Since we don't know in advance which Python version the pickles will be loaded in (e.g., scanned by a CI but loaded on a developer machine), we need the union of all stdlib module names across supported Python versions and use that instead. On the other hand, there's also the risk that a PyPI package was created with the same name as a previously-standard module.
Either generate the table with top-level modules ourselves and check it in, or add stdlib_list back and merge their lists (they contain submodules as well), give some control to the user over the target version, and document the defaults and the associated risks.
Since we don't know in advance which Python version the pickles will be loaded in (e.g., scanned by a CI but loaded on a developer machine), we need the union of all stdlib module names across supported Python versions and use that instead. On the other hand, there's also the risk that a PyPI package was created with the same name as a previously-standard module.
Either generate the table with top-level modules ourselves and check it in, or add
stdlib_listback and merge their lists (they contain submodules as well), give some control to the user over the target version, and document the defaults and the associated risks.