You missed my point about IPython:  forget "In/Out arrays, etc".  What you suggest is inadequate for _just_ changing PS1/PS2 for IPython.  Again, read their `parse()` function.  They support _more than one_ set of PS1/PS2 conventions.  So the code changes you suggest wouldn't help them.  So long as they want to support more than one set, they'd still have to write their own `parse()` function just to deal with multiple PS1/PS2 conventions.

For the rest, it's not the number of people using non-CPython shells that's rare, but the number _of_ non-CPython shells.  IPython is popular, they already solved their problem, and the patch wouldn't help them anyway.  Who would it help?

I don't take "a plain CPython shell with customized ps1/ps2" seriously because, as I said, after some decades now I still don't know of anyone who does that (granting that newbies sometimes try it, go 'ha! ha! - cool!", and never try it again).  If any such people exist, I would oppose changing doctest just for them, because the _point_ of doctest is to help write easily understood tests.  Catering to changing shell syntax in random ways opposes that goal - programming is a social activity.

So, sorry, but I remain opposed.  Something that might change my mind:  find an author of a widely used alternative Python shell who says this change would allow them to _remove_ their own doctest-PS1/PS2 customization code.  That would be a real use case.