Brian Beckman

I narrowed this down to `dtype=numpy.int16` causes the exception, but `dtype=int16` with `from numpy import int16` avoids the error. Here is a full program that avoids the error: ```python from...

got it

``` └─(12:33:29 on vector-backend)──> PYTHONPATH=".:../src/runtime/lpython:.." LPYTHON_PY_MOD_NAME="" LPYTHON_PY_MOD_PATH="_lpython-tmp-test-cpython" python bindc_03.py Traceback (most recent call last): File "/Users/brian/Dropbox/Mac/Documents/GitHub/lpython/integration_tests/bindc_03.py", line 12, in def g(a: CPtr, value: i32, offset_value: bool) -> None: File "/Users/brian/Dropbox/Mac/Documents/GitHub/lpython/src/runtime/lpython/lpython.py",...

I'm leaving this open. bindc_03.py doesn't work in CPython, but most of the other bindc programs do work with the ENVIRONMENT variables set as above. I verified by hand ```...

Another one (Grammarly missed this!) "constant" should be plural "constants" ``` Only those local variables that can be reduced to compile-time constants should be used in dimensions of an array....

Some more possibilities: ```python # The following does not work in lpython (Issue #2492) # A_nm: i16[n, m] = empty((n, m), dtype=int16) # doesn't work # A_nm : i16[:] =...

The following produces an unhandled exception ```python A_nm : Array[i16[:]] = empty((n, m), dtype=int16) ``` ``` (lp) ┌─(~/Documents/GitHub/lpython/integration_tests)────────────────────────────────────────────────────────────────────────────────(brian@MacBook-Pro:s001)─┐ └─(12:48:58 on vector-backend ✹ ✭)──> lpython matmul_integration.py ──(Tue,Feb06)─┘ Internal Compiler Error: Unhandled...

Thanks! I'm leaving it open because of the uncaught exception, but your answer with `Const` unblocks my work!

marking `severe` because the workaround is nasty (reverse-engineering `run_tests.py`

The following is a workaround. Leaving this open as a Design Change Request because I would like to see either documentation or a more discoverable / intuitive design for returning...