TypeError: 'int' object is not subscriptable when I try to pass three arguments in parse.parse_args()

Question

ap = argparse.ArgumentParser()
ap.add_argument("-i", "--image", help="Path to input image", required=True)
ap.add_argument("-p", "--pivot-point", help="Pivot point coordinates x, y separated by comma (,)", required=True)
ap.add_argument("-s", "--scale", help="Scale to zoom", type = int, required=True)
args = vars(ap.parse_args(['image.jpeg', '(144,72)', 3]))

Hello,

I am trying to pass an image in the arguments to pivot it and scale it for zooming through argparse.

But I'm getting the following error:

Please, let me know what you're doing wrong.

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-131-c7157827dce9> in <module>()
      3 ap.add_argument("-p", "--pivot-point", help="Pivot point coordinates x, y separated by comma (,)", required=True)
      4 ap.add_argument("-s", "--scale", help="Scale to zoom", type = int, required=True)
----> 5 args = vars(ap.parse_args(['image.jpeg', '(144,72)', 3]))

~/anaconda3/lib/python3.6/argparse.py in parse_args(self, args, namespace)
   1728     # =====================================
   1729     def parse_args(self, args=None, namespace=None):
-> 1730         args, argv = self.parse_known_args(args, namespace)
   1731         if argv:
   1732             msg = _('unrecognized arguments: %s')

~/anaconda3/lib/python3.6/argparse.py in parse_known_args(self, args, namespace)
   1760         # parse the arguments and exit if there are any errors
   1761         try:
-> 1762             namespace, args = self._parse_known_args(args, namespace)
   1763             if hasattr(namespace, _UNRECOGNIZED_ARGS_ATTR):
   1764                 args.extend(getattr(namespace, _UNRECOGNIZED_ARGS_ATTR))

~/anaconda3/lib/python3.6/argparse.py in _parse_known_args(self, arg_strings, namespace)
   1801             # and note the index if it was an option
   1802             else:
-> 1803                 option_tuple = self._parse_optional(arg_string)
   1804                 if option_tuple is None:
   1805                     pattern = 'A'

~/anaconda3/lib/python3.6/argparse.py in _parse_optional(self, arg_string)
   2087 
   2088         # if it doesn't start with a prefix, it was meant to be positional
-> 2089         if not arg_string[0] in self.prefix_chars:
   2090             return None
   2091 

TypeError: 'int' object is not subscriptable

Answer 1

The parser normally handles sys.argv[1:] which is a list of strings generated by the shell. Testing with your own list requires strings as well. Note the use of '3' instead of 3.

In [182]: ap.parse_args(['image.jpeg', '(144,72)', '3'])
usage: ipython3 [-h] -i IMAGE -p PIVOT_POINT -s SCALE
ipython3: error: the following arguments are required: -i/--image, -p/--pivot-point, -s/--scale
An exception has occurred, use %tb to see the full traceback.

SystemExit: 2

Now it gives the standardized argparse error with usage and useful diagnosis.

If we provide the short option flags that you told it to expect:

In [183]: ap.parse_args(['-i','image.jpeg', '-p','(144,72)', '-s' '3'])
Out[183]: Namespace(image='image.jpeg', pivot_point='(144,72)', scale=3)
In [184]: vars(_)
Out[184]: {'image': 'image.jpeg', 'pivot_point': '(144,72)', 'scale': 3}

Or we can use split to get a clearer idea of what the proper commandline will look like:

In [186]: ap.parse_args("-i image.jpeg -p (144,72) -s 3".split())
Out[186]: Namespace(image='image.jpeg', pivot_point='(144,72)', scale=3)

Since you have defined -s as type=int, it converts the string '3' into an integer.

For -p you might want to try nargs=2, allowing you to use '-p 144 72'.

With

ap.add_argument("-p", "--pivot-point", nargs=2, metavar=('x','y'), 
    type=int, help="Pivot point coordinates")

I get:

In [196]: ap.print_help()
usage: ipython3 [-h] -i IMAGE -p x y -s SCALE

optional arguments:
  -h, --help            show this help message and exit
  -i IMAGE, --image IMAGE
                        Path to input image
  -p x y, --pivot-point x y
                        Pivot point coordinates
  -s SCALE, --scale SCALE
                        Scale to zoom

In [197]: ap.parse_args("-i image.jpeg -p 144 72 -s 3".split())
Out[197]: Namespace(image='image.jpeg', pivot_point=[144, 72], scale=3)

Answer 2

All of the elements in the list that you pass into parse_args need to be strings, it's up to the argument parser to interpret the strings as the type that you set.