---

cpyvke

Curses PYthon Variable and Kernel Explorer

cpyvke is a variable explorer and a kernel manager written in Python 3 ncurses for iPython kernels (Python 2.x or 3.x). cpyvke supports inspection of numpy ndarray among others types, and provides a set of tools to visualize and plot data.

Be aware that cpyvke is still in development : A lot of bugs are definitely there !

Note that this version only works with Python 3.x. The Python 2.7 version is no longer maintened !

---

cpyvke : the Curses interface

Variable explorer and kernel manager.

• Explore variables in IPython kernels
• Manage IPython kernels

---

kd5 : the daemon to communicate with IPython kernels

Communication with ipython kernels.

• Stream variable list each time a change occurs in the kernel
• Listen to request from the client

---

Requirement

• 256 colors terminal is preferred, but cpyvke also works with 8 colors terminals.
• python3-tk : install it with your package manager
• Tested with python 3.5, 3.6 and 3.7

---

Dependencies

• Ipython >= 5.1 (tested with 5.1.0 -- 7.2.0 -- 7.8.0)
• ipykernel (tested with 4.6.1 -- 5.0.1 -- 5.1.3)
• jupyter_client >= 4.4 (tested with 4.4.0 -- 5.2.4 -- 5.3.4)
• psutil (tested with 3.4.2)
• numpy (tested with 1.13.0)
• matplotlib (tested with 1.5.1)

---

Installation

git clone https://github.com/ipselium/cpyvke.git python3 setup.py install

or

pip install cpyvke

---

Quick Start

To start working, just launch cpyvke in a console. It will create a new kernel, start the daemon and launch the client :

cpyvke

If you already have a running kernel, you can use cpyvke last in the console to start working. It will check for the presence of an existing kernel and, if available, start the daemon and launch the client without creating a new kernel:

cpyvke last

You can also launch cpyvke-launch-ipython to open the current kernel :

cpyvke-launch-ipython

You can now work in this Ipython console and cpyvke will display all changes in the associated kernel :

In [1] : run my_program.py

Tips

You can also use : https://github.com/ipselium/vim-cpyvke

vim-cpyvke provides tools to evaluate blocks of code or full scripts directly from vim. The duo cpyvke/vim-cpyvke paired with a vim plugin such as python-mode (www.github.com/klen/python-mode) can provide a complete development environment for Python in console.

---

Manuals

kd5 : The Daemon

Usage: kd5 {start|stop|restart|list} [INTEGER]

• start : start daemon. If no [INTEGER] is provided, a new ipython kernel is created. [INTEGER] is the id of the connection file.
• stop : stop daemon
• restart : restart daemon
• list : list available ipython kernels

cpyvke : The client

Usage: cpyvke [-h] [-L] [-D] [integer]

• positional arguments:

  • [integer] : Start up with existing kernel. INTEGER is the id of the connection file.

• optional arguments:

  • [-h], [--help] : show this help message and exit
  • [-D], [--debug] : Debug mode
  • [-L], [--list] : List all kernels

• bindings:

  • h : help
  • ENTER : Validate/Item menu
  • q|ESC : Previous menu/quit'
  • s : sort by name/type
  • l : limit display to all variable matching the given keyword
  • u : undo limit
  • k : kernel manager
  • / : Search for variable
  • q : previous menu -- quit
  • r : Refresh explorer
  • c-r : Restart Daemon
  • R : Restart connection to daemon
  • D : Disconnect from daemon
  • C : Connect to daemon
  • ↓ : Next line
  • ↑ : Previous line
  • →|↡ Next page
  • ←|↟ Previous page

Setup workspace

• You can directly launch cpyvke. It will create a new kernel, start the daemon and launch the client
• cpyvke-launch-ipython automatically launch the current ipython console
• You can also manually open an existing ipython instance like this : ipython console --existing kernel-xxxxx.json
• For the latest version of Python, you should use jupyter console --existing kernel-xxxxx.json instead of ipython console --existing kernel-xxxxx.json. where xxxxx is the id of the kernel

Note

If you just want to test cpyvke without installing. In cpyvke/ directory :

• launch kd5 first : python3 -m cpyvke.kd5 start
• then launch cpyvke : python3 -m cpyvke.cpyvke
• and launch ipython in another console : python3 -m cpyvke.launch

If you want to use an existing kernel, you can simply run python3 -m cpyvke.kd5 last instead of python3 -m cpyvke.kd5 start.

---

Configuration

A configuration file cpyvke.conf is created in $HOME/.cpyvke/ at first startup. Appearance of the client can be customize (colors, font).

Colors

The available colors are...

• black
• red
• green
• yellow
• blue
• magenta
• cyan
• white

Fonts

cpyvke can also display powerline fonts. You can find them here : https://github.com/powerline/fonts.

Add the following section in $HOME/.cpyvke/cpyvke.conf:

[font]

powerline-font = True

Python kernel version

The cpyvke/kd5 duo handles python 2.x or 3.x kernel equally. To setup the Python kernel you'll want to use :

[kernel version]

version = 3

The version can be 2 or 3 for python 2.x kernel or 3.x kernel, respectively.

---

Known Bugs

cpyvke is still in developpement and may present unexpected behavior !

---