Cookbooks
Lua
Chapter 3. Numbers
Generating random numbers

Lua:
Generating random numbers

How to:

Lua provides built-in support for generating random numbers via the math.random function. This function can be used in multiple ways, depending on the desired output:

  1. Generating a random floating-point number between 0 and 1:
print(math.random())

Sample output might be 0.13117647051304. Each run produces a different value.

  1. Generating a random integer within a specified range:

To produce a random integer between two bounds, inclusive, you first need to set the seed using math.randomseed(os.time()) for variability, then call math.random with two arguments:

math.randomseed(os.time())
print(math.random(1, 10)) -- Generates a random integer between 1 and 10

Sample output could be 7. Again, the output will vary with each execution.

It’s crucial to set the seed with math.randomseed because without it, math.random could generate the same sequence of numbers each time a program runs. Typically, seeding with the current time, os.time(), ensures different sequences per execution.

Deep Dive

The mechanism underlying the generation of random numbers in Lua (and most programming languages) is not truly random but pseudorandom, generated by an algorithm. These pseudorandom number generators (PRNGs) are deterministic and require a seed value to begin the sequence of number generation. The choice of seeding is crucial for the randomness quality, which is why using the current time is a common practice.

Historically, Lua’s random number generation capabilities have evolved. Earlier versions relied on the C standard library’s rand() function, which varied in quality and performance across implementations. The current version of Lua enhances this by possibly using more robust mechanisms depending on the underlying platform, offering greater consistency and utility in generating random numbers.

For projects requiring cryptographic-level randomness, the built-in Lua functionality might not suffice due to the deterministic nature of PRNGs. In such cases, programmers often turn to external libraries or system-specific APIs that can provide non-deterministic random numbers suitable for high-security applications.

Rounding numbers