Skip to content

stephen-soltesz/pipeline

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

45 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Pipeline View & Probe

Pipeline provides a pair of commands: lineview and lineprobe. Together, they make it easy to view streams of data in real time from the command line.

Lineview combines SocketServer to manage client connections and pylab to plot and display data. Multiple lineprobe clients can connect to a single server.

lineview.py

  • Lineview is a threaded display server. It supports a simple, text-based protocol, implemented by lineprobe.

    See: lineview.py --helpshort for more detailed usage notes.

lineprobe.py

  • Lineprobe is a command line client that sends values to the lineview server. By default, lineprobe reads from stdin. However, lineprobe can also run a command and read the command output instead.

    See: lineprobe.py --helpshort for more detailed usage notes.

Install

Pipeline has two dependencies on third-party modules:

  • packaged as python-matplotlib or py-matplotlib or similar.
  • packaged as python-gflags or py-gflags or similar.

Example

Pipeline Example

The above example was created using commands like those below.

First, start the server:

./lineview.py --timestamp

Next run a lineprobe for each line. Please note that the lineprobe commands expect scripts sysctl.py and cpu.py which return a single floating point value and exit.

LAB=("1" "5" "15")
for i in 0 1 2 ; do 
  ./lineprobe.py --axis_ylabel "Load Average" \
      --label load${LAB[$i]} \
      --command "./sysctl.py $i" & 
  sleep 10
done

Finally, read cpu load from stdin and add to a new axis named "cpu_pct".

LAB=("sys" "user")
for i in 0 1 ; do 
  ./cpu.py ${LAB[$i]} | ./lineprobe.py --axis_ylabel "% Load" \
      --label ${LAB[$i]} \
      --axis cpu_pct &
  sleep 1
done

About

plot data from the command line using matplotlib.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published