From a171f0678e7b04e890e39221805ad1d66ad3a224 Mon Sep 17 00:00:00 2001
From: Anton Karpov <30812217+karpovantonme@users.noreply.github.com>
Date: Wed, 12 Aug 2026 16:58:54 +0300
Subject: [PATCH] docs: Starlark ints are not limited to 32 bits

The page says the int type is limited to 32-bit signed integers and that
overflow throws. StarlarkInt has documented the opposite since it moved to
arbitrary precision, and the generated page at /rules/lib/int says integers
may be of any magnitude and arithmetic is exact.

Part of #27257. The float and set part of that issue is already fixed.
---
 site/en/rules/language.md | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/site/en/rules/language.md b/site/en/rules/language.md
index 2284cdb74fa581..3122e6cc127ce3 100644
--- a/site/en/rules/language.md
+++ b/site/en/rules/language.md
@@ -133,7 +133,8 @@ illegal, and 2) `*args` and `**kwargs` arguments are not allowed.
 
 * Recursion is not allowed.
 
-* Int type is limited to 32-bit signed integers. Overflows will throw an error.
+* Integers may be of any magnitude, and arithmetic on them is exact. See
+  [`int`](lib/int).
 
 * Modifying a collection during iteration is an error.