CODEWITHSUNDEEP
bashfilevariablesfor-loopecho
Omnomnom
Omnomnom

Reputation: 111

Bash reads files multiple times & output isn't right

I'm having a problem with reading multiple files and gathering info out of those files. Those files have an input like:

File1:

number1
text
number2
text
number3
text

File2:

number4
text
number5 
...

I only want to get the numbers out of every file, and those numers need to be placed under each other, but only 1 time.

Like this:

number1
number2
number3
number4
number5

What I have:

var0=$(echo "test")
var2=$(echo "test2")

for file in*.txt
do
     var1=$(awk -F .. do something "$file")
done
echo "$var0" "$var1" "$var2" >> myfile

So what i want it to do:

test1
here print var1 (so the numbers)
test2

The input files have something that i want to get out with the awk command, this works perfectly, when i echo the var i get number1,number2,number3,number4,number5,number1,... so it keeps repeating the command.

When i print all the var's he only gives the value of the last file in var1:

test
number4
number5
test2

And I want everything, so also number1, number2,...

Ty in advance!

Upvotes: 0

Views: 33

Answers (1)

Tom Fenech
Tom Fenech

Reputation: 74655

You're overwriting your variable $var1 inside the loop, so it's no surprise that you only get the contents of the last file.

If I understand you correctly, you want something like this:

awk 'BEGIN{print "test"}/^[0-9]+$/;END{print "test2"}' *.txt

This reads all of the .txt files in one awk command and prints any lines that only contain numbers (I'm assuming that's what you mean when you write number1, number2 etc.)

However if you would like to continue using your approach, I guess the only thing you need to do is concatenate the output of each awk, rather than overwrite $var1:

var1+=$(awk -F .. do something "$file")
----^

If you don't actually care about the variable $var1 there's no need to use concatenation, as shown by tripleee in the comments:

{ echo "test"; for f in *.txt; do awk ... "$f"; done; echo "test2"; } >myfile

You can probably skip the loop too (which will be shorter and quicker):

{ echo "test"; awk ... "$f" *.txt; echo "test2"; } >myfile

...and if you can do that, we've arrived pretty much back at my first suggestion, which is to do the whole thing in one go using awk :)

Upvotes: 1

Related Questions