Shell Programming and Scripting

hello,

i have publicly available txt file with little less than 300000 rows. i want to extract from column 1 to column 218 and save it in another text file. i use the cut command but the file is saved with multiple rows from the source file onto a single row in the destination. basically it is not formatted like the source file. the source file is in the below link. i have removed the www since this is my first post and the file name is INP_psf1301.txt

cms.gov/Medicare/Medicare-Fee-for-Service-Payment/ProspMedicareFeeSvcPmtGen/Downloads/INP_psf1301.zip


i use the following

Code:

cut -c1, 218 INP_psf1301.txt > outputfile.txt

but the resulting file is is not formatted like the original. i want the resulting file to be exactly the same except without the last few columns. i want columns 1 to 218 in the resulting file.

is there a line break or something that i should mention or would AWK command work better?

thanks

Try

Code:

cut -c 1-218 INP_psf1301.txt > outputfile.txt

OR

Code:

awk '{ print substr($0,1,218); } ' INP_psf1301.txt > outputfile.txt

Thank you. But i still get my target text file with the multiple rows from source file in one row in target file. i checked to see if the word wrap option is disabled and i doesn't make a difference.

I think you have (windows?) control chars (esp. <CR> = carriage return = ^M = 0x0D) in your file, making subsequent lines overwrite earlier lines. Give it a shot after removing those... (tr -d '\r' <file)

thanks Rudi. i tried your command and it removes the carriage return. i view the file in Notepad++ and there is no carriage return.

but i have a delphi utility that loads the text file and it looks for the carriage return so what i actually need is to remove columns 219 to 236 but keep the [CR][LF]

Then extend bipinajith's script to look like

Code:

awk '{ print substr($0,1,218) "\r"; }' ...

thanks rudi. this works for me