My first Shell that count rows in each file

Question

I'm trying to run a shell script that counts the number of rows in each file, and if the number < 2 I need to move it to a different directory.

shell:

#!/bin/bash
foreach i in (ls *.DAT)
  a=wc -l $i
  if $a=<2 then
    mv $i aux1/pelvar/var/pel/projs/ar/shells/IGUD_OUT/backup

  endif
end

But my shell gets an error:

igud_to_backup.sh: line 8: syntax error near unexpected token `('

igud_to_backup.sh: line 8: `foreach i in (ls *.DAT)'

What is wrong with the shell script?

see also List (or move) only files with a certain number of lines? — don_crissti, Jun 28 '16 at 10:58
Is that a csh or tcsh shell script? (I'm unfamiliar with them but) If it is, the #!-line is wrong and should point to the csh executable instead. — Kusalananda, Jun 28 '16 at 11:46

score 2 · Answer 1 · edited Apr 13 '17 at 12:36

2

There are many problems:

foreach is not a bash keyword: use for
if you want to execute a command use $( ... ) and not just the parenthesis
the command execution in the parenthesis is not needed you can just use shell expansion for i in *.DAT; do (in general see Why *not* parse `ls`?)
to test if a value is less or equal (see man test): if [ $a -le 2 ] ; then
a for is ended by done and not end
an if is ended by fi and not endif
if you give the file name as an argument to wc it will print the number of lines and the file name. Use < to make wc read from standard input

To sum up:

#!/bin/sh
for i in *DAT; do
  a=$( wc -l < "$i" )
  if [ "$a" -le 2 ] ; then
    mv "$i" aux1/pelvar/var/pel/projs/ar/shells/IGUD_OUT/backup
  fi
done

edited Apr 13 '17 at 12:36

Community

1

answered Jun 28 '16 at 10:51

Matteo

9,796
4
51
66

Hey, thanks a lot, now it is much better. but still getting error: unknown operator – Elislu Jun 28 '16 at 10:59
@don_crissti Oops sorry – Matteo Jun 28 '16 at 11:06
@Elislu What do you mean? – Matteo Jun 29 '16 at 10:13
Hey, The Shell:
#!/bin/sh for i in $(ls *.DAT); do if [ "$a" -ge 2 ] ; then mv ${i} /aux1/pelvar/var/pel/projs/ar/shells/IGUD_OUT/backup fi done

He should count how many rows there are in each file and then with two or more rows, he should move it to another directory. But unfortunately it does not work :(
– Elislu Jun 29 '16 at 10:19
The code you posted in the last comment is not what I suggested. But more importantly you are missing the line assigning the number of rows to a a=$( wc -l < "$i" ) – Matteo Jun 29 '16 at 10:20
You're right, now it works! Thank you very much – Elislu Jun 29 '16 at 10:36

score 0 · Answer 2 · answered Jun 28 '16 at 11:36

0

As Matteo mentioned you are trying to use perl syntax. Should be;

for i in $(ls *.DAT); do
  if [ $(wc -l ${i}) -le 2 ]; then
    mv ${i} /path/to/
  fi
done

answered Jun 28 '16 at 11:36

jas-

868

Looks more like csh syntax to me, but anyway, it isn't bash. – Kusalananda Jun 28 '16 at 11:45

My first Shell that count rows in each file

2 Answers2