Fixing BasicIntTypes to allow C Standard Integers and 'bool'

The purpose of this check is to ensure that all integral types used by the code point to some fixed size type (e.g. an unsigned 8-bit integer). However; the previous implementation only allowed JPL style typedefs (i.e. U8) and ignored C standard integer types (i.e. uint8_t). This causes the query to false-positive when a typedef resolves to a C standard int type.

'bool' has also be allowed as part of the exclusions list as it represents distinct values 'true' and 'false' in C++ code.

Author: LeStarch

cpp/ql/src/JPL_C/LOC-3/Rule 17/BasicIntTypes.ql

@@ -12,7 +12,8 @@
 import cpp
 
 predicate allowedTypedefs(TypedefType t) {
-  t.getName() = ["I64", "U64", "I32", "U32", "I16", "U16", "I8", "U8", "F64", "F32"]
+  t.getName() = ["I64", "U64", "I32", "U32", "I16", "U16", "I8", "U8", "F64", "F32",
+                 "int64_t", "uint64_t", "int32_t", "uint32_t", "int16_t", "uint16_t", "int8_t", "uint8_t"]
 }
 
 /**
@@ -38,8 +39,8 @@ Type getAUsedType(Type t) {
 }
 
 predicate problematic(IntegralType t) {
-  // List any exceptions that should be allowed.
-  any()
+  // 'bool' is allowed as it represents a 'true' or 'false' value
+  t.getName() != ["bool"]
 }
 
 from Declaration d, Type usedType

cpp/ql/src/change-notes/2025-03-11-basic-int-types.md

@@ -0,0 +1,4 @@
+---
+category: majorAnalysis
+---
+* The query "Basic Integral Types" in JPL_C has been updated to allow C standard integer types (uint8_t etc.) and 'bool'.