you want a function within "awk" so you can call it multiple times, here is a sample-

echo "hello|10/17/03|abc" | awk -F\| 'function swap(x) { split(x,a,"/"); return a[3] "/" a[2] "/" a[1] };{print swap($2)}'

You can call "swap" multiple times within your awk script. My sample is a one-line but you may want to put the awk script into a seperate file.

HTH

-- Rod Hills