golang_str: bstr/ustr pickle support

Without explicitly overriding __reduce_ex__ pickling was failing for
protocols < 2:

    _________________________ test_strings_pickle __________________________

        def test_strings_pickle():
            bs = b("мир")
            us = u("май")

            #from pickletools import dis
            for proto in range(0, pickle.HIGHEST_PROTOCOL):
    >           p_bs = pickle.dumps(bs, proto)

    golang/golang_str_test.py:282:
    _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

    self = b'\xd0\xbc\xd0\xb8\xd1\x80', proto = 0

        def _reduce_ex(self, proto):
    >       assert proto < 2
    E       RecursionError: maximum recursion depth exceeded in comparison

    /usr/lib/python3.9/copyreg.py:56: RecursionError

See added comments for details.

golang/_golang_str.pyx

@@ -41,6 +41,11 @@ from libc.stdint cimport uint8_t
 pystrconv = None  # = golang.strconv imported at runtime (see __init__.py)
 import types as pytypes
 import functools as pyfunctools
+if PY_MAJOR_VERSION >= 3:
+    import copyreg as pycopyreg
+else:
+    import copy_reg as pycopyreg
+
 
 def pyb(s): # -> bstr
     """b converts object to bstr.
@@ -244,6 +249,18 @@ class pybstr(bytes):
             return self
 
 
+    # override reduce for protocols < 2. Builtin handler for that goes through
+    # copyreg._reduce_ex which eventually calls bytes(bstr-instance) to
+    # retrieve state, which gives bstr, not bytes. Fix state to be bytes ourselves.
+    def __reduce_ex__(self, protocol):
+        if protocol >= 2:
+            return bytes.__reduce_ex__(self, protocol)
+        return (
+            pycopyreg._reconstructor,
+            (self.__class__, self.__class__, _bdata(self))
+        )
+
+
     def __hash__(self):
         # hash of the same unicode and UTF-8 encoded bytes is generally different
         # -> we can't make hash(bstr) == both hash(bytes) and hash(unicode) at the same time.
@@ -345,6 +362,18 @@ class pyustr(unicode):
             return pyb(self)
 
 
+    # override reduce for protocols < 2. Builtin handler for that goes through
+    # copyreg._reduce_ex which eventually calls unicode(ustr-instance) to
+    # retrieve state, which gives ustr, not unicode. Fix state to be unicode ourselves.
+    def __reduce_ex__(self, protocol):
+        if protocol >= 2:
+            return unicode.__reduce_ex__(self, protocol)
+        return (
+            pycopyreg._reconstructor,
+            (self.__class__, self.__class__, _udata(self))
+        )
+
+
     def __hash__(self):
         # see pybstr.__hash__ for why we stick to hash of current str
         if PY_MAJOR_VERSION >= 3:

golang/golang_str_test.py

@@ -31,6 +31,7 @@ import sys
 import six
 from six import text_type as unicode, unichr
 from six.moves import range as xrange
+import pickle
 import array
 
 
@@ -271,6 +272,26 @@ def test_strings_memoryview():
     assert _(5) == 0x80
 
 
+# verify that bstr/ustr can be pickled/unpickled correctly.
+def test_strings_pickle():
+    bs = b("мир")
+    us = u("май")
+
+    #from pickletools import dis
+    for proto in range(0, pickle.HIGHEST_PROTOCOL):
+        p_bs = pickle.dumps(bs, proto)
+        #dis(p_bs)
+        bs_ = pickle.loads(p_bs)
+        assert type(bs_) is bstr
+        assert bs_ == bs
+
+        p_us = pickle.dumps(us, proto)
+        #dis(p_us)
+        us_ = pickle.loads(p_us)
+        assert type(us_) is ustr
+        assert us_ == us
+
+
 # verify that ord on bstr/ustr works as expected.
 def test_strings_ord():
     with raises(TypeError): ord(b(''))