hitting return once does not execute awk script 2

[pter]

Hello Forum,
I work under linux kernel 2.6.20-1.2320.fc5
I wrote an awk script (named prim_fail) for generating a sequence of draws from a binomial distribution:

BEGIN {
        srand()
}
{       for(i=1;i<=100;i++)
        {       if (rand()<0.2)
                        print $1=1
                else
                        print $1=0
        }
        exit
}

I run the script as follows:
$ awk -f prim_fail > sim_sheet
After entering the command, I press "return" (cursor jumps to the next line) , but I need to press "return" again to get to the shell prompt again. This is not convenient when running this awk script from within a python script loop, because for every iteration I need to press the "return" key.
Can anyone give me a tip how to avoid pressing "enter" manually? It seems that this problem does not occur when awk gets data from a file: The problem occurs when the data generated by awk are written to a newly created file.
Thank you,
pter

 

[feherke]

Hi

Your script has a block which will be executed for every input record. In the command line you do not specify any input file, so [tt]awk[/tt] will read the standard input.

What is your goal ? How will you run that script from [tt]python[/tt] ?

Feherke.

http://rootshell.be/~feherke/

 

[pter]

My goal is to create a column of 0s and 1s and write it to a file (which works by hitting return twice). Once I have this file, I would like to process it with short awk scripts further (by adding more columns of data).
Finally, I would like collect some statistics from the resulting file (in the python script below denoted as sim_sheet7), and append the stats report to another file (in the python script below denoted as test2)
Here is the python script:

#!/usr/bin/python
import os
for i in range(10):
        os.system("awk -f   prim_fail  > sim_sheet")
        os.system("awk -f   back_fail  sim_sheet > sim_sheet2")
        os.system("awk -f   check_freq  sim_sheet2 > sim_sheet3")
        os.system("awk -f   backup_state  sim_sheet3 > sim_sheet4")
        os.system("awk -f   system_state  sim_sheet4 > sim_sheet5")
        os.system("awk -f   backup_found_failed  sim_sheet5 > sim_sheet6")
        os.system("awk -f   multi_fail  sim_sheet6 > sim_sheet7")
        os.system("awk -f   sim_sum sim_sheet7 >> test2")

...and now I need to hit enter ten times to execute the python script loop.

pter

 

[p5wizard]

Try putting all your code in the BEGIN block, then awk doesn't need an input file and won't read from stdin either...

BEGIN {
        srand()

        for(i=1;i<=100;i++)
        {       if (rand()<0.2)
                        print $1=1
                else
                        print $1=0
        }
        exit
}

HTH,

p5wizard

 

[feherke]

Hi

The simplest way is to modify the first line of execution, according to what you need :

os.system("awk -f prim_fail [red]/dev/null[/red] > sim_sheet")

[gray]# or[/gray]

os.system("[red]echo |[/red] awk -f prim_fail > sim_sheet")

The complicated way is to make us understand what is your goal.

Sorry, what you are doing there is just stupid. You use a lot of temporary files and a lot of execution of external commands, which slows down the script. At least execute them all in one :

#!/usr/bin/python
import os
for i in range(10):
        os.system("awk -f prim_fail | awk -f back_fail | awk -f check_freq | awk -f backup_state | awk -f system_state | awk -f backup_found_failed | awk -f multi_fail | awk -f sim_sum >> test2")

But the best would be to do it all in [tt]python[/tt].

My goal is to create a column of 0s and 1s and write it to a file

Then why not simply create them ?

BEGIN {
  srand()
  for(i=1;i<=100;i++) print rand()<0.2
}

Feherke.

http://rootshell.be/~feherke/