**DEPRECATED**
Checkout https://github.com/ludbek/bro instead. It is a subset of 'easypy'.
It supports project management and task execution. Rest of the features can
be implemented as custom tasks.
WARNING
this package is a protoype. it is created for sole purpose of exploring the
possibilities of combining 'virtualenv', 'virtualenvwrapper', 'pip',
'invoke'and 'twine'. once it gets enough feedbacks at its
'project home'[1], work for stable version will begin. so use it and please
report bugs and provide feedbacks.
the document at pypi ignores white space, so it is suggested to visit the
github 'repo'[1] instead.
HOW IT WORKS
it uses 'invoke', the pythonic task executor at its core. it sits on top of
'invoke', 'pip', 'virtualenv', 'virtualenvwrapper' and 'twine'. it provides
tasks for creating and managing virtualenv, installing, removing, updating
and automatically recording the package requirements at 'meta.json' file and
finally registering and deploying python packages at 'pypi'. it can be used
for creating generic python projects as well as a python package.
INSTALLATION
1. $ pip install easypy
AVAILABLE TASK
GLOBAL
start
starts a new project. creates project directory at specified path,
creates virtualenv for development. after the environment has been
created one can work on it by issuing 'workon <project_name>' at the
terminal. if 'workon' command is not available, source it by
issueing following command:
'source ~/.bashrc'
syntax:
$ py start <project_name> -d <path/to/project/home/> [-p -f]
options
-d = directory
-p = if it is python package
-f = overwrite existing directory at project home
end
it removes virtual environments associated with a project. if --all
option is specified it removes the project directory as well.
syntax
$ py end <project_name> -a
options
-a = if all, removes project directory as well
setup
set ups virtual environments for already existing project. the
project must be created with easypy.
syntax
$ py setup [--test --dev --prod]
options
--dev = for development environment
--test = for test environment
--prod = for production environment
register
it registers a python package at pypi. it uses
'python setup register' behind the scene.
syntax
$ py register
deploy
it bundles and uploads the python package to pypi.
it uses 'python setup sdist' and 'twine upload dist/<package_dist_name>'
syntax
$ py deploy
LISTING AVAILABLE TASKS
1. easypy tasks
$ py
2. local tasks
$ py .
3. a package tasks
$ py <package_name>
4. list all the tasks at once (TODO)
$ py -a
INSIDE VIRTUALENV
add
installs a python package and registers it as a requirements to the
project. if an environment is not specified, it installs the package
as common requirements for all the environments.
syntax
$ py add <package_name> [--dev --test --prod]
options
--dev = installs in development environment
--test = installs in test environment
--prod = installs in production environment
remove
uninstalls a python package and removes it as a requirements to the
project. for now an environment has to be specified.
syntax
$ py remove <package_name> [--dev --test --prod]
options
--dev = installs in development environment
--test = installs in test environment
--prod = installs in production environment
update
updates a package. an environment has to be specified if the package
is not a common requirement.
syntax
$ py update <package_name> [--dev --test --prod]
meta
displays and sets the meta information for a project.
syntax
$ py meta <property_name>[=property_value]
available properties
- name
- version
- description
[1] https://github.com/ludbek/easypy
Checkout https://github.com/ludbek/bro instead. It is a subset of 'easypy'.
It supports project management and task execution. Rest of the features can
be implemented as custom tasks.
WARNING
this package is a protoype. it is created for sole purpose of exploring the
possibilities of combining 'virtualenv', 'virtualenvwrapper', 'pip',
'invoke'and 'twine'. once it gets enough feedbacks at its
'project home'[1], work for stable version will begin. so use it and please
report bugs and provide feedbacks.
the document at pypi ignores white space, so it is suggested to visit the
github 'repo'[1] instead.
HOW IT WORKS
it uses 'invoke', the pythonic task executor at its core. it sits on top of
'invoke', 'pip', 'virtualenv', 'virtualenvwrapper' and 'twine'. it provides
tasks for creating and managing virtualenv, installing, removing, updating
and automatically recording the package requirements at 'meta.json' file and
finally registering and deploying python packages at 'pypi'. it can be used
for creating generic python projects as well as a python package.
INSTALLATION
1. $ pip install easypy
AVAILABLE TASK
GLOBAL
start
starts a new project. creates project directory at specified path,
creates virtualenv for development. after the environment has been
created one can work on it by issuing 'workon <project_name>' at the
terminal. if 'workon' command is not available, source it by
issueing following command:
'source ~/.bashrc'
syntax:
$ py start <project_name> -d <path/to/project/home/> [-p -f]
options
-d = directory
-p = if it is python package
-f = overwrite existing directory at project home
end
it removes virtual environments associated with a project. if --all
option is specified it removes the project directory as well.
syntax
$ py end <project_name> -a
options
-a = if all, removes project directory as well
setup
set ups virtual environments for already existing project. the
project must be created with easypy.
syntax
$ py setup [--test --dev --prod]
options
--dev = for development environment
--test = for test environment
--prod = for production environment
register
it registers a python package at pypi. it uses
'python setup register' behind the scene.
syntax
$ py register
deploy
it bundles and uploads the python package to pypi.
it uses 'python setup sdist' and 'twine upload dist/<package_dist_name>'
syntax
$ py deploy
LISTING AVAILABLE TASKS
1. easypy tasks
$ py
2. local tasks
$ py .
3. a package tasks
$ py <package_name>
4. list all the tasks at once (TODO)
$ py -a
INSIDE VIRTUALENV
add
installs a python package and registers it as a requirements to the
project. if an environment is not specified, it installs the package
as common requirements for all the environments.
syntax
$ py add <package_name> [--dev --test --prod]
options
--dev = installs in development environment
--test = installs in test environment
--prod = installs in production environment
remove
uninstalls a python package and removes it as a requirements to the
project. for now an environment has to be specified.
syntax
$ py remove <package_name> [--dev --test --prod]
options
--dev = installs in development environment
--test = installs in test environment
--prod = installs in production environment
update
updates a package. an environment has to be specified if the package
is not a common requirement.
syntax
$ py update <package_name> [--dev --test --prod]
meta
displays and sets the meta information for a project.
syntax
$ py meta <property_name>[=property_value]
available properties
- name
- version
- description
[1] https://github.com/ludbek/easypy
Metadata
Release files for easypy 0.0.4
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| easypy-0.0.4.tar.gz | 8.5 kB | Without provenance | Details |
Release files / easypy-0.0.4.tar.gz
| Download URL | easypy-0.0.4.tar.gz |
|---|---|
| Size | 8.5 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
0806138eeb62316bbebb7d92012b1e5a80118fa540e5492eb5e00ace71ca8159
|
|
BLAKE2b-256 checksum How to use checksums |
6e0489959e69c22f8c3ac47cfc6107ae2629ea432ebc022dc6abbfc371c29496
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
Security for easypy 0.0.4
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.