Translate access to flexible array member (#165)

Changed the access pattern of flexible arrays from

```rs
let bytes: [u8; 1] = ...;
bytes[10] = ...
```

to

```rs
let bytes: [u8; 1] = ...;
*bytes.as_mut_ptr().add(10) = ...;
```

The original access panics because the bounds are violated, i.e.
accessing the 10th element in an array declared with 1 element. The new
access does not panic, it preserves the C/C++ behavior.

Author: lucic71

cpp2rust/converter/converter.cpp

@@ -5,6 +5,7 @@
 
 #include <clang/AST/APValue.h>
 #include <clang/AST/ParentMapContext.h>
+#include <clang/Basic/LangOptions.h>
 #include <clang/Basic/SourceManager.h>
 #include <llvm/ADT/DenseMap.h>
 #include <llvm/Support/ConvertUTF.h>
@@ -3421,8 +3422,42 @@ void Converter::ConvertPointerOffset(clang::Expr *base, clang::Expr *idx,
   computed_expr_type_ = ComputedExprType::FreshPointer;
 }
 
+static bool IsFlexibleArrayMemberAccess(clang::ASTContext &ctx,
+                                        clang::Expr *array) {
+  return array->isFlexibleArrayMemberLike(
+      ctx, clang::LangOptions::StrictFlexArraysLevelKind::OneZeroOrIncomplete,
+      /*IgnoreTemplateOrMacroSubstitution=*/true);
+}
+
+void Converter::EmitFlexibleArrayElementPtr(clang::Expr *array,
+                                            clang::Expr *idx, bool is_mut) {
+  {
+    PushExplicitAutoref no_autoref(*this, std::nullopt);
+    Convert(array);
+  }
+  StrCat(is_mut ? ".as_mut_ptr()" : ".as_ptr()", ".add");
+  {
+    PushParen call(*this);
+    {
+      PushParen paren(*this);
+      Convert(idx);
+    }
+    StrCat(keyword::kAs, "usize");
+  }
+}
+
 void Converter::ConvertArraySubscript(clang::Expr *base, clang::Expr *idx,
                                       clang::QualType type) {
+  if (auto inner = base->IgnoreImplicit()) {
+    if (inner->getType()->isArrayType() &&
+        IsFlexibleArrayMemberAccess(ctx_, inner)) {
+      PushParen outer(*this);
+      StrCat(token::kStar);
+      EmitFlexibleArrayElementPtr(inner, idx,
+                                  !inner->getType().isConstQualified());
+      return;
+    }
+  }
   if (IsUniquePtr(base->getType())) {
     PushExplicitAutoref no_autoref(*this, std::nullopt);
     Convert(base->IgnoreImplicit());
@@ -3725,6 +3760,19 @@ void Converter::ConvertUnsignedArithBinaryOperator(clang::BinaryOperator *op,
 
 void Converter::ConvertAddrOf(clang::Expr *expr, clang::QualType pointer_type) {
   assert(pointer_type->isPointerType());
+  if (auto ase =
+          clang::dyn_cast<clang::ArraySubscriptExpr>(expr->IgnoreParens())) {
+    auto base = ase->getBase();
+    auto inner = base->IgnoreImplicit();
+    if (base->IgnoreCasts()->getType()->isArrayType() &&
+        IsFlexibleArrayMemberAccess(ctx_, inner)) {
+      EmitFlexibleArrayElementPtr(
+          inner, ase->getIdx(),
+          !pointer_type->getPointeeType().isConstQualified());
+      computed_expr_type_ = ComputedExprType::FreshPointer;
+      return;
+    }
+  }
   if (IsReferenceType(expr) || pointer_type->isFunctionPointerType()) {
     PushExprKind push(*this, ExprKind::AddrOf);
     Convert(expr);

cpp2rust/converter/converter.h

@@ -443,6 +443,9 @@ class Converter : public clang::RecursiveASTVisitor<Converter> {
   virtual void ConvertArraySubscript(clang::Expr *base, clang::Expr *idx,
                                      clang::QualType type);
 
+  void EmitFlexibleArrayElementPtr(clang::Expr *array, clang::Expr *idx,
+                                   bool is_mut);
+
   virtual void ConvertAssignment(clang::Expr *lhs, clang::Expr *rhs,
                                  std::string_view assign_operator);
 

tests/unit/out/unsafe/union_field_alignment.rs

@@ -31,7 +31,7 @@ pub fn main() {
 unsafe fn main_0() -> i32 {
     let mut n: node = <node>::default();
     n.next = std::ptr::null_mut();
-    n.x.bytes[(0) as usize] = 171_u8;
-    assert!(((((n.x.bytes[(0) as usize] as i32) == (171)) as i32) != 0));
+    (*n.x.bytes.as_mut_ptr().add((0) as usize)) = 171_u8;
+    assert!((((((*n.x.bytes.as_mut_ptr().add((0) as usize)) as i32) == (171)) as i32) != 0));
     return 0;
 }

tests/unit/out/unsafe/union_flex_array_member.rs

@@ -0,0 +1,63 @@
+extern crate libc;
+use libc::*;
+extern crate libcc2rs;
+use libcc2rs::*;
+use std::collections::BTreeMap;
+use std::io::{Read, Seek, Write};
+use std::os::fd::{AsFd, FromRawFd, IntoRawFd};
+use std::rc::Rc;
+#[repr(C)]
+#[derive(Copy, Clone)]
+pub union anon_0 {
+    pub bytes: [u8; 1],
+    pub aligner: *mut ::libc::c_void,
+}
+impl Default for anon_0 {
+    fn default() -> Self {
+        unsafe { std::mem::zeroed() }
+    }
+}
+#[repr(C)]
+#[derive(Copy, Clone, Default)]
+pub struct node {
+    pub len: u64,
+    pub pos: u64,
+    pub x: anon_0,
+}
+pub fn main() {
+    unsafe {
+        std::process::exit(main_0() as i32);
+    }
+}
+unsafe fn main_0() -> i32 {
+    let mut tail_size: u64 = 32_u64;
+    let mut n: *mut node = (libcc2rs::malloc_unsafe(
+        (::std::mem::size_of::<node>() as u64 as u64).wrapping_add(tail_size),
+    ) as *mut node);
+    (*n).len = tail_size;
+    let mut i: u64 = 0_u64;
+    'loop_: while ((((i) < (tail_size)) as i32) != 0) {
+        (*(*n).x.bytes.as_mut_ptr().add((i) as usize)) = (((i) & (255_u64)) as u8);
+        i.postfix_inc();
+    }
+    let mut i: u64 = 0_u64;
+    'loop_: while ((((i) < (tail_size)) as i32) != 0) {
+        assert!(
+            (((((*(*n).x.bytes.as_mut_ptr().add((i) as usize)) as i32)
+                == ((((i) & (255_u64)) as u8) as i32)) as i32)
+                != 0)
+        );
+        i.postfix_inc();
+    }
+    let mut p: *mut u8 = ((*n).x.bytes.as_mut_ptr().add((10) as usize));
+    assert!((((((*p) as i32) == (10)) as i32) != 0));
+    (*p) = 170_u8;
+    assert!((((((*(*n).x.bytes.as_mut_ptr().add((10) as usize)) as i32) == (170)) as i32) != 0));
+    (*n).pos = 20_u64;
+    let mut q: *mut u8 = ((*n).x.bytes.as_mut_ptr().add(((*n).pos) as usize));
+    assert!((((((*q) as i32) == (20)) as i32) != 0));
+    (*q) = 187_u8;
+    assert!((((((*q) as i32) == (187)) as i32) != 0));
+    libcc2rs::free_unsafe((n as *mut node as *mut ::libc::c_void));
+    return 0;
+}

tests/unit/union_flex_array_member.c

@@ -1,4 +1,4 @@
-// no-compile
+// no-compile: refcount
 #include <assert.h>
 #include <stddef.h>
 #include <stdint.h>
@@ -7,6 +7,7 @@
 
 struct node {
   size_t len;
+  size_t pos;
   union {
     uint8_t bytes[1];
     void *aligner;
@@ -30,6 +31,12 @@ int main(void) {
   *p = 0xAA;
   assert(n->x.bytes[10] == 0xAA);
 
+  n->pos = 20;
+  uint8_t *q = &n->x.bytes[n->pos];
+  assert(*q == 20);
+  *q = 0xBB;
+  assert(*q == 0xBB);
+
   free(n);
   return 0;
 }