Eric Snow

As to why we destroy the shared lock in the original interpreter, it is because the lock should be deallocated using the same allocator that allocated its memory. For now...

Thanks, @benedwards14. I updated the BPO title and component and am starting a review of the PR.

See: https://hg.python.org/cpython/file/09b223827f63/Lib/test/test_capi.py#l344 https://hg.python.org/cpython/file/default/Programs/_testembed.c https://hg.python.org/cpython/file/09b223827f63/Lib/test/support/__init__.py#l2326 https://hg.python.org/cpython/file/09b223827f63/Modules/_testcapimodule.c#l2615

Thanks @Mariatta. What do you mean by "the second PR number". miss-islington creates the PR with only one PR number in the title (the original PR). The only case currently...

Thanks for the update, @Mariatta (and @methane). If this has already been discussed/decided then I say don't worry about it. Maybe no one considered moving the original PR number to...

FYI, I'm hoping to wrap this up (or get really close) during the PyCon US sprints. I'll be here through Thursday. @pganssle and @erlend-aasland are here too, so I'll get...

FYI, I chatted with @pganssle and @erlend-aasland (and @encukou) at the PyCon US sprints and we resolved on the following plan for beta 2: 1. basic multi-phase init 2. convert...

Here's my tentative plan for the datetime C-API: * all 5 types will stay static and we'll use the trick we use for builtin types to address cross-interpreter issues *...

@neonene does that plan make sense based on the work you have done?

tp_dict, tp_sublasses, and the weakrefs are all a problem. I'd like to see if we can use the same solution as for the builtin static types.