Handle PyLong_FromLong() and PyUnicode_FromString() failures

on memory allocation failure

Instead of ignoring the memory allocation failure and create invalid objects.

allocating a bigger table failed

PyMemoTable destructor does crash if mt_table is NULL.

memory allocation failure

Replace also PyMem_Free() with PyMem_FREE() to be consistent with the rest of
the function.

fails with an assertion error

exception (when getting NULL) to an assertion to detect bugs earlier

the newly created pickler, and not PyObject_GC_Del().

raised if the result is NULL to help to find bugs in C mode (get the error
earlier than the SystemError in ceval.c).

The caller gets an error code and can raise a classic Python exception.

 * create_new_element() initializes all attributes before handling errors,
   to fix a crash in the destructor
* create_new_element() calls PyObject_GC_Del() on error, instead of
  PyObject_Del(), because the object was created by PyObject_GC_New()
* subelement() now handles create_new_element() failure
* element_getattro() now handles element_get_text() failure
* makeuniversal() now handles PyBytes_FromStringAndSize() failure

attributes before handling errors

_Pickler_New() now calls PyObject_GC_Del() instead of Py_DECREF() on error,
because the pickle object is created using PyObject_GC_New().

Fix a crash in the destructor when an attribute is not initiallized.

The caller gets an error code and can raise a classic Python exception.

If an attribute is not initialized, the destructor can crash

type->tp_str must not point to slot_tp_str() if type has no __str__ attribute,
so there is no reason for slot_tp_str() to fallback on slot_tp_str() on lookup
error. Moreover, calling PyErr_Clear() may hide a real bug like MemoryError.

If __str__ attribute is removed, slots must be updated (which is done by
type_setattro()).

The problem was I was only checking for decimal digits after the third '?',
not for *hex* digits :(.

This changeset also fixes a couple of comment typos, deletes an unused
function relating to encoded word parsing, and removed an invalid
'if' test from the folding function that was revealed by the tests
written to validate this issue.

more bytes than requested.

more bytes than requested.

do with byte strings.

Added tests for Tcl.split() and Tcl.splitline().

do with byte strings.

Added tests for Tcl.split() and Tcl.splitline().

not to stderr.  Patch by Berker Peksag and Michael Dickens.

Patch by Dmi Baranov.

PyModule_GetDef() returns an error.

PyModule_GetDef() returns an error.

Also fixes a resource warning that occurred when the fallback is taken.

Patch by Serhiy Storchaka.

(We couldn't figure out how to write tests for this.)