allows int -> Natural or Positive

changelog.md

@@ -33,7 +33,7 @@ errors.
 
 - Bitshift operators (`shl`, `shr`, `ashr`) now apply bitmasking to the right operand in the C/C++/VM/JS backends.
 
-- Adds a new warning enabled by `--warning:ImplicitRangeConversion` that detects downsizing implicit conversions to range types (e.g., `int -> range[0..255]` or `range[1..256] -> range[0..255]`) that could cause runtime panics. Safe conversions like `range[0..255] -> range[0..65535]` and explicit casts are not warned on.
+- Adds a new warning `--warning:ImplicitRangeConversion` that detects downsizing implicit conversions to range types (e.g., `int -> range[0..255]` or `range[1..256] -> range[0..255]`) that could cause runtime panics. Safe conversions like `range[0..255] -> range[0..65535]` and explicit casts are not warned on. `int` to `Natural` and `Positive` conversions are not warned.
 
 ## Standard library additions and changes
 

compiler/nim.cfg

@@ -65,3 +65,7 @@ define:useStdoutAsStdmsg
 @if nimHasVtables:
   experimental:vtables
 @end
+
+@if nimHasImplicitRangeConversion:
+  warning[ImplicitRangeConversion]:off
+@end

compiler/sempass2.nim

@@ -176,7 +176,12 @@ proc shouldWarnRangeConversion(conf: ConfigRef; formalType, argType: PType): boo
   if f.kind == tyRange:
     # Only warn if formal range doesn't fully contain argument range
     # Check if the ranges don't perfectly overlap
-    result = not isRangeSupertype(conf, f, a)
+    if f.sym != nil and (f.sym.name.s == "Positive" or
+          f.sym.name.s == "Natural"):
+      # Positive and Natural are special cases that we never warn on
+      result = false
+    else:
+      result = not isRangeSupertype(conf, f, a)
   else:
     result = false
 

doc/manual.md

@@ -1144,6 +1144,8 @@ semantic analysis). Assignments from the base type to one of its subrange types
 A subrange type has the same size as its base type (`int` in the
 Subrange example).
 
+Implicit "downsizing" conversions to range types (for example, `int -> range[0..255]` or `range[1..256] -> range[0..255]`) emit the `ImplicitRangeConversion` warning. Conversions that are clearly safe (for example, `range[0..255] -> range[0..65535]`) and any explicit casts do not trigger this warning. Conversions from `int` to common subranges such as `Natural` or `Positive` are also exempt.
+
 
 Pre-defined floating-point types
 --------------------------------