在意外令牌附近获取语法错误`;' 在python中

Question

I am a Python Novice so please help me out...

#!/usr/bin/python -tt

import sys
import commands

def runCommands():
  f = open("a.txt", 'r')
  for line in f:  # goes through a text file line by line
    cmd = 'ls -l ' + line 
    print "printing cmd = " + cmd,
    (status, output) = commands.getstatusoutput(cmd)
  if status:    ## Error case, print the command's output to stderr and exit
      print "error"
      sys.stderr.write(output)
      sys.exit(1)
  print output
  f.close()

def main():
  runCommands()

# Standard boilerplate at end of file to call main() function.
if __name__ == '__main__':
  main()

I run it as follows:

$python demo.py
sh: -c: line 1: syntax error near unexpected token `;'
sh: -c: line 1: `; } 2>&1'
error

Running less $(which python) says:

#!/bin/sh bin=$(cd $(/usr/bin/dirname "$0") && pwd) exec -a "$0" "$bin/python2.5" "$@"

If i remove for loop then it works fine

$cat a.txt
dummyFile


$ls -l dummyFile
-rw-r--r-- 1 blah blah ...................

$python demo.py
printing cmd = ls -l dummyFile
sh: -c: line 1: syntax error near unexpected token `;'
sh: -c: line 1: `; } 2>&1'
error

I am using 'ls' just for showing the problem. Actually i wanna use some internal shell scripts so i have to run this python script in this way only.

Answer 1

The problem is caused by this line:

    cmd = 'ls -l ' + line

it should be modified to:

    cmd = 'ls -l ' + line.strip() 

When you read the line from your text file, you also read the trailing \\n . You need to strip this so that it works. The getstatusoutput() doesn't like the trailing newline. See this interactive test (which is how I verified it):

In [7]: s, o = commands.getstatusoutput('ls -l dummyFile')

In [8]: s, o = commands.getstatusoutput('ls -l dummyFile\n')
sh: Syntax error: ";" unexpected

Answer 2

This seems to be a problem with the "python" command, perhaps it's a shell wrapper script or something.

Run

$ less $(which python)

UPDATE :

Try calling the Python executable directly, it seems to be at /usr/bin/python2.5 :

$ /usr/bin/python2.5 demo.py

Answer 3

The documentation for the commands module states that when you run getstatusoutput(cmd) ,

cmd is actually run as { cmd ; } 2>&1 { cmd ; } 2>&1

This should explain where the ; } 2>&1 ; } 2>&1 is coming from.

My first guess is that the problem is being caused by not stripping the newlines off the end of each line you read from the file, and so the command you're actually running is something like

{ ls -l somedir
; } 2>&1

However, I don't know shell programming very well so I don't know how sh will cope with the contents of the { ... } split over two lines, nor why it reports the problem on line 1 when there are now two lines.

A second guess is that there's a blank line in your file, in which case sh may be complaining because it's looking for an argument for ls and it found ; } 2>&1 ; } 2>&1 instead.

A third guess is that one of the files contains a } , or maybe a ; followed by a } .

Ultimately, I can't say for sure what the problem is without seeing the contents of your file a.txt .

Incidentally, I hope this file doesn't contain a line / && sudo rm -rf / , as that might cause you one or two problems.

Answer 4

Got this answer from somewhere else:

When you iterate through a file as an iterator, newlines are NOT stripped. The following is ACTUALLY what your script is executing. The fact that you have a trailing comma on your print statement (and a newline on your output) is the giveaway.

ls -l dummyFile \n

Which commands interprets as

{ ls -l dummyFile
; } 2>&1

Call line.rstrip() (or just strip) to fix it.

cmd = 'ls -l ' + line.strip()