Make numerous improvements

- Python static hello world now 1.8mb
- Python static fully loaded now 10mb
- Python HTTPS client now uses MbedTLS
- Python REPL now completes import stmts
- Increase stack size for Python for now
- Begin synthesizing posixpath and ntpath
- Restore Python \N{UNICODE NAME} support
- Restore Python NFKD symbol normalization
- Add optimized code path for Intel SHA-NI
- Get more Python unit tests passing faster
- Get Python help() pagination working on NT
- Python hashlib now supports MbedTLS PBKDF2
- Make memcpy/memmove/memcmp/bcmp/etc. faster
- Add Mersenne Twister and Vigna to LIBC_RAND
- Provide privileged __printf() for error code
- Fix zipos opendir() so that it reports ENOTDIR
- Add basic chmod() implementation for Windows NT
- Add Cosmo's best functions to Python cosmo module
- Pin function trace indent depth to that of caller
- Show memory diagram on invalid access in MODE=dbg
- Differentiate stack overflow on crash in MODE=dbg
- Add stb_truetype and tools for analyzing font files
- Upgrade to UNICODE 13 and reduce its binary footprint
- COMPILE.COM now logs resource usage of build commands
- Start implementing basic poll() support on bare metal
- Set getauxval(AT_EXECFN) to GetModuleFileName() on NT
- Add descriptions to strerror() in non-TINY build modes
- Add COUNTBRANCH() macro to help with micro-optimizations
- Make error / backtrace / asan / memory code more unbreakable
- Add fast perfect C implementation of μ-Law and a-Law audio codecs
- Make strtol() functions consistent with other libc implementations
- Improve Linenoise implementation (see also github.com/jart/bestline)
- COMPILE.COM now suppresses stdout/stderr of successful build commands

third_party/python/Include/abstract.h

@@ -14,7 +14,6 @@ PyObject *PyObject_Call(PyObject *callable_object, PyObject *args,
 
 #ifndef Py_LIMITED_API
 PyObject *_PyStack_AsTuple(PyObject **stack, Py_ssize_t nargs);
-
 PyObject *_PyStack_AsDict(PyObject **values, PyObject *kwnames);
 
 int _PyStack_UnpackDict(PyObject **args, Py_ssize_t nargs, PyObject *kwargs,

third_party/python/Include/ceval.h

@@ -124,6 +124,7 @@ const char * PyEval_GetFuncDesc(PyObject *);
 PyObject * PyEval_GetCallStats(PyObject *);
 PyObject * PyEval_EvalFrame(struct _frame *);
 PyObject * PyEval_EvalFrameEx(struct _frame *f, int exc);
+#define PyEval_EvalFrameEx(fr,st) PyThreadState_GET()->interp->eval_frame(fr,st)
 #ifndef Py_LIMITED_API
 PyObject * _PyEval_EvalFrameDefault(struct _frame *f, int exc);
 #endif

third_party/python/Include/ezprint.h

@@ -0,0 +1,29 @@
+#ifndef COSMOPOLITAN_THIRD_PARTY_PYTHON_INCLUDE_EZPRINT_H_
+#define COSMOPOLITAN_THIRD_PARTY_PYTHON_INCLUDE_EZPRINT_H_
+#include "libc/calls/calls.h"
+#include "third_party/python/Include/bytesobject.h"
+#include "third_party/python/Include/pyerrors.h"
+#include "third_party/python/Include/unicodeobject.h"
+#if !(__ASSEMBLER__ + __LINKER__ + 0)
+COSMOPOLITAN_C_START_
+
+static void EzPrint(PyObject *x, const char *s) {
+  PyObject *u;
+  if (!s) {
+    dprintf(2, "%s = NULL\n", s);
+  } else if (PyBytes_Check(x)) {
+    dprintf(2, "%s = b%`'.*s\n", s, PyBytes_GET_SIZE(x), PyBytes_AS_STRING(x));
+  } else if ((u = PyUnicode_AsUTF8String(x))) {
+    dprintf(2, "%s = u%`'.*s\n", s, PyBytes_GET_SIZE(u), PyBytes_AS_STRING(u));
+    Py_DECREF(u);
+  } else {
+    PyErr_Clear();
+    dprintf(2, "%s = !!!\n", s);
+  }
+}
+
+#define EZPRINT(x) EzPrint(x, #x)
+
+COSMOPOLITAN_C_END_
+#endif /* !(__ASSEMBLER__ + __LINKER__ + 0) */
+#endif /* COSMOPOLITAN_THIRD_PARTY_PYTHON_INCLUDE_EZPRINT_H_ */

third_party/python/Include/longintrepr.h

@@ -61,6 +61,7 @@ typedef long stwodigits; /* signed variant of twodigits */
 #else
 #error "PYLONG_BITS_IN_DIGIT should be 15 or 30"
 #endif
+
 #define PyLong_BASE	((digit)1 << PyLong_SHIFT)
 #define PyLong_MASK	((digit)(PyLong_BASE - 1))
 

third_party/python/Include/object.h

@@ -790,7 +790,7 @@ void _Py_Dealloc(PyObject *);
     do {                                                \
         PyObject *_py_decref_tmp = (PyObject *)(op);    \
         if (_Py_DEC_REFTOTAL  _Py_REF_DEBUG_COMMA       \
-        --(_py_decref_tmp)->ob_refcnt != 0)             \
+            --(_py_decref_tmp)->ob_refcnt != 0)         \
             _Py_CHECK_REFCNT(_py_decref_tmp)            \
         else                                            \
             _Py_Dealloc(_py_decref_tmp);                \

third_party/python/Include/objimpl.h

@@ -9,12 +9,12 @@
 COSMOPOLITAN_C_START_
 /* clang-format off */
 
-void * PyObject_Malloc(size_t size);
+void * PyObject_Malloc(size_t);
 #if !defined(Py_LIMITED_API) || Py_LIMITED_API+0 >= 0x03050000
-void * PyObject_Calloc(size_t nelem, size_t elsize);
+void * PyObject_Calloc(size_t, size_t);
 #endif
-void * PyObject_Realloc(void *ptr, size_t new_size);
-void PyObject_Free(void *ptr);
+void * PyObject_Realloc(void *, size_t);
+void PyObject_Free(void *);
 
 #ifndef Py_LIMITED_API
 /* This function returns the number of allocated memory blocks, regardless of size */
@@ -24,7 +24,7 @@ Py_ssize_t _Py_GetAllocatedBlocks(void);
 /* Macros */
 #ifdef WITH_PYMALLOC
 #ifndef Py_LIMITED_API
-void _PyObject_DebugMallocStats(FILE *out);
+void _PyObject_DebugMallocStats(FILE *);
 #endif /* #ifndef Py_LIMITED_API */
 #endif
 
@@ -43,8 +43,7 @@ void _PyObject_DebugMallocStats(FILE *out);
 
 /* Functions */
 PyObject * PyObject_Init(PyObject *, PyTypeObject *);
-PyVarObject * PyObject_InitVar(PyVarObject *,
-                                                 PyTypeObject *, Py_ssize_t);
+PyVarObject * PyObject_InitVar(PyVarObject *, PyTypeObject *, Py_ssize_t);
 PyObject * _PyObject_New(PyTypeObject *);
 PyVarObject * _PyObject_NewVar(PyTypeObject *, Py_ssize_t);
 
@@ -243,8 +242,11 @@ extern PyGC_Head *_PyGC_generation0;
 #endif /* Py_LIMITED_API */
 
 #ifndef Py_LIMITED_API
-PyObject * _PyObject_GC_Malloc(size_t size);
-PyObject * _PyObject_GC_Calloc(size_t size);
+PyObject * _PyObject_GC_Alloc(int, size_t);
+PyObject * _PyObject_GC_Malloc(size_t);
+PyObject * _PyObject_GC_Calloc(size_t);
+#define _PyObject_GC_Malloc(sz) _PyObject_GC_Alloc(0, sz)
+#define _PyObject_GC_Callac(sz) _PyObject_GC_Alloc(1, sz)
 #endif /* !Py_LIMITED_API */
 PyObject * _PyObject_GC_New(PyTypeObject *);
 PyVarObject * _PyObject_GC_NewVar(PyTypeObject *, Py_ssize_t);
@@ -257,6 +259,12 @@ void PyObject_GC_Del(void *);
 #define PyObject_GC_NewVar(type, typeobj, n) \
                 ( (type *) _PyObject_GC_NewVar((typeobj), (n)) )
 
+#define PyObject_GC_UnTrack(ARG) do {                   \
+    void *opArg = (ARG);                                \
+    if (_PyGC_REFS(opArg) != _PyGC_REFS_UNTRACKED) {    \
+      _PyObject_GC_UNTRACK(opArg);                      \
+    }                                                   \
+  } while (0)
 
 /* Utility macro to help write tp_traverse functions.
  * To use this macro, the tp_traverse function must name its arguments

third_party/python/Include/pyatomic.h

@@ -1,6 +1,5 @@
 #ifndef Py_ATOMIC_H
 #define Py_ATOMIC_H
-#ifdef Py_BUILD_CORE
 #include "libc/assert.h"
 #include "third_party/python/Include/dynamic_annotations.h"
 #include "third_party/python/pyconfig.h"
@@ -242,5 +241,4 @@ _Py_ANNOTATE_MEMORY_ORDER(const volatile void *address, _Py_memory_order order)
 #define _Py_atomic_load_relaxed(ATOMIC_VAL) \
     _Py_atomic_load_explicit(ATOMIC_VAL, _Py_memory_order_relaxed)
 
-#endif  /* Py_BUILD_CORE */
 #endif  /* Py_ATOMIC_H */

third_party/python/Include/pyerrors.h

@@ -101,7 +101,7 @@ void PyErr_SetExcInfo(PyObject *, PyObject *, PyObject *);
 #endif
 
 /* Defined in Python/pylifecycle.c */
-void Py_FatalError(const char *message) _Py_NO_RETURN;
+void Py_FatalError(const char *message) relegated _Py_NO_RETURN;
 
 #if defined(Py_DEBUG) || defined(Py_LIMITED_API)
 #define _PyErr_OCCURRED() PyErr_Occurred()
@@ -247,7 +247,7 @@ PyObject * PyErr_SetFromErrnoWithFilename(
     PyObject *exc,
     const char *filename   /* decoded from the filesystem encoding */
     );
-#if defined(MS_WINDOWS) && !defined(Py_LIMITED_API)
+#if !defined(Py_LIMITED_API)
 PyObject * PyErr_SetFromErrnoWithUnicodeFilename(
     PyObject *, const Py_UNICODE *);
 #endif /* MS_WINDOWS */
@@ -349,7 +349,6 @@ PyObject * _PyErr_TrySetFromCause(
     );
 #endif
 
-
 /* In sigcheck.c or signalmodule.c */
 int PyErr_CheckSignals(void);
 void PyErr_SetInterrupt(void);
@@ -477,20 +476,6 @@ int PyUnicodeTranslateError_SetReason(
     const char *reason          /* UTF-8 encoded string */
     );
 
-/* These APIs aren't really part of the error implementation, but
-   often needed to format error messages; the native C lib APIs are
-   not available on all platforms, which is why we provide emulations
-   for those platforms in Python/mysnprintf.c,
-   WARNING:  The return value of snprintf varies across platforms; do
-   not rely on any particular behavior; eventually the C99 defn may
-   be reliable.
-*/
-#if defined(MS_WIN32) && !defined(HAVE_SNPRINTF)
-# define HAVE_SNPRINTF
-# define snprintf _snprintf
-# define vsnprintf _vsnprintf
-#endif
-
 int PyOS_snprintf(char *str, size_t size, const char  *format, ...)
                         Py_GCC_ATTRIBUTE((format(printf, 3, 4)));
 int PyOS_vsnprintf(char *str, size_t size, const char  *format, va_list va)

third_party/python/Include/pystate.h

@@ -206,6 +206,9 @@ PyThreadState * PyThreadState_Get(void);
 /* Similar to PyThreadState_Get(), but don't issue a fatal error
  * if it is NULL. */
 PyThreadState * _PyThreadState_UncheckedGet(void);
+#define _PyThreadState_UncheckedGet() \
+  ((PyThreadState *)_Py_atomic_load_relaxed(&_PyThreadState_Current))
+
 #endif /* !Py_LIMITED_API */
 
 PyThreadState * PyThreadState_Swap(PyThreadState *);

third_party/python/Include/tupleobject.h

@@ -37,7 +37,7 @@ extern PyTypeObject PyTupleIter_Type;
                  PyType_FastSubclass(Py_TYPE(op), Py_TPFLAGS_TUPLE_SUBCLASS)
 #define PyTuple_CheckExact(op) (Py_TYPE(op) == &PyTuple_Type)
 
-PyObject * PyTuple_New(Py_ssize_t size);
+PyObject * PyTuple_New(Py_ssize_t);
 Py_ssize_t PyTuple_Size(PyObject *);
 PyObject * PyTuple_GetItem(PyObject *, Py_ssize_t);
 int PyTuple_SetItem(PyObject *, Py_ssize_t, PyObject *);
@@ -45,7 +45,7 @@ PyObject * PyTuple_GetSlice(PyObject *, Py_ssize_t, Py_ssize_t);
 #ifndef Py_LIMITED_API
 int _PyTuple_Resize(PyObject **, Py_ssize_t);
 #endif
-PyObject * PyTuple_Pack(Py_ssize_t, ...);
+PyObject *PyTuple_Pack(Py_ssize_t, ...);
 #ifndef Py_LIMITED_API
 void _PyTuple_MaybeUntrack(PyObject *);
 #endif

third_party/python/Include/unicodeobject.h

@@ -118,7 +118,7 @@ typedef uint8_t Py_UCS1;
 #define Py_UNICODE_MATCH(string, offset, substring) \
     ((*((string)->wstr + (offset)) == *((substring)->wstr)) && \
      ((*((string)->wstr + (offset) + (substring)->wstr_length-1) == *((substring)->wstr + (substring)->wstr_length-1))) && \
-     !memcmp((string)->wstr + (offset), (substring)->wstr, (substring)->wstr_length*sizeof(Py_UNICODE)))
+     !bcmp((string)->wstr + (offset), (substring)->wstr, (substring)->wstr_length*sizeof(Py_UNICODE)))
 
 #endif /* Py_LIMITED_API */