ctplot is a tool to create various kinds plots from and to perform statistical analysis on tabular data stored in HDF5 files. The plots are created from the command line or via an interactive web interface.
Included is an extensible tool to convert raw data into HDF5 tables.
To install ctplot, download the ZIP, extract it and run setup.py or just run
# pip install https://github.com/quantenschaum/ctplot/archive/master.zip
To use ctplot with an Apache webserver, create a ctplot.wsgi somewhere on your server containing
#!/usr/bin/env python
from ctplot.wsgi import application
Set the following environment variables to configure the data directories
CTPLOT_BASEDIR=/data
CTPLOT_DATADIR=/data/data
CTPLOT_CACHEDIR=/data/cache
CTPLOT_PLOTDIR=/data/plots
CTPLOT_SESSIONDIR=/data/sessions
It's only neccessary to set CTPLOT_BASEDIR. The other paths are subdirectories of basedir, which can be overridden by setting them explicitly.
Enable mod_wsgi and in your apache config set a WSGIScriptAlias like
WSGIApplicationGroup %{GLOBAL}
WSGIDaemonProcess ctplot processes=2 threads=20
WSGIScriptAlias /ctplot /path/to/ctplot.wsgi
Set processes to the number of plot creating processes that are allowed to run in parallel (number of cores).
To run ctplot as simple CGI script with mod_cgi, create ctplot.py containing
#!/usr/bin/env python
from ctplot.wsgi import application
# this allows the WSGI app to be run as CGI script
if __name__ == '__main__':
import wsgiref.handlers
wsgiref.handlers.CGIHandler().run(application)
and put it into your server tree and register it with a CGI handler.
Run ctserver (depends on tornado) to run ctplot as standalone webserver. You may set the environment variable CTPLOT_PORT to set a port different from the default of 8080 and CTPLOT_ADDRESS to specify a listening address. If CTPLOT_ADDRESS is not set, the webserver will listen on all addresses.
Use the Dockerfile to create a Docker image.
docker build -t ctplot .
To run the image do
docker run -p 8080:8080 -v /path/to/data:/data ctplot