I've got a parser with some sub-parsers. I setup a global argument to be used on all subparser. Here's the relevant snippet
parser = argparse.ArgumentParser(prog="my_prog", add_help=False)
parser.add_argument('-d', '--debug', action='store_true', help='debug flag')
subparsers = parser.add_subparsers(dest="subparser_name", help='some help notes')
parser_cmd1 = subparsers.add_parser('cmd1', parents=[parser])
parser_cmd1.add_argument('-f', '-foo', type=str, action=foo, required=False, help='foo command')
parser_cmd2 = subparsers.add_parser('cmd2', parents=[parser])
parser_cmd2.add_argument('-b', '-bar', type=str, action=bar, required=False, help='bar command')
args = parser.parse_args()
parser = args.subparser_name
print(args)
if args.debug:
logging.basicConfig(level=logging.INFO)
if parser == 'cmd1':
if args.foo:
//do foo stuff
if parser == 'cmd2':
if args.bar:
//do bar stuff
So you can a command like such my_prog.py cmd1 -d -f inp_str. Here's the problem: subparser_name is None. The output of print(args) looks kind of like this
Namespace(debug=True, foo="inp_str", subparser_name=None)
Before I added the global debug argument, subparser_name would be the name of the command I ran, i.e. 'cmd1' or 'cmd2'. Now, it's 'None'. Even with the parents=[parser] addition in the subparser creation. How can I fix this? How do I know which command was called?
Split out the common args to a separate ArgumentParser, which is then used as parent for the sub parsers. Also your foo and bar options were specified using -foo and -bar whi should be --foo and --bar. Also you didn't have default values for these so e.g. when -f/--foo wasn't specified args.foo correctly didn't exist.
This works better:
import argparse
common_args = argparse.ArgumentParser(prog="my_prog", add_help=False)
common_args.add_argument('-d', '--debug', action='store_true', help='debug flag')
parser = argparse.ArgumentParser(prog="my_prog", add_help=True)
subparsers = parser.add_subparsers(dest="subparser_name", help='some help notes')
parser_cmd1 = subparsers.add_parser('cmd1', parents=[common_args])
parser_cmd1.add_argument('-f', '--foo', type=str, default='', required=False, help='foo command')
parser_cmd2 = subparsers.add_parser('cmd2', parents=[common_args])
parser_cmd2.add_argument('-b', '--bar', type=str, default='', required=False, help='bar command')
args = parser.parse_args()
parser = args.subparser_name
print(args)
if args.debug:
logging.basicConfig(level=logging.INFO)
if parser == 'cmd1':
if args.foo:
#//do foo stuff
print( f"foo {args.foo}" )
if parser == 'cmd2':
if args.bar:
#//do bar stuff
print( f"bar {args.bar}" )
run with:
args.py cmd1 -f asd
output:
Namespace(subparser_name='cmd1', debug=False, foo='asd')
foo asd
Update:
If you want to be able to use e.g. args.py -d cmd1 then on the creation of parser, specify parents=[common_args]
parser = argparse.ArgumentParser(prog="my_prog", add_help=True, parents=[common_args])
Next time you ask a question ensure you only post code as a minimal reproducible example - i.e. that can be run without adding anything
The subparser's defaults have priority of any values set by the main parser - default or user input. The main does set the subparser_name to 'cmd1', but the subparser changes it back to the default None.
While not evident in your test case, defining debug at both levels has the same problem. The subparser's default overwrites anything set in the main.
In general, it is not a good idea to use the same dest in the main and subparsers. Flags can be the same, but the dest should be different - at least if you want to see anything set by the main.
And using the main parser as a parent to the sub, is just asking for confusion.
I have the following
import argparse
parser = argparse.ArgumentParser(prog='macc', usage='macc [options] [address]')
parser.add_argument('-l', '--list', help='Lists MAC Addresses')
args = parser.parse_args()
print(args)
def list_macs():
print("Found the following MAC Addresses")
I get an error when running with python macc.py -l it says that an argument was expected. Even when I change my code to parser.add_argument('-l', '--list', help='Lists MAC Addresses' default=1) I get the same error.
The default action for an argument is store, which sets the value of the attribute in the namespace returned by parser.parse_args using the next command line argument.
You don't want to store any particular value; you just want to acknowledge that -l was used. A quick hack would be to use the store_true action (which would set args.list to True).
parser = argparse.ArgumentParser(prog='macc')
parser.add_argument('-l', '--list', action='store_true', help='Lists MAC Addresses')
args = parser.parse_args()
if args.list:
list_macs()
The store_true action implies type=bool and default=False as well.
However, a slightly cleaner approach would be to define a subcommand named list. With this approach, your invocation would be macc.py list rather than macc.py --list.
parser = argparse.ArgumentParser(prog='macc')
subparsers = parser.add_subparsers(dest='cmd_name')
subparsers.add_parser('list')
args = parser.parse_args()
if args.cmd_name == "list":
list_macs()
If you use the argument -l on the cli you need to specify an argument, like:
python macc.py -l something
If you set default = 1 on the -l argument you can run your script without using it like this:
python macc.py
I have the following code:
# Get parsed arguments
args = argparse.ArgumentParser(description=Messages().Get(112))
# Get the arguments for sinit
args.add_argument('init', help=Messages().Get(100), action="store_true")
args.add_argument('--url', default=None, help=Messages().Get(101))
# Get the arguments for schema import
args.add_argument('schema-import', help=Messages().Get(104), action="store_true")
args.add_argument('--file', default=None, help=Messages().Get(104))
The --url argument should only be used with init. For example: script.py schema-import --url should not be accepted but script.py schema-import --file should.
How to set arguments as child arguments?
As mentioned there might be a way to do this with argparse, I'm not sure, but in any event I find it more transparent to explicitly handle argument dependencies in application logic. This should achieve what I think you want:
import argparse
import sys
args = argparse.ArgumentParser(description="please only use the '--url' argument if you also use the 'init' argument")
# Going to use aliases here it's more conventional. So user can use, eg,
# -i or --init for the first argument.
args.add_argument('-i', '--init', help='init help', action="store_true")
args.add_argument('-u', '--url', default=None, help='init help')
args.add_argument('-s', '--schema-import', help='schema-import help', action="store_true")
args.add_argument('-f', '--file', help='file help')
def main():
arguments = args.parse_args()
if arguments.url and not arguments.init:
# You can log an output or raise an exception if you want
# But most likely a print statment is most appropriate
# Since you are interacting with the CLI.
print("You can only use the URL option with init. Exiting")
sys.exit(0)
print("gaurd clauses passed. Here is my code...")
...
if __name__ == "__main__":
main()
Test results (my file called temp.py):
$python temp.py -u https://www.google.com
You can only use the URL option with init. Exiting
$
$python temp.py -i -u https://www.google.com
Gaurd clauses passed. Here is my code...
Why bother with doing all the logic when you can let argparse do all the work for you?
Simply use Sub-commands to define different "branches" of execution:
args = argparse.ArgumentParser(description=Messages().Get(112))
subparsers = args.add_subparsers()
parser_init = subparsers.add_parser('init', help=Messages().Get(100))
parser_init.add_argument('--url', default=None, help=Messages().Get(101))
parser_schema = subparsers.add_parser('schema-import', help=Messages().Get(104))
parser_schema.add_argument('--file', default=None, help=Messages().Get(104))
And this will give you what you want without any logic added:
>>> print(args.parse_args(['schema-import', '--url', "some.url"]))
usage: args.py [-h] {init,schema-import} ...
args.py: error: unrecognized arguments: --url some.url
>>> print(args.parse_args(['schema-import', '--file', "some.file"]))
Namespace(file='some.file')
Basic intended usage:
my_framework create Project_title /path/to/project
OR
my_framework create Project_title (ie. use current working directory)
OR
my_framework update (ie. update my_framework rather than creating a new project)
I know I can make name optional by providing it with a default, but, in reality name is not optional provided the user has entered create as the first argument.
Best solution I've come up with is to use a default value for name and then, if the argument name equals its default value, throw an error. But if there's a way to make argparse do this work for me I'd rather learn to do it.
Writing two scripts, my_framework_create and my_framework_update doesn't appeal to me aesthetically.
#!/usr/bin/env python
import argparse
import os
import shutil
from subprocess import call
template_path = "/usr/local/klibs/template"
parser = argparse.ArgumentParser("MY_FRAMEWORK CLI", description='Creates a new MY_FRAMEWORK project or updates MY_FRAMEWORK')
parser.add_argument('action', choices=['create', 'update'], type=str, help='<help text>')
parser.add_argument('name', type=str, help='<help text>')
parser.add_argument('path', default=os.getcwd(), nargs="?", type=str, help='<help text>')
args = parser.parse_args()
if args.action == "create":
# do the create stuff
if args.action == "update":
# do the update stuff
The best way to do this is with a subparser
An example from the docs:
>>> parser = argparse.ArgumentParser()
>>> subparsers = parser.add_subparsers(title='subcommands',
... description='valid subcommands',
... help='additional help')
>>> subparsers.add_parser('foo')
>>> subparsers.add_parser('bar')
>>> parser.parse_args(['-h'])
usage: [-h] {foo,bar} ...
optional arguments:
-h, --help show this help message and exit
subcommands:
valid subcommands
{foo,bar} additional help
In your case you would have create and update as separate subparsers.
Example:
def create(args):
# do the create stuff
print(args)
def update(args):
# do the update stuff
print(args)
parser = argparse.ArgumentParser()
subparsers = parser.add_subparsers(title='subcommands',
description='valid subcommands',
help='additional help')
create_parser = subparsers.add_parser('create')
create_parser.add_argument('name', type=str)
create_parser.set_defaults(func=create)
update_parser = subparsers.add_parser('update')
update_parser.set_defaults(func=update)
Assume I have a program that uses argparse to process command line arguments/options. The following will print the 'help' message:
./myprogram -h
or:
./myprogram --help
But, if I run the script without any arguments whatsoever, it doesn't do anything. What I want it to do is to display the usage message when it is called with no arguments. How is that done?
This answer comes from Steven Bethard on Google groups. I'm reposting it here to make it easier for people without a Google account to access.
You can override the default behavior of the error method:
import argparse
import sys
class MyParser(argparse.ArgumentParser):
def error(self, message):
sys.stderr.write('error: %s\n' % message)
self.print_help()
sys.exit(2)
parser = MyParser()
parser.add_argument('foo', nargs='+')
args = parser.parse_args()
Note that the above solution will print the help message whenever the error
method is triggered. For example, test.py --blah will print the help message
too if --blah isn't a valid option.
If you want to print the help message only if no arguments are supplied on the
command line, then perhaps this is still the easiest way:
import argparse
import sys
parser=argparse.ArgumentParser()
parser.add_argument('foo', nargs='+')
if len(sys.argv)==1:
parser.print_help(sys.stderr)
sys.exit(1)
args=parser.parse_args()
Note that parser.print_help() prints to stdout by default. As init_js suggests, use parser.print_help(sys.stderr) to print to stderr.
Instead of writing a class, a try/except can be used instead
try:
options = parser.parse_args()
except:
parser.print_help()
sys.exit(0)
The upside is that the workflow is clearer and you don't need a stub class. The downside is that the first 'usage' line is printed twice.
This will need at least one mandatory argument. With no mandatory arguments, providing zero args on the commandline is valid.
With argparse you could use ArgumentParser.print_usage():
parser.argparse.ArgumentParser()
# parser.add_args here
# sys.argv includes a list of elements starting with the program
if len(sys.argv) < 2:
parser.print_usage()
sys.exit(1)
Printing help
ArgumentParser.print_usage(file=None)
Print a brief description of how the ArgumentParser should be invoked on the command line. If file is None, sys.stdout is assumed.
The cleanest solution will be to manually pass default argument if none were given on the command line:
parser.parse_args(args=None if sys.argv[1:] else ['--help'])
Complete example:
import argparse, sys
parser = argparse.ArgumentParser()
parser.add_argument('--host', default='localhost', help='Host to connect to')
# parse arguments
args = parser.parse_args(args=None if sys.argv[1:] else ['--help'])
# use your args
print("connecting to {}".format(args.host))
This will print complete help (not short usage) if called w/o arguments.
If you associate default functions for (sub)parsers, as is mentioned under add_subparsers, you can simply add it as the default action:
parser = argparse.ArgumentParser()
parser.set_defaults(func=lambda x: parser.print_usage())
args = parser.parse_args()
args.func(args)
Add the try-except if you raise exceptions due to missing positional arguments.
If you have arguments that must be specified for the script to run - use the required parameter for ArgumentParser as shown below:-
parser.add_argument('--foo', required=True)
parse_args() will report an error if the script is run without any arguments.
Throwing my version into the pile here:
import argparse
parser = argparse.ArgumentParser()
args = parser.parse_args()
if not vars(args):
parser.print_help()
parser.exit(1)
You may notice the parser.exit - I mainly do it like that because it saves an import line if that was the only reason for sys in the file...
There are a pair of one-liners with sys.argv[1:] (a very common Python's idiom to refer the command line arguments, being sys.argv[0] the script's name) that can do the job.
The first one is self-explanatory, clean and pythonic:
args = parser.parse_args(None if sys.argv[1:] else ['-h'])
The second one is a little hackier. Combining the previously evaluated fact that an empty list is False with the True == 1 and False == 0 equivalences you get this:
args = parser.parse_args([None, ['-h']][not sys.argv[1:]])
Maybe too many brackets, but pretty clear if a previous argument selection was made.
_, *av = sys.argv
args = parser.parse_args([None, ['-h']][not av])
parser.print_help()
parser.exit()
The parser.exit method also accept a status (returncode), and a message value (include a trailing newline yourself!).
an opinionated example,
:)
#!/usr/bin/env python3
""" Example argparser based python file
"""
import argparse
ARGP = argparse.ArgumentParser(
description=__doc__,
formatter_class=argparse.RawTextHelpFormatter,
)
ARGP.add_argument('--example', action='store_true', help='Example Argument')
def main(argp=None):
if argp is None:
argp = ARGP.parse_args() # pragma: no cover
if 'soemthing_went_wrong' and not argp.example:
ARGP.print_help()
ARGP.exit(status=64, message="\nSomething went wrong, --example condition was not set\n")
if __name__ == '__main__':
main() # pragma: no cover
Example calls:
$ python3 ~/helloworld.py; echo $?
usage: helloworld.py [-h] [--example]
Example argparser based python file
optional arguments:
-h, --help show this help message and exit
--example Example Argument
Something went wrong, --example condition was not set
64
$ python3 ~/helloworld.py --example; echo $?
0
Most of the answers here required another module, such as sys, to be imported or were using optional arguments. I wanted to discover an answer that used only argparse, worked with required arguments, and if possible worked without catching exceptions. I ended up with the following:
import argparse
if __name__ == '__main__':
arg_parser = argparse.ArgumentParser(add_help=False)
arg_parser.add_argument('input_file', type=str, help='The path to the input file.')
arg_parser.add_argument('output_file', type=str, help='The path to the output file.')
arg_parser.add_argument('-h','--help', action='store_true', help='show this help message and exit')
arg_parser.usage = arg_parser.format_help()
args = arg_parser.parse_args()
The main idea was to use the format_help function in order to provide the help string to the usage statement. Setting add_help to False in the call to ArgumentParser() prevents the help statement from printing twice in certain circumstances. However, I had to create an argument for the optional help argument that mimicked the typical help message once it was set to False in order to display the optional help argument in the help message. The action is set to store_true in the help argument to prevent the help message from filling in a value like HELP for the parameter when it prints the help message.
So for a really simple answer. Most of the time with argparse you are checking to see if parameters are set anyway, to call a function that does something.
If no parameters, just else out at the end and print the help. Simple and works.
import argparse
import sys
parser = argparse.ArgumentParser()
group = parser.add_mutually_exclusive_group()
group.add_argument("--holidays", action='store_true')
group.add_argument("--people", action='store_true')
args=parser.parse_args()
if args.holidays:
get_holidays()
elif args.people:
get_people()
else:
parser.print_help(sys.stderr)
Here is another way to do it, if you need something flexible where you want to display help if specific params are passed, none at all or more than 1 conflicting arg:
import argparse
import sys
def main():
parser = argparse.ArgumentParser()
parser.add_argument('-d', '--days', required=False, help="Check mapped inventory that is x days old", default=None)
parser.add_argument('-e', '--event', required=False, action="store", dest="event_id",
help="Check mapped inventory for a specific event", default=None)
parser.add_argument('-b', '--broker', required=False, action="store", dest="broker_id",
help="Check mapped inventory for a broker", default=None)
parser.add_argument('-k', '--keyword', required=False, action="store", dest="event_keyword",
help="Check mapped inventory for a specific event keyword", default=None)
parser.add_argument('-p', '--product', required=False, action="store", dest="product_id",
help="Check mapped inventory for a specific product", default=None)
parser.add_argument('-m', '--metadata', required=False, action="store", dest="metadata",
help="Check mapped inventory for specific metadata, good for debugging past tix", default=None)
parser.add_argument('-u', '--update', required=False, action="store_true", dest="make_updates",
help="Update the event for a product if there is a difference, default No", default=False)
args = parser.parse_args()
days = args.days
event_id = args.event_id
broker_id = args.broker_id
event_keyword = args.event_keyword
product_id = args.product_id
metadata = args.metadata
make_updates = args.make_updates
no_change_counter = 0
change_counter = 0
req_arg = bool(days) + bool(event_id) + bool(broker_id) + bool(product_id) + bool(event_keyword) + bool(metadata)
if not req_arg:
print("Need to specify days, broker id, event id, event keyword or past tickets full metadata")
parser.print_help()
sys.exit()
elif req_arg != 1:
print("More than one option specified. Need to specify only one required option")
parser.print_help()
sys.exit()
# Processing logic here ...
Cheers!
I like to keep things as simple as possible, this works great:
#!/usr/bin/env python3
Description = """Tool description"""
Epilog = """toolname.py -a aflag -b bflag with these combined it does blah"""
arg_parser = argparse.ArgumentParser(
formatter_class=argparse.RawDescriptionHelpFormatter,
description=Description,
epilog=Epilog,
)
try:
if len(sys.argv) == 1:
arg_parser.print_help()
except Exception as e:
print(e)
This is how I start all my tools as its always good to include some examples
When call add_subparsers method save the first positional argument to dest= and check value after argparse has been initialized, like this:
subparsers = parser.add_subparsers(dest='command')
And just check this this variable:
if not args.command:
parser.print_help()
parser.exit(1) # If exit() - exit code will be zero (no error)
Full example:
#!/usr/bin/env python
""" doc """
import argparse
import sys
parser = argparse.ArgumentParser(description=__doc__)
subparsers = parser.add_subparsers(dest='command',
help='List of commands')
list_parser = subparsers.add_parser('list',
help='List contents')
list_parser.add_argument('dir', action='store',
help='Directory to list')
create_parser = subparsers.add_parser('create',
help='Create a directory')
create_parser.add_argument('dirname', action='store',
help='New directory to create')
create_parser.add_argument('--read-only', default=False, action='store_true',
help='Set permissions to prevent writing to the directory')
args = parser.parse_args()
if not args.command:
parser.print_help()
parser.exit(1)
print(vars(args)) # For debug
This approach is a lot more elegant than most others. Instead of overriding error(), you can control the behaviour a lot more precisely by wrapping the parse_args() method:
import sys
import argparse
HelpFlags = ('help', '--help', '-h', '/h', '?', '/?', )
class ArgParser (argparse.ArgumentParser):
def __init__(self, *args, **kws):
super().__init__(*args, **kws)
def parse_args(self, args=None, namespace=None):
if args is None:
args = sys.argv[1:]
if len(args) < 1 or (args[0].lower() in HelpFlags):
self.print_help(sys.stderr)
sys.exit()
return super().parse_args(args, namespace)
Set your positional arguments with nargs, and check if positional args are empty.
import argparse
parser = argparse.ArgumentParser()
parser.add_argument('file', nargs='?')
args = parser.parse_args()
if not args.file:
parser.print_help()
Reference Python nargs
If your command is something where a user needs to choose some action, then use a mutually exclusive group with required=True.
This is kind of an extension to the answer given by pd321.
import argparse
parser = argparse.ArgumentParser()
group = parser.add_mutually_exclusive_group(required=True)
group.add_argument("--batch", action='store', type=int, metavar='pay_id')
group.add_argument("--list", action='store_true')
group.add_argument("--all", action='store_true', help='check all payments')
args=parser.parse_args()
if args.batch:
print('batch {}'.format(args.batch))
if args.list:
print('list')
if args.all:
print('all')
Output:
$ python3 a_test.py
usage: a_test.py [-h] (--batch pay_id | --list | --all)
a_test.py: error: one of the arguments --batch --list --all is required
This only give the basic help. And some of the other answers will give you the full help. But at least your users know they can do -h
This isn't good (also, because intercepts all errors), but:
def _error(parser):
def wrapper(interceptor):
parser.print_help()
sys.exit(-1)
return wrapper
def _args_get(args=sys.argv[1:]):
parser = argparser.ArgumentParser()
parser.error = _error(parser)
parser.add_argument(...)
...
Here is definition of the error function of the ArgumentParser class.
As you see, the following signature takes two arguments. However, functions outside the class know nothing about first argument self, because, roughly speaking, this argument is for the class.
def _error(self, message):
self.print_help()
sys.exit(-1)
def _args_get(args=sys.argv[1:]):
parser = argparser.ArgumentParser()
parser.error = _error
...
will output:
"AttributeError: 'str' object has no attribute 'print_help'"
You can pass parser (self) in _error function, by calling it:
def _error(self, message):
self.print_help()
sys.exit(-1)
def _args_get(args=sys.argv[1:]):
parser = argparser.ArgumentParser()
parser.error = _error(parser)
...
But if you don't want exit the program right now, return it:
def _error(parser):
def wrapper():
parser.print_help()
sys.exit(-1)
return wrapper
Nonetheless, parser doesn't know that it has been modified. Thus, when an error occurs, it will print the cause of it (by the way, it's a localized translation). So intercept it:
def _error(parser):
def wrapper(interceptor):
parser.print_help()
sys.exit(-1)
return wrapper
Now, when an error occurs, parser will print the cause of it, and you'll intercept it, look at it, and... throw out.