Skip to main content

Argparse generator from dataclass

Project description

Current PyPi Version Supported Python Versions codecov docs tests style

Inspired by Simple Parsing, simplified and extended to build extensive, extendable command line interface without much code.

pip install argklass

Features

  • Automatic cli discovery and command plugin

    # Folder structure
    project/cli/
    ├── __init__.py         <= empty
    ├── editor/
       ├── __init__.py     <= ParentCommand(editor)
       ├── cook.py         <= Command(cook)
       ├── client.py       <= Command(client)
       ├── game.py         <= Command(game)
       └── open.py         <= Command(open)
    └── uat/
       ├── __init__.py     <= ParentCommand(uat)
       ├── localize.py     <= Command(localize)
       └── test.py         <= Command(test)
    
    #  editor/__init__.py
    from argklass.command import ParentCommand
    
    class Editor(ParentCommand):
       name = "editor"
    
    
    COMMANDS = Editor
    
    # cook.py
    from argklass.command import Command
    
    class Cook(Command):
       name = "cook"
    
       @staticmethod
       def execute(args) -> int:
          print("cook")
    
    COMMANDS = Cook
    
    #
    cli = CommandLineInterface(project.cli)
    cli.run()
    
    # or
    cli.run(["editor", "cook", "--help"])
    
    #
    cli editor cook --help
    cli uat localize --help
  • New Argument format
    • able to show the entire command line interface with all its subparsers

    • new format mirror dataclass syntax

    editor                                           Set of commands to launch the editors in different modes
       server                                       Parameters added to the Map URL
       game                                         docstring ...
       client                                       docstring ...
       resavepackages                               docstring ...
       cook                                         docstring ...
       ml                                           Launch unreal engine with mladapter setup
       editor                                       Other arguments
       open                                         docstring ...
       localize                                     docstring ...
       worldpartition                               Convert a UE4 map using world partition
       -h, --help                                   Show help
    engine                                           Set of commands to manage engine installation/source
          add                                          docstring ...
          update                                       Update the engine source code
    format                                             docstring ...
          --profile: str                               docstring ...
          --file: str                                  docstring ...
          --fail_on_error: bool = False                docstring ...
          --col: int = 24                              docstring ...
  • Compact argparse definition

    def workdir():
       d = os.getcwd()
       if os.access(d, os.W_OK):
          return d
       return None
    
    
    @dataclass
    class MyArguments:
       a  : str                                                    # Positional
       b  : int                = 20                                # My argument
       c  : bool               = False                             # My argument
       d  : int                = choice(0, 1, 2, 3, 4, default=1)  # choices
       e  : List[int]          = argument(default=[0])             # list
       f  : Optional[int]      = None                              # Optional
       p  : Tuple[int, int]    = (1, 1)                            # help p
       g  : Color              = Color.RED                         # help g
       s  : SubArgs            = SubArgs                           # helps group
       cmd: Union[cmd1, cmd2]  = subparsers(cmd1=cmd1, cmd2=cmd2)  # Command subparser
       de : str                = deduceable(workdir)
    
    parser = ArgumentParser()
    parser.add_arguments(MyArguments)
    args = parser.parse_args()
  • Save and load arguments from configuration files

    parser = build_parser(commands)
    
    # load/save defaults before parsing
    save_defaults(parser, "config.hjson")
    apply_defaults(parser, "config.hjson")
    
    args = parser.parse_args(["editor", "editor"])
    
    # load save arguments after parsing
    save_as_config(parser, args, "dump.hjson")
    apply_config(parser, args, "dump.hjson")
  • Lower level interface, that gives you back all of argparse power

    @dataclass
    class SubArgs:
       aa: str = argument(default="123")
    
    
    @dataclass
    class cmd1:
       args: str = "str1"
    
    
    @dataclass
    class cmd2:
       args: str = "str2"
    
    
    @dataclass
    class MyArguments:
       a: str                  = argument(help="Positional")
       b: int                  = argument(default=20, help="My argument")
       c: bool                 = argument(action="store_true", help="My argument")
       d: int                  = argument(default=1, choices=[0, 1, 2, 3, 4], help="choices")
       e: List[int]            = argument(default=[0], help="list")
       f: Optional[int]        = argument(default=None, help="Optional")
       p: Tuple[int, int]      = argument(default=(1, 1), help="help p")
       g: Color                = argument(default=Color.RED, help="help g")
       s: SubArgs              = group(default=SubArgs, help="helps group")
       cmd: Union[cmd1, cmd2]  = subparsers(cmd1=cmd1, cmd2=cmd2)
    
    
    parser = ArgumentParser()
    parser.add_arguments(MyArguments)
    args = parser.parse_args()

Architecture

argklass works by building the argument parser as a tree, adding metadata to each nodes when necessary.

One of the core component is ArgumentParserIterator which traverse the parsing tree. Each features, such as argument grouping into dataclasses or saving/loading configuration, are implemented as a simple traversal.

This enable us to implement each feature independently from each other and make them optional.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

argklass-1.4.4.tar.gz (20.3 kB view details)

Uploaded Source

Built Distribution

argklass-1.4.4-py3-none-any.whl (21.1 kB view details)

Uploaded Python 3

File details

Details for the file argklass-1.4.4.tar.gz.

File metadata

  • Download URL: argklass-1.4.4.tar.gz
  • Upload date:
  • Size: 20.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.1 CPython/3.9.19

File hashes

Hashes for argklass-1.4.4.tar.gz
Algorithm Hash digest
SHA256 2a41210f2d0bea280b7d5cde65bdadda7797c9ef1b6e2bf73089eb920c5203eb
MD5 6f4ba49ca5ac9a0c5e66362b12d23e6b
BLAKE2b-256 a4a57ecdd392ee15f054797b58f1965e7b1a24543ad4d3bef9010f6e5fc627d7

See more details on using hashes here.

File details

Details for the file argklass-1.4.4-py3-none-any.whl.

File metadata

  • Download URL: argklass-1.4.4-py3-none-any.whl
  • Upload date:
  • Size: 21.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.1 CPython/3.9.19

File hashes

Hashes for argklass-1.4.4-py3-none-any.whl
Algorithm Hash digest
SHA256 5964d54d61daf61bede7a2f25947a75d121894aef95818b1556056193d07e39b
MD5 87e6e4c5aae067a5a001d6be4f052de9
BLAKE2b-256 e0560949f0ba02fad14f8d50adc5a39cbcb3a857a157c2567cd5739732adcd7c

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page