Soundness: Creating MapArray from ArrayData does not perform bound checks on reading offsets

[jorgecarleitao]

Equivalent to #773 for the MapArray.

use std::convert::TryFrom;
use std::sync::Arc;

use arrow::array::*;
use arrow::buffer::*;
use arrow::datatypes::*;

fn main() {
    let data = ArrayData::new(
        DataType::Map(
            Box::new(Field::new(
                "entries",
                DataType::Struct(vec![
                    Field::new("keys", DataType::Utf8, false),
                    Field::new("values", DataType::Utf8, false),
                ]),
                false,
            )),
            false,
        ),
        10,
        None,
        None,
        0,
        vec![Buffer::from_slice_ref(&[0i32, 10])],
        vec![StructArray::try_from(vec![
            (
                "keys",
                Arc::new(StringArray::from(vec!["hello", "", "parquet"])) as ArrayRef,
            ),
            (
                "values",
                Arc::new(StringArray::from(vec!["hello", "", "parquet"])) as ArrayRef,
            ),
        ])
        .unwrap()
        .data_ref()
        .clone()],
    );
    let a = MapArray::from(data);
    let b = a.value(1);
    let b = b.as_any().downcast_ref::<StructArray>().unwrap();
    println!("{:?}", b);
}

error: Undefined Behavior: using uninitialized data, but this operation requires initialized memory
  --> /home/azureuser/projects/arrow-rs/arrow/src/array/array_map.rs:73:19
   |
73 |         let end = self.value_offsets()[i + 1] as usize;
   |                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ using uninitialized data, but this operation requires initialized memory
   |
   = help: this indicates a bug in the program: it performed an invalid operation, and caused Undefined Behavior
   = help: see https://doc.rust-lang.org/nightly/reference/behavior-considered-undefined.html for further information
           
   = note: inside `arrow::array::MapArray::value` at /home/azureuser/projects/arrow-rs/arrow/src/array/array_map.rs:73:19
note: inside `main` at arrow/examples/unsafe.rs:41:13
  --> arrow/examples/unsafe.rs:41:13
   |
41 |     let b = a.value(1);
   |             ^^^^^^^^^^
   = note: inside `<fn() as std::ops::FnOnce<()>>::call_once - shim(fn())` at /home/azureuser/.rustup/toolchains/nightly-2021-07-09-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/core/src/ops/function.rs:227:5
   = note: inside `std::sys_common::backtrace::__rust_begin_short_backtrace::<fn(), ()>` at /home/azureuser/.rustup/toolchains/nightly-2021-07-09-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/std/src/sys_common/backtrace.rs:125:18
   = note: inside closure at /home/azureuser/.rustup/toolchains/nightly-2021-07-09-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/std/src/rt.rs:63:18
   = note: inside `std::ops::function::impls::<impl std::ops::FnOnce<()> for &dyn std::ops::Fn() -> i32 + std::marker::Sync + std::panic::RefUnwindSafe>::call_once` at /home/azureuser/.rustup/toolchains/nightly-2021-07-09-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/core/src/ops/function.rs:259:13
   = note: inside `std::panicking::r#try::do_call::<&dyn std::ops::Fn() -> i32 + std::marker::Sync + std::panic::RefUnwindSafe, i32>` at /home/azureuser/.rustup/toolchains/nightly-2021-07-09-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/std/src/panicking.rs:401:40
   = note: inside `std::panicking::r#try::<i32, &dyn std::ops::Fn() -> i32 + std::marker::Sync + std::panic::RefUnwindSafe>` at /home/azureuser/.rustup/toolchains/nightly-2021-07-09-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/std/src/panicking.rs:365:19
   = note: inside `std::panic::catch_unwind::<&dyn std::ops::Fn() -> i32 + std::marker::Sync + std::panic::RefUnwindSafe, i32>` at /home/azureuser/.rustup/toolchains/nightly-2021-07-09-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/std/src/panic.rs:434:14
   = note: inside closure at /home/azureuser/.rustup/toolchains/nightly-2021-07-09-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/std/src/rt.rs:45:48
   = note: inside `std::panicking::r#try::do_call::<[closure@std::rt::lang_start_internal::{closure#2}], isize>` at /home/azureuser/.rustup/toolchains/nightly-2021-07-09-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/std/src/panicking.rs:401:40
   = note: inside `std::panicking::r#try::<isize, [closure@std::rt::lang_start_internal::{closure#2}]>` at /home/azureuser/.rustup/toolchains/nightly-2021-07-09-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/std/src/panicking.rs:365:19
   = note: inside `std::panic::catch_unwind::<[closure@std::rt::lang_start_internal::{closure#2}], isize>` at /home/azureuser/.rustup/toolchains/nightly-2021-07-09-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/std/src/panic.rs:434:14
   = note: inside `std::rt::lang_start_internal` at /home/azureuser/.rustup/toolchains/nightly-2021-07-09-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/std/src/rt.rs:45:20
   = note: inside `std::rt::lang_start::<()>` at /home/azureuser/.rustup/toolchains/nightly-2021-07-09-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/std/src/rt.rs:62:5

error: aborting due to previous error

[alamb]

Updating title of the ticket to make it clear this affects misusing the lower level APIs, as described in https://github.com/apache/arrow-rs/tree/master/arrow#safety

[alamb]

This is a specific case of the general issue described in #817, so closing this one as a duplicate in favor of the more general ticket.