JK Utils

These are my own Python utilities that I have been using over the last years. Because they might be useful for others too, I decided to make them freely available for anyone.

License

The libraries are released under LGPL 3, which means that you can use in commercial/closed-source software if you want.

What is in the package?

So far, there are utilities for the following things:

• graphs: Utility library to construct graphs, find sub-graphs, search paths, etc...
• fuzzy_hashing: My own fuzzy hashing algorithms.
• factor: Primes and factorization utilities.
• simple_log: Extremely simple logging utilities.
• process: Simple process management utilities.
• query_utils: Utilities for constructing SQL language's "where" clauses.
• web_db: Wrappers around web.py database utilities, for MySQL and SQLite.

How to use them?

All python modules are independent. If you want to use one of the mentioned utilities in your project, just copy the .py file into your project's folder. That's all.

Examples

Graphs

This module has support for constructing graphs and performing some basic and common operations with them.

Finding sub-graphs

This is an easy example usage of the graphs library showing how to find if a graph is a sub-graph of another graph (isomorphism):

def test_is_subgraph():
  """
  G1)
           A
          / \
         B   C
        / \ / \
       D  E F  G

  G2)
           A
          / 
         B
        / \
       D  E
  """

  a = CNode("a")
  b = CNode("b")
  c = CNode("c")
  d = CNode("d")
  e = CNode("e")
  f = CNode("f")
  g = CNode("g")

  g1 = CGraph()
  g1.add_edge(a, b)
  g1.add_edge(a, c)
  g1.add_edge(b, d)
  g1.add_edge(b, e)
  g1.add_edge(c, f)
  g1.add_edge(c, g)

  g2 = CGraph()
  g2.add_edge(a, b)
  g2.add_edge(b, d)
  g2.add_edge(b, e)

  # Check if it's a subgraph
  assert g1.is_subgraph(g2) 
  
  # Change the graph and check again
  g2.add_edge(a, d)
  assert g1.is_subgraph(g2) == False

Finding paths between nodes

This is an example showing how to find if there is a path between 2 nodes, finding the shortest and the longest paths:

def test_find_paths():
  a = CNode("a")
  b = CNode("b")
  c = CNode("c")
  d = CNode("d")
  e = CNode("e")
  f = CNode("f")
  g = CNode("g")

  g1 = CGraph()
  g1.add_edge(a, b)
  g1.add_edge(a, c)
  g1.add_edge(b, d)
  g1.add_edge(b, e)
  g1.add_edge(c, f)
  g1.add_edge(f, g)
  g1.add_edge(a, g)

  path = g1.search_path(a, g)
  if path:
    print "Path found between %s and %s" % (a, g), path
    print "Shortest path:", g1.search_shortest_path(a, g)
    print "Longest path:", g1.search_longest_path(a, g)

Process

This module contains only 1 function and 1 class: process_manager and TimeoutCommand.

Running multiple process in parallel

The module function process_manager maintains running in parallel a total ammount of specified processes running some function and waiting for each thread to finish for a specified number of second(s).

Example:

import time
from process import process_manager

def do_nothing():
  try:
    log("Doing nothing...")
    time.sleep(1)
  except KeyboardInterrupt:
    print "Aborted."

if __name__ == "__main__":
  process_manager(2, do_nothing, [], 1)

Running an operating system command with a timeout

The class TimeoutCommand supports executing an operating system command with a specified timeout in seconds in Linux, Unix and Windows. Example:

from process import TimeoutCommand

cmd = TimeoutCommand("/long/time/running_command -with -options")
ret = cmd.run(timeout)
print "Exit code %d" % ret