Weighted Random Choice or Treasure Tables with Rarity

Project Goal: Develop an Abstraction for a random treasure generator, where some treasure is more rare than others.

Define a data structure to represent a sequence of weighted values. Then design an algorithm to make random selections from a weighted sequence;  such that the output distribution reflects the desired weights of the values in the sequence.

It is unknown at this time if a relative weighted strategy or a cumulative weighted strategy will be preferred, we should consider both.

Cumulative Weighted Table

#!/usr/bin/python3 
# file name: weighted_choice.py
from dice import d

treasure_table = (
     [40, "Healing Potion"],
     [70, "Dagger +1"],
     [90, "Sword +2"],
     [95, "Bag of Holding"],
     [96, "Holy Avenger"],
     [97, "Blackrazor"],
     [98, "Hammer of Thunder"],
     [99, "Wand of the War Mage"],
     [100, "Orb of Annihilation"] 
)

def cumulative_weighted_choice(cumulative_table):
    max_weight = cumulative_table[-1][0]
    rand_val = d(max_weight)
    for weight, value in cumulative_table:
        if rand_val <= weight:
             return value

for _ in range(20):
    print(cumulative_weighted_choice(treasure_table))

Relative Weighted Table

rel_treasure_table = (
     [40, "Healing Potion"],
     [30, "Dagger +1"],
     [20, "Sword +2"],
     [5, "Bag of Holding"],
     [1, "Holy Avenger"],
     [1, "Blackrazor"],
     [1, "Hammer of Thunder"],
     [1, "Wand of the War Mage"],
     [1, "Orb of Annihilation"]
)

def cumulative_transform(relative_table):
    cumulative_table = sorted(list(relative_table), reverse=True)
    cum_weight = 0
    for pair in cumulative_table:
        cum_weight += pair[0]
        pair[0] = cum_weight
    return cumulative_table

transformed_treasure_table = cumulative_transform(rel_treasure_table)
for _ in range(20):
    print(cumulative_weighted_choice(transformed_treasure_table))