How we can use correctly pipe in Linux?

Question

I have a project, the goal it is to extract phone numbers of a data.txt file, so I did a program. Like that :

   grep -E '^[ ]{0,9}\([0-9]{3}\) [0-9]{3}-[0-9]{4}[ ]{0,9}$' $1 | sed 's/^[ \t]*//' > all-result-phonenumber-filter.txt
count=$(wc -l all-result-phonenumber-filter.txt)

echo "The number of line is :$count" >> all-result-phonenumber-filter.txt

My problem is when I want to use this program and to execute on my terminal I had to use pipe for my terminal command. I tried many different commands on my terminal and the last one is :

cat data.txt | ./all-phone-number-filter.sh | cat all-result-phonenumber-filter.txt

But this command doesn't work, and I don't know why. So what is the correct command I had to use with this above format please?

I had to use the following format SDTIN | STDOUT for the pipe.

I give you the data.txt file :

(512) 258-6589

(205) 251-6584

(480) 589-9856

(303) 548-9874

(808) 547-3215

(270) 987-6547

(225) 258-9887

(314) 225-2543

(979) 547-6854

(276) 225-6985

les numeros suivants ne sont pas valables pour ce programme :

+512 325

+512 251 2545654654

+512 6546 6464646

+512546546646564646463313

(314) sgf225-2543

(314) 225-2543fsgaf

(314afd) 225-2543

FSd(314) 225-2543

The result I need to have is :

  (512) 258-6589
(205) 251-6584
(480) 589-9856
(303) 548-9874
(808) 547-3215
(270) 987-6547
(225) 258-9887
(314) 225-2543
(979) 547-6854
(276) 225-6985
The number of line is :10 all-result-phonenumber-filter.txt

Answer 1

Try this:

cat data.txt | ./all-phone-number-filter.sh > all-result-phonenumber-filter.txt

To see the results on too, make the next command

cat all-result-phonenumber-filter.txt

Answer 2

A filter will:

• Read from stdin
• Write to stdout

Your program will:

• Read from stdin
• Write to a hard-coded filename

So your program is not a filter and can not be used similarly to what you want.

The easiest/worst way of turning your program into a filter is to write to a temporary file instead, and then cat that:

#!/bin/sh
# TODO: Rewrite to be a nicer filter that doesn't write to files
tmp=$(mktemp)
grep -E '^[ ]{0,9}\([0-9]{3}\) [0-9]{3}-[0-9]{4}[ ]{0,9}$' $1 | sed 's/^[ \t]*//' > "$tmp"
count=$(wc -l < "$tmp")

echo "The number of line is :$count" >> "$tmp"
cat "$tmp"
rm "$tmp"

Now you can treat it as a filter:

$ cat data.txt | ./all-phone-number-filter.sh | tail -n 3
(979) 547-6854
(276) 225-6985
The number of line is :10

$ cat data.txt | ./all-phone-number-filter.sh > myfile.txt
(no output)

$ nl myfile.txt | tail -n 5
 7  (225) 258-9887
 8  (314) 225-2543
 9  (979) 547-6854
10  (276) 225-6985
11  The number of line is :10