* Initialize _PyPreConfig.dev_mode to -1.
* _PyPreConfig_Read(): coreconfig has the priority over preconfig.
* _PyCoreConfig_Read() now calls _PyPreCmdline_Read() internally.
* config_from_cmdline() now pass _PyPreCmdline to config_read().
* Add _PyPreCmdline_Copy().

Prepare code to move some _PyPreConfig parameters into _PyPreCmdline.
Changes:

* _PyCoreConfig_ReadFromArgv(): remove preconfig parameter,
  use _PyRuntime.preconfig.
* Add _PyPreCmdline_GetPreConfig() (called by _PyPreConfig_Read()).
* Rename _PyPreCmdline_Init() to _PyPreCmdline_SetArgv()
* Factorize _Py_PreInitializeFromPreConfig() code: add
  pyinit_preinit().
* _PyPreConfig_Read() now sets coerce_c_locale to 2 if it must be
  coerced.
* Remove _PyCoreConfig_ReadPreConfig().
* _PyCoreConfig_Write() now copies updated preconfig into _PyRuntime.

bpo-35884: Add string-keys-only microbenchmark for dict access to var_access_benchmark.py (GH-11905)

Set type_attr to NULL after the assignment to stgdict->proto (like
what is done with stgdict after the Py_SETREF() call) so that it is
not decrefed twice on error.

Fix error in commit 2b751555 noticed by Serhiy Storchaka.

Check for sys.abiflags before using since not all platforms have it defined.

We will remove int support from 3.10 or 4.0.

* Add tests for grep findfiles.
* Move findfiles to module function.
* Change findfiles to use os.walk.

Based on a patch by Al Sweigart.

_PyPreConfig_Write() now writes the applied pre-configuration into
_PyRuntimeState.preconfig.

Remove now unneeded imports.

Clarify capturing or suppressing stdout and stderr on the old call APIs.

Do not state that they are equivalent to run() calls when they are not implemented using run as that was misleading. Unlike run they cannot handle stdout or stderr being set to PIPE without a risk of deadlock.

This param was only used once and changed the return type.

Before, an `AttributeError` was raised due to trying to access an attribute that exists on specs but having received `None` instead for a non-existent module.


https://bugs.python.org/issue36298

Clarify that the naming of protocol handler methods shouldn't be literally called "protocol" but should be named after the actual protocol.



https://bugs.python.org/issue35155

`Task.current_task()` and `Task.all_tasks()` will be removed in 3.9.

compiler_call() needs to check if an error occurred during the
maybe_optimize_method_call() call.

If the first PyUnicode_DecodeUTF8() call fails in structseq_repr(),
_PyUnicodeWriter_Dealloc() will be called on an uninitialized
_PyUnicodeWriter.

bpo-36256: Fix bug in parsermodule when parsing if statements

In the parser module, when validating nodes before starting the parsing with to create a ST in "parser_newstobject" there is a problem that appears when two arcs in the same DFA state has transitions with labels with the same type. For example, the DFA for if_stmt has a state with
two labels with the same type: "elif" and "else" (type NAME). The algorithm tries one by one the arcs until the label that starts the arc transition has a label with the same type of the current child label we are trying to accept. In this case, the arc for "elif" comes before the arc for "else"and passes this test (because the current child label is "else" and has the same type as "elif"). This lead to expecting a namedexpr_test (305) instead of a colon (11). The solution is to compare also the string representation (in case there is one) of the labels to see if the transition that we have is the correct one.

Also fixes venvs from the build directory on Windows.

Currently, when arguments on Parser/asdl_c.py are parsed
``ìf`` sentence is used. This PR Propose to use ``elif``
to avoid multiple evaluting of the ifs.





https://bugs.python.org/issue36385

* Improve coverage.
* Note inherent limitations of the accuracy tests


https://bugs.python.org/issue36324