Suppose we have:

<script type="py" src="./code.py"></script>

code.py:

async def func(arg: int) -> None:
    assert arg

async def main() -> None:
    await func(0)

if __name__ == '__main__':
    await main()

At execution, the following stack trace appears in console:

PythonError: Traceback (most recent call last):
  File "/lib/python313.zip/_pyodide/_base.py", line 597, in eval_code_async
    await CodeRunner(
    ...<9 lines>...
    .run_async(globals, locals)
  File "/lib/python313.zip/_pyodide/_base.py", line 413, in run_async
    await coroutine
  File "<exec>", line 8, in <module>
  File "<exec>", line 5, in main
  File "<exec>", line 2, in func
AssertionError

    at new_error (pyodide.asm.js:8:578922)
    at pyodide.asm.wasm:0x10f082
...

Those <exec> patterns instead of file names are not very informative.

Of course, when we only have one Python <script>, it's pretty evident that <exec> means code.py.

But in a larger project, with workers, it could be difficult to quickly understand which process crashed, and which file to lookup those line numbers in.

As far as I understand Python, those <exec> things appear because file name was not specified when Python code was compiled.

As far as I understand PyScript, is starts Python scripts using (await loadPyodide(...)).runPythonAsync(), which, as far as I could find, is always called with code only, without any options. But if {filename: './code.py'} is passed to it, it should replace those <exec> strings, and should also be reflected in global __file__ variable and in sys.modules['__main__'].__file__ too.

Is it possible to pass <script>'s src attribute as filename to runPythonAsync() options?
That sounds logical and probably would resolve the problem.