Command Line Manager + Interactive Shell for Python Projects
Project description
Free software: ISC license
Documentation: https://manage.readthedocs.io.
Features
With manage you add a command line manager to your Python project and also it comes with an interactive shell with iPython support.
All you have to do is init your project directory (creating the manage.yml file)
$ pip install manage
$ cd /my_project_root_folder
$ manage init
creating manage.yml....
The file manage.yml describes how manage command should discover your app modules and custom commands and also it defines which objects should be loaded in to the shell
The Shell
By default the command manage shell
is included, it is a simple Python REPL console with some
configurable options.
You can change the banner message to say anything you want, e.g: “Welcome to my shell!” and you can also specify some objects to be automatically imported in to the shell context so when you enter in to the shell you already have your project’s common objects available.
Also you can specify a custom function to run or a string based code block to run, useful to init and configure the objects.
Consoles
manage shell
can start different consoles by passing the options
manage shell --ipython
- This is the default (if ipython installed)manage shell --ptpython
manage shell --bpython
manage shell --python
- This is the default Python console including support for autocomplete. (will be default when no other is installed)
The first thing you can do with manage is customizing the objects that will be automatically loaded in to shell, saving you from importing and initializing a lot of stuff every time you need to play with your app via console.
Edit manage.yml with:
project_name: My Awesome Project
help_text: |
This is the {project_name} interactive shell!
shell:
console: bpython
readline_enabled: false # MacOS has no readline completion support
banner:
enabled: true
message: 'Welcome to {project_name} shell!'
auto_import:
display: true
objects:
my_system.config.settings:
my_system.my_module.MyClass:
my_system.my_module.OtherClass:
as: NiceClass
sys.path:
as: sp
init:
insert:
args:
- 0
- /path/to/be/added/automatically/to/sys/path
init_script: |
from my_system.config import settings
print("Initializing settings...")
settings.configure()
Then the above manage.yaml will give you a shell like this:
$ manage shell
Initializing settings...
Welcome to My Awesome Project shell!
Auto imported: ['sp', 'settings', 'MyClass', 'NiceCLass']
>>> NiceClass. <tab> # autocomplete enabled
Watch the demo:
Check more examples in:
https://github.com/rochacbruno/manage/tree/master/examples/
The famous naval fate example (used in docopt and click) is in:
https://github.com/rochacbruno/manage/tree/master/examples/naval/
Projects using manage
Quokka CMS (A Flask based CMS) is using manage
Red Hat Satellite QE tesitng framework (robottelo) is using manage
Custom Commands
Sometimes you need to add custom commands in to your project e.g: A command to add users to your system:
$ manage create_user --name=Bruno --passwd=1234 Creating the user...
manage has some different ways for you to define custom commands, you can use click commands defined in your project modules, you can also use function_commands defined anywhere in your project, and if really needed can define inline_commands inside the manage.yml file
1. Using a custom click_commands module (single file)
Lets say you have a commands module in your application, you write your custom command there and manage will load it
# myproject/commands.py
import click
@click.command()
@click.option('--name')
@click.option('--passwd')
def create_user(name, passwd):
"""Create a new user"""
click.echo('Creating the user...')
mysystem.User.create(name, password)
Now you go to your manage.yml or .manage.yml and specify your custom command module.
click_commands:
- module: commands
Now you run manage –help
$ manage --help
...
Commands:
create_user Create a new user
debug Shows the parsed manage file
init Initialize a manage shell in current...
shell Runs a Python shell with context
Using a click_commands package (multiple files)
It is common to have different files to hold your commands so you may prefer having a commands/ package and some python modules inside it to hold commands.
# myproject/commands/user.py
import click
@click.command()
@click.option('--name')
@click.option('--passwd')
def create_user(name, passwd):
"""Create a new user"""
click.echo('Creating the user...')
mysystem.User.create(name, password)
# myproject/commands/system.py
import click
@click.command()
def clear_cache():
"""Clear the system cache"""
click.echo('The cache will be erased...')
mysystem.cache.clear()
So now you want to add all those commands to your manage editing your manage file with.
click_commands:
- module: commands
Now you run manage –help and you have commands from both modules
$ manage --help
...
Commands:
create_user Create a new user
clear_cache Clear the system cache
debug Shows the parsed manage file
init Initialize a manage shell in current...
shell Runs a Python shell with context
Custom click_command names
Sometimes the name of commands differ from the name of the function so you can customize it.
click_commands:
- module: commands.system
config:
clear_cache:
name: reset_cache
help_text: This resets the cache
- module: commands.user
config:
create_user:
name: new_user
help_text: This creates new user
Having different namespaces
If customizing the name looks too much work for you, and you are only trying to handle naming conflicts you can user namespaced commands.
namespaced: true
click_commands:
- module: commands
Now you run manage –help and you can see all the commands in the same module will be namespaced by modulename_
$ manage --help
...
Commands:
user_create_user Create a new user
system_clear_cache Clear the system cache
debug Shows the parsed manage file
init Initialize a manage shell in current...
shell Runs a Python shell with context
And you can even customize namespace for each module separately
click_commands:
- module: commands.system
namespace: sys
- module: commands.user
namespace: user
Now you run manage –help and you can see all the commands in the same module will be namespaced.
$ manage --help
...
Commands:
user_create_user Create a new user
sys_clear_cache Clear the system cache
debug Shows the parsed manage file
init Initialize a manage shell in current...
shell Runs a Python shell with context
2. Defining your inline commands in manage file directly
Sometimes your command is so simple that you do not want (or can’t) have a custom module, so you can put all your commands in yaml file directly.
inline_commands:
- name: clear_cache
help_text: Executes inline code to clear the cache
context:
- sys
- pprint
options:
--days:
default: 100
code: |
pprint.pprint({'clean_days': days, 'path': sys.path})
Now running manage –help
$ manage --help
...
Commands:
clear_cache Executes inline code to clear the cache
debug Shows the parsed manage file
init Initialize a manage shell in current...
shell Runs a Python shell with context
And you can run using
$ manage clear_cache --days 15
3. Using general functions as commands
And if you already has some defined function (any callable works).
# my_system.functions.py
def create_user(name, password):
print("Creating user %s" % name)
function_commands:
- function: my_system.functions.create_user
name: new_user
help_text: Create new user
options:
--name:
required: true
--password:
required: true
Now running manage –help
$ manage --help
...
Commands:
new_user Create new user
debug Shows the parsed manage file
init Initialize a manage shell in current...
shell Runs a Python shell with context
$ manage new_user --name=Bruno --password=1234
Creating user Bruno
Further Explanations
You can say, how this is useful?, There’s no need to get a separate package and configure everything in yaml, just use iPython to do it. Besides, IPython configuration has a lot more options and capabilities.
So I say: Nice! If you don’t like it, dont’t use it!
Credits
This is inspired by Django’s manage.py command
This is based on click
This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.
Similar projects
Cobra is a manage for Go language https://github.com/spf13/cobra
History
0.1.15 (2021-02-07)
Fix yaml message.
Fix description in pypi.
0.1.12 (2016-08-15)
Readline is optional because of MacOS compat
0.1.11 (2016-08-15)
Submodules import
multiple ‘as’ names
0.1.10 (2016-07-04)
Bpython added
0.1.9 (2016-07-03)
Bpython added
0.1.8 (2016-07-03)
Python 3 dict fixes
0.1.7 (2016-06-26)
exec bug fix for py3
0.1.6 (2016-06-26)
exec bug fix
0.1.5 (2016-06-25)
added support to commands collector (see quokka cms)
0.1.4 (2016-06-22)
Fix missing deps
0.1.3 (2016-06-19)
Support to function_commands
0.1.2 (2016-06-17)
Support to inline_commands
0.1.1 (2016-06-14)
Support to custom and hidden manage_file
0.1.0 (2016-06-09)
First release on PyPI.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for manage-0.1.15-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 9fdb62db5ea4b3dc34e5e6ae9603eb958afb22837b373a0ddc3ed4452b0ef4c4 |
|
MD5 | 41231c882efbf9e26a0a96aef11eb435 |
|
BLAKE2b-256 | 264d6fbf9daf506c61c96c78171c8e906253dab5c3b5df4199edcb3ea22afe08 |