date field manipulation

Shell Programming and Scripting
1

I have a data file. Seperated by "|". The 19 th filed is a date field that occurs like this

11/02/2001

i need to convert into the below format

2001-11-02

for e.g..

i/p

o/p should be

can somebody throw some light

I need to do this for all records in the file

2

First off, your date field is not 19, but rather 18 (based on your sample).

 nawk -F'|' '{split($18,a,"/");$18=a[3]"-"a[1]"-"a[2]}1' OFS="|" myFile
3

Thanks Vgresh. Is there a way we can do it in sed. I am using couple of other sed commands on the file. Hence would like to see if we can do it in sed.

4

Probably, but why bother - use the tool better suited for the task.
Most likely your other sed 'commands' might be migrated to awk.

5

If you are sure there is only one date field per line in field 18 that needs to be changed, or any other field that contains a date, needs to be changed as well, you can could use this:

sed 's|\(..\)/\(..\)/\(....\)|\3-\1-\2|g' infile

But like vgersh99 says, awk is more suited for the task. It is more accurate since it will only modify field 18.

6 
# cat datet
1|147431|3284924|Superliner|1652382|765x125||Delta flight|1 x 1 Default Ad, 34|Superliner 765x125|ion|rt|rt|332|0|0.0|Clicks|05/07/2010|0|332|0|0|Y

# sed 's@|\([0-9][0-9]\)/\([0-9][0-9]\)/\([0-9][0-9]*\)@|\3-\1-\2@g' datet
1|147431|3284924|Superliner|1652382|765x125||Delta flight|1 x 1 Default Ad, 34|Superliner 765x125|ion|rt|rt|332|0|0.0|Clicks|2010-05-07|0|332|0|0|Y