R08. When the interpreter stops

[tamnd]

R08 is the last lesson about the runtime as a thing with a lifetime. R01 asked what has already happened before your first line, and this one asks what is still to come after your last one.

Every cell starts a child interpreter, because a notebook cannot watch its own shutdown. The cells walk the order of the end with sys.is_finalizing() printed next to each step, show atexit callbacks coming back newest first and one registered during shutdown that never runs at all, ask a late __del__ what it can still reach, let an atexit callback and a finalizer both raise so you can read CPython's own ignored exception wording and watch the exit status stay at zero, catch the single generation two collection with gc.callbacks, and leave a subinterpreter open so its warning and its own atexit callback both turn up.

The case that earns the lesson is the daemon thread. Pass time.sleep and your finalizers run. Pass a function defined in your own module and its stack frame holds your module's globals, so the module dict is never cleared and nothing in it is freed. The two Tier 1 recordings run nine endings on a release build and on a debug build, and the debug one puts a number on it: 12680 references still alive at the end, against zero for the same five thousand objects held any other way.

Twelve citations, all resolving against the pinned source. Six diagrams. Three new glossary terms in the startup and shutdown group: interpreter finalisation, atexit callback, and finalizing flag.

One drive-by fix in R07, which said four of seven sample extensions were refused where the table has eight rows.

Checked with just check. Under Pyodide all nine code cells run and each one says it cannot start a process, which is the honest answer in a browser tab.