I suggest to turn calcExperience({_level}) to an expression instead of function. I suggest : [total] [e]xp[erience] at level %number%

We can aslo add this expression :

expression [e]xp[erience] to level up %number%:
    return type: number
    get:
        set {_level} to expression-1
        if {_level} <= 15:
            return 2*{_level}+7
        else if {_level} <= 30:
     

Current coverage is not bad (70+%) and in general and I don't want to embark on one of those 100%-coverage crusades. But I know there are parts of the interpreter that are undertested (and don't generally need a whole lot of tests) and it might be nice to get some quick wins in terms of coverage.

The idea would be to be able to select a specifc code segment given its PP (page pointer), also to slice a code segment (some can be as long as 2000 instructions, it could be great to say "hey I want the code segment 12, from instruction n°144 to instruction n°244".

Could also be great to be able to slice the constants table/symbols table and also to print only the constants table/symbols table