merge columns from multiple files

grail · 12-21-2010, 05:58 AM

Thanks for the compliment

I did add a gensub in to take those spaces out but actually quite like the output as it was

I also tried adding:

Code:

if(_[FNR])
    _[FNR]=_[FNR] sprintf("%-8s",$2)
else
    _[FNR]=$2

But it comes out wonky because it is missing the length of the input and so agreed with you it would then become a lot more messy

grail · 12-21-2010, 09:52 AM

Just an update -- turns out it was easier than I thought (change in red):

Code:

{
    while( length(_[FNR]) < ((ARGIND - 1) * 8))
        _[FNR]=_[FNR] sprintf("%8s","")

    _[FNR]=_[FNR] sprintf("%-8s",$2)
}

END{
    for(x=1; x < ARGC; x++)
        print _[x]
}

ramram29 · 01-25-2011, 10:59 PM

How about this. Let me know if this is what you are looking for:

Joining Two Columns Using Bash

ghostdog74 · 01-26-2011, 05:04 AM

Quote:

Originally Posted by ramram29

How about this. Let me know if this is what you are looking for:

Joining Two Columns Using Bash

that script is long winded and inefficient, making use of unnecessary external tools. You can easily use paste or just awk

Code:

awk '{getline a<"file2"; print a FS $0}' file1

or even just bash (don't use this method if file is large )

Code:

#!/bin/bash
exec 4<"file2"
while read -r LINE
do
  read L <&4
  echo $L $LINE
done <"file1"
exec 4<&-

ramram29 · 01-26-2011, 07:15 AM

Quote:

Originally Posted by ghostdog74

or even just bash (don't use this method if file is large )

What if the file "IS" very large? Then what? Do you think the external commands will be more useful then?

ghostdog74 · 01-26-2011, 09:35 AM

Quote:

Originally Posted by ramram29

What if the file "IS" very large? Then what? Do you think the external commands will be more useful then?

it has been proven that external commands like awk works better on big files as compared to the shell's while read loop.

pdsun · 06-24-2014, 06:59 AM

Hi

Further modification into the code, i want to print starting 2 columns only for the first file while column two only for the remaining files. The concept is is to use the first column (from first file) as header for the remaining columns. Thanks for the help!!

Quote:

Originally Posted by grail

Just an update -- turns out it was easier than I thought (change in red):

Code:

{
    while( length(_[FNR]) < ((ARGIND - 1) * 8))
        _[FNR]=_[FNR] sprintf("%8s","")

    _[FNR]=_[FNR] sprintf("%-8s",$2)
}

END{
    for(x=1; x < ARGC; x++)
        print _[x]
}