[foreign-memaccess+abi] RFR: Improve exception messages and simplify LayoutPath [v8]

[Maurizio Cimadamore]

On Wed, 20 Sep 2023 16:42:07 GMT, Per Minborg <pminborg at openjdk.org> wrote:

>> This PR proposes to improve the exception messages in the internal `LayoutPath` class and also do some code simplifications.
>
> Per Minborg has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Rework bredcrumbs appearance

> Could I make a suggestion?
> 
> For the following `MemoryLayout`:
> 
> ```java
>         MemoryLayout struct = MemoryLayout.structLayout(
>                 MemoryLayout.sequenceLayout(2, MemoryLayout.structLayout(
>                                 JAVA_INT.withOrder(ByteOrder.LITTLE_ENDIAN).withName("3a"),
>                                 JAVA_INT.withOrder(ByteOrder.LITTLE_ENDIAN).withName("3b")
>                         ).withName("2")
>                 ).withName("1")
>         ).withName("0");
> ```
> 
> I find the notation like:
> 
> ```
>                 "[i4(3a)i4(3b)](2), selected from: " +
>                 "[2:[i4(3a)i4(3b)](2)](1), selected from: " +
>                 "[[2:[i4(3a)i4(3b)](2)](1)](0)";
> ```
> 
> Difficult to interpret. Maybe something like the below could be clearer/easier-to-read?
> 
> ```
> (0: struct)
> ├── (1: sequence) x 2
> │   ├── (2: struct)      ERROR: Invalid sequence selection!
> │   │   ├── (3a: i4 LE)
> │   │   └── (3b: i4 LE)
> ```

Honestly I find the notation hard to parse too - but, I have to admit, I wonder if a structural representation, such as the one in this PR, or the one in the comment above really "fit" an exception message - they seem more akin to compiler messages to me.

-------------

PR Comment: https://git.openjdk.org/panama-foreign/pull/886#issuecomment-1731747920