AWK error 1

[curious01]

Hi,
I am very lost with the errors I get when I execute a previously used command. Though it worked before but now a problem. Is single quote not allowed on some machines? Can anyone help me?

The error message is:
syntax error near line 1
bailing out near line 1

The script I used is:

awk '
NR==FNR {a[$1]=$1;b[$1]=$2; next}
$1 in a {print a[$1],$2}
'TableA.txt TableB.txt
Thank you

curious1

 

[feherke]

Hi

Though it worked before but now a problem.

What happened since then ? Any input file changed ?

Is single quote not allowed on some machines?

That is matter of shell.

Works for me with [tt]gawk[/tt] and [tt]mawk[/tt] runned from [tt]bash[/tt].

Maybe you have a strange Unix version of [tt]awk[/tt], which does not support blank lines. Try to remove them :

awk 'NR==FNR {a[$1]=$1;b[$1]=$2; next}
$1 in a {print a[$1],$2}' TableA.txt TableB.txt

By the way, you had no space between the code and the first file name. But probably that was just a typo in the posted copy.

Feherke.

http://rootshell.be/~feherke/

 

[p5wizard]

on first glance, you need a space between the final ' and the first file name like so:

awk '
NR==FNR {a[$1]=$1;b[$1]=$2; next}
$1 in a {print a[$1],$2}
' TableA.txt TableB.txt

also, the ' chars can be on the same line as the first/last AWK code characters like so:

awk 'NR==FNR {a[$1]=$1;b[$1]=$2; next}
$1 in a {print a[$1],$2}' TableA.txt TableB.txt



HTH,

p5wizard

 

[curious01]

Thank you Feherke and p5wizard.
I 've tried both your suggestions but I still get the error message. When I execute using nawk I get error saying:
syntax error at source line 1.
context is awk >>> ' <<<
nawk bailing out at source line 1

So does that mean single quote is an invalid char to use?
How can I substitute this? Your help will be much much appreciated
Thanks again

curious1

 

[feherke]

Hi

As I wrote, the single quote's ( ' ) role of string delimiter depends on the shell. So what shell you use ?

Feherke.

http://rootshell.be/~feherke/

 

[PHV]

Get rid of the #! line you seem to have in your script.

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ181-2886

 

[curious01]

Hi Feherke and PHV,
Thanks for your response. I am using the bash shell and do not have #! in my script. Do I need to modify my script?

Thanks

curious1

 

[PHV]

Why not posting the WHOLE content of the script file inside [ignore]

 and

[/ignore] tags and how you launch it from bash ?

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ181-2886

 

[curious01]

Hi PHV,
I'm sorry, I am still very new to AWK etc. May I ask what does it mean inside

 and

tags? Well I lauch using : awk -f scriptfile > outfile

Thanks

curious01

 

[feherke]

Hi

Use tags as described here, to mark you code :

http://tipmaster.com/includes/tgmlinfo.cfm

awk -f scriptfile > outfilep

That is not good. That way [tt]awk[/tt] will wait for input data to be available on standard input.

awk -f scriptfile [red]/input/file[/red] > outfilep

Feherke.

http://rootshell.be/~feherke/

 

[p5wizard]

Also then the word "awk" and the single quotes should not be in your scriptfile

scriptfile:
NR==FNR {a[$1]=$1;b[$1]=$2; next}
$1 in a {print a[$1],$2}

launch:
awk -f scriptfile TableA.txt TableB.txt >outfile

without a script file, launch like so:
awk 'NR==FNR {a[$1]=$1;b[$1]=$2; next}
$1 in a {print a[$1],$2}' TableA.txt TableB.txt >outfile



HTH,

p5wizard