Evaluating boolean expressions in script interpreter

Question

I'm currently working on a script interpreter for a game I'm porting as a personal project - I have asked a few questions in regard to it earlier.

It's starting to work pretty well, I did however came to the point where I have to evaluate boolean expressions, for example the ones in while-loops, if-statements, etcetera.

My current code tokenizes the script, and passes the boolean expression to a function called ProcessExpression(string[] tokens) when it encounters one. An example token-array could be:

local.i
<=
local.arrayLength
+
1
&&
(
local.i
<
local.notThis
||
local.i
>
local.notThis
)

Thinking about how to tackle the actual evaluation, I figured I'd start by replacing the variables with actual values, for example: local.i would become 0, local.arrayLength would become 16 and local.notThis would become 10.

Then I thought about pulling the resulting tokens through a postfix calculator to get rid of all the math, leaving just the logical conditions.

However, I'm kind of on the fence if this is the approach I should take or not. Seeing as the postfix calculation alone would add a lot code, I was wondering if there are some tricks I didn't think off? If postfix is indeed the way to go, can I somehow also make it handle the logical conditions alongside with the math operators?

I'm making this in C#, but I am however unwilling to use external libraries or some kind of "Eval"-function. I'm also not interested in making a grammar for this using things like ANTLR, as there must be less intrusive ways for something seemingly "simple" as an expression solver (if that would be the term for it).

Answer 1

So, to answer my own question: I ended up using a postfix calculator approach.

I pass on all tokens to my ExpressionSolver class, this will convert it from Infix to Postfix and then evaluate the Postfix. The operator-list also contains the && and || operators and handles them accordingly when they are encountered.

I also ended up making a simple Regex-based lexer for tokenizing my input after what Kirk Woll said. It certainly made my parsing before I actually get to solving an expression much easier, I used the following resource for that (also see my comment there): Poor man's "lexer" for C#