3

I am trying to use the output of a cut command into a head command to output a specific line. I find that if the number being fed into the head command 10+ it works fine. So I am wondering if the cut -c1-2 is including white space for the single digits that is tripping up the head command?

my code

#!/bin/bash
echo "Enter your name"
read input
cut -c5-19 filelist | grep -n "$input" | cut -c1-2 > cat

while read cat
do
head -$cat filelist | tail -1 > filelist2
done < cat

Any advice or suggestions will be greatly appreciated! Thank you :) Edit

FULLRanjit Singh   Marketing  Eagles       Dean Johnson   
FULLKen Whillans   Marketing  Eagles       Karen Thompson 
PARTPeter RobertsonSales      Golden TigersRich Gardener  
CONTSandeep Jain   President  Wimps        Ken Whillans   
PARTJohn Thompson  Operations Hawks        Cher           
CONTCher           Operations Vegans       Karen Patel    
FULLJohn Jacobs    Sales      Hawks        Davinder Singh 
FULLDean Johnson   Finance    Vegans       Sandeep Jain   
PARTKaren Thompson EngineeringVegans       John Thompson  
FULLRich Gardener  IT         Golden TigersPeter Robertson
FULLKaren Patel    IT         Wimps        Ranjit Singh

This is 'filelist' The error I am getting is "head: invalid trailing option -- :" If I type in 'Patel' as the name, it works.

Rubén
  • 34,714
  • 9
  • 70
  • 166
Tanya
  • 33
  • 3
  • It's probably best to use a filename other than `cat` because of the `cat` command in bash. Also, what do you mean by "So I am wondering if the cut -c1-2 is including white space for the single digits that is tripping up the head command" – Travis Oct 07 '14 at 05:31
  • I am just confused why it is working for double digits and not single. Since cut -c1-2 would include a space beside a single digit number, I was wondering if that was why I was getting an error when it was being inputted into the head command. – Tanya Oct 07 '14 at 05:36
  • 1
    Can you show more of your code or your exact input, output, and error? For instance, `filelist` is not defined – Travis Oct 07 '14 at 05:41

2 Answers2

0

Answer to revised question

Replace:

cut -c5-19 filelist | grep -n "$input" | cut -c1-2 > cat

With:

cut -c5-19 filelist | grep -n "$input" | cut -d: -f1 >cat

grep -n places a colon between the line number and the text of the line. So, it is natural to use a colon as field delimiter for cut and ask cut to return the first field.

From what you have shown, the script can be further simplified to:

read -p "Enter your name: " input
linenum=$(cut -c5-19 filelist | grep -n "$input" | cut -d: -f1)
head -$linenum filelist | tail -1 > filelist2

Answer to original question

Since you haven't shown us filelist, we can only guess at the problem. If you are right about filelist containing spaces, then this is the solution. Replace:

head -$cat filelist | tail -1 > filelist2

With:

head -${cat## } filelist | tail -1 > filelist2

The construction ${cat## } removes all leading spaces from the varialbe cat.

John1024
  • 109,961
  • 14
  • 137
  • 171
0

Some simplyfications:

First, you can shorten the echo ... read into one read using the -p prompt parameter like:

read -r -p 'Enter your name: ' name

always use the -r otherwise escape sequences get processed.

second, the remaining of the script doing the following: print every line, where there the entered name is within positions 5 .. 19. This can be done easily with awk, like:

awk 'substr($0,5,15)~/pattern/'

where the substr return from the $0 (full input line), 15 characters (5-19 is 15 characters), starting from position 5. the ~/pattern/ matches the pattern. The name must be variable, e.g. the full command could be:

awk -v name=John 'substr($0,5,15)~name'

So, the your script can be reduced to the following two lines:

#!/bin/sh
read -r -p 'Enter your name: '
awk -v name="${REPLY}" 'substr($0,5,15)~name' file1 > file2

further, if you want case insensitive matching of the entered names, use the

awk -v name="${REPLY}" 'tolower(substr($0,5,15))~tolower(name)'
clt60
  • 62,119
  • 17
  • 107
  • 194