This Docker image is based on dockerfile/ubuntu-desktop to provide a GUI desktop
for running the BLLIP parser's NLTK + X11/Tk parsing shell. It starts up in CLI
mode though for those who don't want the GUI.
The home for the BLLIP parser PyPi package bllipparser is:
Once you've started the Docker container with this image you should be able to
just type in those commands without any further installation steps.
The home for this Dockerfile is:
The source for the BLLIP parser (including Python wrapper) used in the build is
from the official BLLIP GitHub repository:
I add Firefox and Lynx to the minimal desktop for web access and NumPy since some
NLTK features can use it.
A blog post on using this Docker image along with demo video and screen shots at jimwhite.github.io.
Once you've installed Docker, start me is just:
docker run -it --rm -p 5901:5901 jimwhite/bllip-parser-python`
Note that the Docker Hub Registry means you don't even need a manual download step.
Then at the prompt you can start VNC if you want graphics:
That will prompt you for a password and start up the VNC server.
On a Mac an easy way to use the builtin VNC client is to open an OS X Terminal and type:
Or just enter the URL
vnc://192.168.59.103:5901 in the Safari location bar.
That (192.168.59.103) is the default VirtualBox IP address. You can display it with this command:
When you're done using this application container, just exit from at the shell prompt you got when Docker first launched it:
[ root@4683cabf356e:~ ]$ exit
Docker greatly simplifies running Linux applications and works on most any platform including
Mac and Windows.
To run Docker on Mac or Windows, follow the instructions for installing boot2docker, which
is mostly just a few clicks to install deal, here: