Taking line count from file and printing in file

rohit_shinez · 08-29-2013, 10:36 AM

hi,

i am having a file which contains the below content, i need to take the line count of if and print the file name also

Code:
inputfile.txt
file_name1.txt,type_name1.txt
file_name2.txt,type_name2.txt

i would need the line count of the files like this
Code:
if file_name*.txt
then
wc -l file_name*.txt - 1
if type_name*.txt
then
wc -l type_name*.txt -2

so my output file format should have
Code:
outputfile.txt
file_name1.txt - 10,type_name1.txt - 20
file_name2.txt - 20,type_name2.txt - 30

considering the line count of the files like below
Code:
file_name1.txt - 11
file_name2.txt - 12
type_name1.txt-22
type_name2.txt - 32

schneidz · 08-29-2013, 10:43 AM

you should use code tags to preserve your formatting.

i would the output of wc -l in a variable and use expr to subtract what you dont want.

rohit_shinez · 08-29-2013, 10:50 AM

i tried that but cant able to format according to my output which i have mentioned
and also not able to check in if conditions
can u help me with the code please

grail · 08-29-2013, 10:58 AM

I would have to say I am a little confused (maybe a lot)

You have a file called inputfile.txt with 2 lines in it:

Code:

file_name1.txt,type_name1.txt
file_name2.txt,type_name2.txt

With you so far

But then you pull an 'if' on what exactly?

Code:

if file_name*.txt

The only file you have mentioned is inputfile.txt, so I would expect neither 'if' to work at all???

As for your output file, as I am lost on the above I cannot see how came up with any of the figures you have displayed

rohit_shinez · 08-29-2013, 11:00 AM

my if conditions are it should apply line count for certain files for eg for names containing file_name*.txt(i.e file_name1.txt,file_name2.txt,....so on) in inputfile.txt, i need to apply line count with subtracting 1 and another if condition names containing like
type_name*.txt(i.e type_name1.txt,type_name2.txt,...so on) i need the to apply line count with subtracting 2

grail · 08-29-2013, 11:41 AM

And how exactly are you getting the data out of inputfile.txt so the 'if' statements can see the data?

Quote:

i need to apply line count with subtracting 1

This makes little sense, line count of what? There are 2 lines in inputfile.txt, are we subtracting 1 from that count?

You need to be clearer on how this is supposed to work and how and what data is being manipulated.

rohit_shinez · 08-29-2013, 11:50 AM

let me make it simple now with an example
i will read the files from my inputfile.txt which contains like below
file_name1.txt,type_name1.txt
file_name2.txt,type_name2.txt
.
.
.

i need to take line count of file_name1.txt,file_name1.txt,type_name1.txt,file_name2.txt,type_name2.txt these files are already present in my directory,

here i need to apply a logic either in if condition or case condition like if it contains like file_name*.txt then take line count and subtract -1 from the line count(wc -l file_name*.txt -1) and if contains line type_name*.txt then take line count and subract 2 from the line count(wc -l type_name*.txt -2)

this will be redirected to my output file in the below format

file_name1.txt - 10,type_name1.txt - 20
file_name2.txt - 20,type_name2.txt - 30

actual line counts of the file:
considering the line count of the files like below
Code:
file_name1.txt - 11
file_name2.txt - 12
type_name1.txt-22
type_name2.txt - 32

grail · 08-29-2013, 11:56 AM

Ok ... that is becoming clearer

Please show how you are reading / getting the data out of inputfile.txt? (ie. before the if can be used we need to have the data from inputfile.txt stored in variable(s))

rohit_shinez · 08-29-2013, 12:01 PM

while IFS=',' read -r filename data; do
...
done < "inputfile.txt"

this like i thought i would read it hope this would help

rohit_shinez · 08-29-2013, 12:04 PM

or can use the below logic as well

for i `çat inputfile.txt`
do
read -a FOO <<< $(wc -l $i)
echo "${FOO[1]} - ${FOO[0]}"
done

the above one can be used for reference

schneidz · 08-29-2013, 03:36 PM

Quote:

Originally Posted by schneidz

you should use code tags to preserve your formatting.

i would the output of wc -l in a variable and use expr to subtract what you dont want.

Code:

for item in *
do
 if [ "`echo $item | grep ^file_name`" ]
 then
  n=`wc -l $item | awk '{print $1}'` # 'grep -c . $item' is probably more elegant here
  n=`expr $n - 1`
  echo $item - $n
 fi
 if [ "`echo $item | grep ^type_name`" ]
 then
  n=`wc -l $item | awk '{print $1}'`
  n=`expr $n - 2`
  echo $item - $n
 fi
done

grail · 08-29-2013, 05:37 PM

Well schneidz has presented a solution but I actually have two more questions:

1. Is the format of the data in inputfile.txt known? (ie. is it always file_name and then type_name)

2. Your example shows mock data of 10 lines in file_name1.txt, my question is, how do you then finish with output total of 11 if you are "subtracting" 1, should it not be 9?

rohit_shinez · 08-30-2013, 12:53 PM

the format in the input file remains same only it may extend with row level

for eg:
file_name1.txt,type1_name.txt,type3_name.txt
.
.
.
.
so on

how can i achieve the output in below format by applying the if condition logic
outputfile.txt
file_name1.txt - 10,type_name1.txt - 20
file_name2.txt - 20,type_name2.txt - 30

PTrenholme · 08-30-2013, 01:19 PM

On most systems printf is a command you can call to create formatted output. For example:

Code:

for item in file_name*
do
  name=${item:9}
  n=$(wc -l ${item} | awk '{print $1}') # 'grep -c . $item' is probably more elegant here
  n=$((n - 1))
  if [ -f "type_name${name}" ]
  then
   m=$(wc -l "type_name${name}" | awk '{print $1}')
   m=$((m-2))
   printf "file_name%s = %d, type_name%s = %d\n" "${name}" ${n} "${name}" ${m}
 else
  printf "\nfile_name%s = %d, type_name%s not found.\n" "${name}" ${n} "${name}"
 fi
done

might work.

Warning: UNTESTED CODE

grail · 08-31-2013, 03:01 AM

As there are potentially multiple type_name options I would read the line split into an array so you can then work of the individual entries and apply your calculations as necessary.

Code:

#!/bin/bash

while IFS=, read -a arr
do
	for item in ${arr[*]}
	do
		[[ $item =~ file_name ]] && { s=1; c= ; } || { s=2; c=,; }

		printf "%c%s = %d" "$c" $item $(($(cat $item | wc -l) - s))
	done
	echo
done<$1