Constraint Design Problem

Let us say you have a piece of JSON data as follows.
basket = {'grape': 160,
		  'banana': 80,
		  'pear': 320,
		  'orange': 120,
		  'red apple': 290,
		  'green apple': 380,
		  'papaya': 40}

And you have a price constraint as follows.

constraints = {'grape': ('<', 200),
			   'banana': ('<', 100),
			   'green apple': ('<',500), '
			   red_apple': ('>', 1000)}

When you mentally apply the constraint to the data the answer would be,

{"grape": 160, "banana": 80, "green apple": 380"}

Now let us extend it to another level of nesting.

fuel = {'diesel': ('<', 100),
		'petrol': ('<', 120),
		'cng': ('<', 80)}
    
fuel_prices = {'bangalore': { 'diesel': 85,
						'petrol': 100,
						'cng': 75},
		 'delhi': {'diesel': 120,
				   'petrol': 130,
				   'cng': '70'}}

Now if you apply it the "bangalore" element will be selected and "delhi" wont be.

Come up with a generic design approach to solve constraint problems like this. The things to discuss
and have in your design are,

1. A DSL or generic "path" selector for applying the price constraint on to the data
2. How to find the data at the right nesting if you are not using a selector ?
3. How to apply "pattern" matching to the problem ?
4. How to do this in a performant way ?