modeval
Modeval (or Modular Eval) is a modular and secure string evaluation library that can be used to create custom parsers or interpreters.
Basic Use
from modeval import Parser
# Create a new parser with the default ruleset.
p = Parser()
# Evalute string. Spaces are automatically removed.
print( p.eval('1 * (2-3)') )
Rulesets
The Parser class will use a basic mathematical ruleset if no specific ruleset is specified. Use the default ruleset as a guide on how to make custom ones.
from modeval import Parser, Ruleset
import operator # (standard library)
default_ruleset = Ruleset()
# Notice the order of the array follows order of operations.
default_ruleset.operators = [
[('^', operator.pow), ('**', operator.pow)],
[('*', operator.mul), ('/', operator.truediv)],
[('+', operator.add), ('-', operator.sub)]
]
p = Parser(ruleset = default_ruleset)
Operator behavior is defined by the function attached to the sign/symbol in the tuple.
Note that the attached methods must have two inputs in the correct order (L + R is parsed as add(L, R)).
Modeval also supports functions like sin(), but they are not included in the default ruleset. To add them, reference the following:
from modeval import Parser, Ruleset
import math # (standard library)
custom_ruleset = Ruleset()
# Function order does not matter, so an extra layer of grouping is not needed.
custom_ruleset.functions = [
('sin', math.sin),
('cos', math.cos),
('tan', math.tan)
]
p = Parser(ruleset = custom_ruleset)
# You can now use "sin(...)" in the input string for eval().
Speaking of sin(), what about pi? Modeval also supports custom variables. They can be set like this:
from modeval import Parser, Ruleset
import math # (standard library)
custom_ruleset = Ruleset()
custom_ruleset.variables = [
('pi', math.pi) # Keep in mind this needs to be a value and not a function.
]
p = Parser(ruleset = custom_ruleset)
# Now you can use pi as you would expect (pi*3/2)
Super Technical Limitations
If you're planning on doing something crazy with this library, I'd read this.
Multi-character operators and functions are assigned unicode characters while being processed so there is a limit of around 4000 variables (and also a very high upper limit of functions). If this is a problem you can increase the offset of the function unicode character conversion. Decreasing the offset of the operators runs the risk of an operator getting translated into a regular number.
A possible fix for this is automatically allocating unicode characters based on the number of operators and variables but this is not implemented as of now.
3 Feb 28, 2022
11 Nov 26, 2022
7 Apr 21, 2022
187 Dec 31, 2022
3 May 27, 2022
2 Jan 5, 2022
1 Dec 14, 2021
81 Dec 19, 2022
116 Dec 21, 2022
12.9k Jan 5, 2023
27 Jul 2, 2022
4 Sep 4, 2021
3 Feb 8, 2022
3 Mar 9, 2022
3 Oct 28, 2022
3 Jan 24, 2022
1 Oct 21, 2021
185 Dec 22, 2022
6 Nov 20, 2022