JsonComb

Inspired by having to manually create config files in JSON that enumerated all possible values that I wanted, I decided to create a way to denote many item combinations in a compact form. The idea is simple:

Write this:

{
   "name": "basic",
   "count": ["!C",1,2],
   "type": ["!C","mobility","cabs"]
}

And JsonComb will turn it into this:

[{
   "name": "basic",
   "count": 1,
   "type": "mobility"
},
{
   "name": "basic",
   "count": 1,
   "type": "cabs"
},
{
   "name": "basic",
   "count": 2,
   "type": "mobility"
},
{
   "name": "basic",
   "count": 2,
   "type": "cabs"
}]

Check out the tests for more examples, like nesting!

###Usage

#####Source Directly:

from comb import find_tokens
result_obj = find_tokens(source_obj)

#####Utility (JSON file processing) Source:

from utility import expand
expand(source_file, True)

#####Utility (JSON file processing) Command Line:

python utility.py source.json output.json

or

python utility.py source.json output.json showres

to display (in the terminal) the result

###A Note on Ordering

You probably noticed that JsonComb takes a regular obj and turns it into a list of items (each combination being a list item). Ordering might be important in an application and in JsonComb it is enforced alphabetically. So if you have 3 keys for value expansion:

{
   "alpha": ["!C",1,2],
   "zeta": ["!C","one","two"],
   "charlie": ["!C","a","b"]
}

The first item is alpha in the ordering so the combinations will be as follows:

alpha | charlie | zeta
  1        a       one
  1        a       two
  1        b       one
  1        b       two
  2        a       one
  2        a       two
  2        b       one
  2        b       two

Note that the first alphabetic key is fixed for as long as the other values vary.

###Roadmap

• Allow for customizable ordering (not just alphabetical as current default)
• More robust test suite