Help with sed replace

Hello,

I have a comman separated file lets day data.txt in following format

,:000002 CH XIN9I.INDX, 34.7534909645,:000002 CH,:Index XIN9I.INDX
,:000063 CH XIN9I.INDX, 6.3062924781,:000063 CH,:Index XIN9I.INDX
,:000776 CH XIN9I.INDX, 2.7001954832,:000776 CH,:Index XIN9I.INDX

I would like to replace all occurences of [0-9][0-9][0-9][0-9][0-9][0-9] CH with [0-9][0-9][0-9][0-9][0-9][0-9].CH

i.e. following

:000002.CH XIN9I.INDX, 34.7534909645,:000002.CH,:Index XIN9I.INDX
,:000063.CH XIN9I.INDX,  6.3062924781,:000063.CH,:Index XIN9I.INDX
,:000776.CH XIN9I.INDX,  2.7001954832,:000776.CH,:Index XIN9I.INDX

I tried

cat data.txt | sed "s/[0-9][0-9][0-9][0-9][0-9][0-9] CH/[0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9].CH/g" > out.txt

and it did not work as expected.

I would really appreciate any help.

try this,

sed 's/\(......\) \+CH/\1\.CH/g'  infile

sed 's/\([0-9]\{6\}\) \+CH/\1\.CH/g' infile

OR

 perl -pe 's/(.+?)\s+CH/$1.CH/g' infile

Another one:

sed 's/\([0-9]\{6\}\) /\1./g' file

I tried it, the out put is unchanged, please advise, I would really appreciate it

Or just ... (if apply .... depending on the rest of the file)

sed 's/ CH/\.CH/g' infile

?

Hi ctsgnb,

Thank you so much it worked, m I reading it correctly that you ares suggesting to replace space CH with dot CH, correct?

Thanks

---------- Post updated at 01:15 PM ---------- Previous update was at 01:13 PM ----------

Now when I think of it, I believe we would want it to be stricter than that

There will be exactly 6-digits then space and CH and requirement is to replace it with 6-digits then dot CH

Please advise, I would appreciate it

If you're referring to pravin27's sed suggestions, it could be because they use a gnu extension (\+) and you are not using gnu sed.

Regards,
Alister

---------- Post updated at 02:19 PM ---------- Previous update was at 02:17 PM ----------

Try:

sed 's/\([[:digit:]]\{6\}\) CH/\1.CH/g'

Regards,
Alister

Hi Franklin and ctsgnb,

I modified the version Franklin send and made some changes and modified it to

sed 's/\([0-9]\{6\} CH/\1.CH/g' data.txt

And it worked, thanks.

Thanks to everyone who replied, you guys are life savers.

+1 for alister's suggestion

Ruby

$ ruby -ne 'puts $_.gsub(/:(\d{6})\s+(CH)/,"\\1.\\2")' file
,:000002.CH XIN9I.INDX, 34.7534909645,:000002.CH,:Index XIN9I.INDX
,:000063.CH XIN9I.INDX, 6.3062924781,:000063.CH,:Index XIN9I.INDX
,:000776.CH XIN9I.INDX, 2.7001954832,:000776.CH,:Index XIN9I.INDX