Include pointer offset in counterexample output

[tautschnig]

While the bit-level pointer output would convey the information, the
human-readable expression previously did not consider non-zero offsets except
for null pointers.

• Each commit message has a non-empty body, explaining why the change was made.
• My contribution is formatted in line with CODING_STANDARD.md.
• n/a Methods or procedures I have added are documented, following the guidelines provided in CODING_STANDARD.md.
• Regression or unit tests are included, or existing tests cover the modified code (in this case I have detailed which ones those are in the commit message).
• n/a My commit message includes data points confirming performance improvements (if claimed).
• My PR is restricted to a single feature or bugfix.
• n/a White-space or formatting changes outside the feature-related changed lines are in commits of their own.

[martin-cs]

Well this seems like an unpleasant omission. Are you sure that pointer_expr() is only used in output? If it is used elsewhere we might want to see if this fixes other things.

I guess if we really wanted to show off we could check if the base object was an array and turn it into &(a[i]). Would this be more human friendly than a + i ?

[peterschrammel]

Remove?

[allredj]

Passed Diffblue compatibility checks (cbmc commit: 8652d9c).
Build URL: https://travis-ci.com/diffblue/test-gen/builds/87494837

[kroening]

You probably meant pointer.offset / *object_size

[tautschnig]

Ooops - that clearly shows that my test case is not good enough. Thanks for catching that! I'll extend my test.

[tautschnig]

@martin-cs said:

[...] If it is used elsewhere we might want to see if this fixes other things.

I need to review #2161 as that has modifications in the same area of code.

[tautschnig]

Marking do-not-merge as this needs to be built on top of the fixes in #2161.

[allredj]

✔️
Passed Diffblue compatibility checks (cbmc commit: e140d16).
Build URL: https://travis-ci.com/diffblue/test-gen/builds/89854653