As a simple package manager, Appypi is used to install, remove, upgrade Pypi packages you may need. Here are the ways to do it.
Easy peasy:
$ appypi install <package>
It may be a bit long the first time, as it constructs the package list. You can install multiple packages in one line, each one will be installed in an independant virtualenv:
$ appypi install <package1> <package2> ...
Of course, if you try to install a package that is not on Pypi, appypi will gently decline your proposition. Same thing, if you try to install a package that don’t have any launcher to be used with, like in a library, appypi will stop the installation. Unhappily, there is no way to know it before going far into the installation process.
As with pip, you can use a requirements file to install many packages at once:
$ appypi install --requirements=requirements.txt
But contrary to pip, you can not specify the version of a package like django==1.4. You can not install stuff from git or any other VCS as well. Maybe one day...
Note that each package will be installed in an independant virtualenv. They can’t be installed in conjunction in a unique one.
Sad, but true:
$ appypi remove <package>
A confirmation for deletion is asked to the user. Again, you can remove multiple packages at the same time:
$ appypi remove <package1> <package2> ...
You guessed right:
$ appypi upgrade <package>
The given package is upgraded to the last version available on Pypi.
Yes, it’s easy:
$ appypi upgrade
All packages are upgraded to their last versions available on Pypi.
$ appypi show <package>
This prints details found on Pypi about the installed version of the package. The full description is printed, so it may be a bit verbose. For example, for the Projy package:
Package: Projy
Version: 0.2
Author: Stéphane Péchard
Homepage: https://github.com/stephanepechard/projy
Summary: Projy is a template-based skeleton generator.
Description: Projy
=====
**Projy is a template-based skeleton generator**.
In one command line, you can generate project skeletons
like Python packages, LaTeX document or any file structure
composed of directories and files.
Each file is generated by a different template.
It uses the simple core templating system from Python,
nothing fancy on that part. You can easily add new templates
and new ways to collect data to insert in the created files.
As much as possible, Projy tries to be simple to use and extend.
See [the complete documentation](http://projy.readthedocs.org/).
Installation
============
If you are familiar with Python, it is strongly suggested that you
install Projy in [virtualenv](http://pypi.python.org/pypi/virtualenv).
Pip and Distribute
------------------
If you are on Linux or Mac OS X, just type:
$ sudo pip install projy
If no pip available, try ``easy_install``:
$ sudo easy_install projy
Install from git
----------------
If you prefer git, that is ok too. You can get the very latest code at
Github:
$ git clone http://github.com/stephanepechard/projy
Usage
=====
As an example, let's create a Python package. The Projy template mostly
follows recommendations from
[The Hitchhiker's Guide to Packaging](http://guide.python-distribute.org/).
A simple example
----------------
Use simply:
$ projy PythonPackage TowelStuff
In the same directory as you typed this command, you now have a
*TowelStuff* directory, with the following structure:
TowelStuff/
bin/
bootstrap
CHANGES.txt
docs/
index.rst
LICENSE.txt
MANIFEST.in
README.txt
setup.py
towelstuff/
__init__.py
Each file has been created with a specific template, so the package is
fully functional, yet empty. Now, let's give a little explanation
on each file. You can find [further
information here](http://guide.python-distribute.org/creation.html).
Specify substitutions
---------------------
You can specify template substitutions directly from the command line.
Add them at the end of the previous example:
$ projy PythonPackage TowelStuff "author,Emmett Brown" "date,1851-09-24"
Then the substitutes *author* and *date* (defaulted to the current day)
are defined by the given values, not those computed by Projy.
The format of such substitutions should be "key,value". Neither the
key or the value should therefore include a comma.
Leading and trailing spaces are removed from both key and value.
To know which substitutions can be overwritten this way, use the *-i*
option as described in the dedicated section. You can add substitutions
that are not listed with the *-i* but they won't have any effect if the
template file does not consider them.
Options
-------
Projy comes with some command line options. Type:
$ projy -l
and you'll see the list of available templates in your installation.
That's an easy way to copy/paste the name of the template you want
to use next.
Type
$ projy -i PythonPackage
and you'll see the details of the Python package template. It shows
the created directories and files, with the substitutions included in
the template.
Available templates
===================
The currently available templates are:
* a [Fabric](http://fabfile.org) file ;
* a [LaTeX](http://www.latex-project.org/) book ;
* a [Python](http://python.org/) package ;
* a Python script ;
* a [Projy](https://github.com/stephanepechard/projy) template, meta-style.
* a bootstrap file, to manage your virtualenv happiness ;
See the official doc for more details on created files into these
templates. Soon to come, more templates around Django. Of course,
anyone can propose some templates, they'll be integrated into Projy.
Obviously:
$ appypi list
The output shows the name, the version and the summary of every installed package. Here is an example output:
appypi - 4 installed packages
------------------------------
Name - Version - Summary
------------------------------
Projy - 0.2 - Projy is a template-based skeleton generator.
Fabric - 1.5.0 - Fabric is a simple, Pythonic tool for remote execution and deployment.
subliminal - 0.6.2 - Subtitles, faster than your thoughts
Glances - 1.5.1 - CLI curses-based monitoring tool
You guessed right:
$ appypi update
Note that this update is forced programmatically every 7 days.