Tracking PR: intermediate state of completed str/bytes cleaning up PRs

include/pybind11/cast.h

@@ -1631,6 +1631,14 @@ struct pyobject_caster {
 
     template <typename T = type, enable_if_t<std::is_base_of<object, T>::value, int> = 0>
     bool load(handle src, bool /* convert */) {
+#ifndef PYBIND11_DISABLE_IMPLICIT_STR_FROM_BYTES
+        if (std::is_same<T, str>::value && isinstance<bytes>(src)) {
+            PyObject *str_from_bytes = PyUnicode_FromEncodedObject(src.ptr(), "utf-8", nullptr);
+            if (!str_from_bytes) throw error_already_set();
+            value = reinterpret_steal<type>(str_from_bytes);
+            return true;
+        }
+#endif
         if (!isinstance<type>(src))
             return false;
         value = reinterpret_borrow<type>(src);

include/pybind11/pytypes.h

@@ -752,8 +752,6 @@ inline bool PyIterable_Check(PyObject *obj) {
 inline bool PyNone_Check(PyObject *o) { return o == Py_None; }
 inline bool PyEllipsis_Check(PyObject *o) { return o == Py_Ellipsis; }
 
-inline bool PyUnicode_Check_Permissive(PyObject *o) { return PyUnicode_Check(o) || PYBIND11_BYTES_CHECK(o); }
-
 inline bool PyStaticMethod_Check(PyObject *o) { return o->ob_type == &PyStaticMethod_Type; }
 
 class kwargs_proxy : public handle {
@@ -796,7 +794,9 @@ PYBIND11_NAMESPACE_END(detail)
         Name(handle h, stolen_t) : Parent(h, stolen_t{}) { } \
         PYBIND11_DEPRECATED("Use py::isinstance<py::python_type>(obj) instead") \
         bool check() const { return m_ptr != nullptr && (bool) CheckFun(m_ptr); } \
-        static bool check_(handle h) { return h.ptr() != nullptr && CheckFun(h.ptr()); }
+        static bool check_(handle h) { return h.ptr() != nullptr && CheckFun(h.ptr()); } \
+        template <typename Policy_> \
+        Name(const ::pybind11::detail::accessor<Policy_> &a) : Name(object(a)) { }
 
 #define PYBIND11_OBJECT_CVT(Name, Parent, CheckFun, ConvertFun) \
     PYBIND11_OBJECT_COMMON(Name, Parent, CheckFun) \
@@ -806,9 +806,7 @@ PYBIND11_NAMESPACE_END(detail)
     { if (!m_ptr) throw error_already_set(); } \
     Name(object &&o) \
     : Parent(check_(o) ? o.release().ptr() : ConvertFun(o.ptr()), stolen_t{}) \
-    { if (!m_ptr) throw error_already_set(); } \
-    template <typename Policy_> \
-    Name(const ::pybind11::detail::accessor<Policy_> &a) : Name(object(a)) { }
+    { if (!m_ptr) throw error_already_set(); }
 
 #define PYBIND11_OBJECT(Name, Parent, CheckFun) \
     PYBIND11_OBJECT_COMMON(Name, Parent, CheckFun) \
@@ -899,7 +897,7 @@ class bytes;
 
 class str : public object {
 public:
-    PYBIND11_OBJECT_CVT(str, object, detail::PyUnicode_Check_Permissive, raw_str)
+    PYBIND11_OBJECT_CVT(str, object, PyUnicode_Check, raw_str)
 
     str(const char *c, size_t n)
         : object(PyUnicode_FromStringAndSize(c, (ssize_t) n), stolen_t{}) {
@@ -920,7 +918,7 @@ class str : public object {
         Return a string representation of the object. This is analogous to
         the ``str()`` function in Python.
     \endrst */
-    explicit str(handle h) : object(raw_str(h.ptr()), stolen_t{}) { }
+    explicit str(handle h) : object(raw_str(h.ptr()), stolen_t{}) { if (!m_ptr) throw error_already_set(); }
 
     operator std::string() const {
         object temp = *this;
@@ -945,8 +943,8 @@ class str : public object {
     /// Return string representation -- always returns a new reference, even if already a str
     static PyObject *raw_str(PyObject *op) {
         PyObject *str_value = PyObject_Str(op);
-        if (!str_value) throw error_already_set();
 #if PY_MAJOR_VERSION < 3
+        if (!str_value) throw error_already_set();
         PyObject *unicode = PyUnicode_FromEncodedObject(str_value, "utf-8", nullptr);
         Py_XDECREF(str_value); str_value = unicode;
 #endif

include/pybind11/stl.h

@@ -144,7 +144,7 @@ template <typename Type, typename Value> struct list_caster {
     using value_conv = make_caster<Value>;
 
     bool load(handle src, bool convert) {
-        if (!isinstance<sequence>(src) || isinstance<str>(src))
+        if (!isinstance<sequence>(src) || isinstance<bytes>(src) || isinstance<str>(src))
             return false;
         auto s = reinterpret_borrow<sequence>(src);
         value.clear();

tests/test_pytypes.cpp

@@ -80,6 +80,7 @@ TEST_SUBMODULE(pytypes, m) {
     m.def("str_from_bytes", []() { return py::str(py::bytes("boo", 3)); });
     m.def("str_from_object", [](const py::object& obj) { return py::str(obj); });
     m.def("repr_from_object", [](const py::object& obj) { return py::repr(obj); });
+    m.def("str_from_handle", [](py::handle h) { return py::str(h); });
 
     m.def("str_format", []() {
         auto s1 = "{} + {} = {}"_s.format(1, 2, 3);
@@ -197,6 +198,7 @@ TEST_SUBMODULE(pytypes, m) {
     // test_constructors
     m.def("default_constructors", []() {
         return py::dict(
+            "bytes"_a=py::bytes(),
             "str"_a=py::str(),
             "bool"_a=py::bool_(),
             "int"_a=py::int_(),
@@ -210,6 +212,7 @@ TEST_SUBMODULE(pytypes, m) {
 
     m.def("converting_constructors", [](py::dict d) {
         return py::dict(
+            "bytes"_a=py::bytes(d["bytes"]),
             "str"_a=py::str(d["str"]),
             "bool"_a=py::bool_(d["bool"]),
             "int"_a=py::int_(d["int"]),
@@ -225,6 +228,7 @@ TEST_SUBMODULE(pytypes, m) {
     m.def("cast_functions", [](py::dict d) {
         // When converting between Python types, obj.cast<T>() should be the same as T(obj)
         return py::dict(
+            "bytes"_a=d["bytes"].cast<py::bytes>(),
             "str"_a=d["str"].cast<py::str>(),
             "bool"_a=d["bool"].cast<py::bool_>(),
             "int"_a=d["int"].cast<py::int_>(),
@@ -369,4 +373,11 @@ TEST_SUBMODULE(pytypes, m) {
             buf, static_cast<ssize_t>(strlen(buf)));
     });
 #endif
+
+    m.def("isinstance_pybind11_bytes", [](py::object o) { return py::isinstance<py::bytes>(o); });
+    m.def("isinstance_pybind11_str", [](py::object o) { return py::isinstance<py::str>(o); });
+
+    m.def("pass_to_pybind11_bytes", [](py::bytes b) { return py::len(b); });
+    m.def("pass_to_pybind11_str", [](py::str s) { return py::len(s); });
+    m.def("pass_to_std_string", [](std::string s) { return s.size(); });
 }

tests/test_pytypes.py

@@ -104,11 +104,22 @@ def __repr__(self):
 
     assert m.str_from_object(A()) == "this is a str"
     assert m.repr_from_object(A()) == "this is a repr"
+    assert m.str_from_handle(A()) == "this is a str"
 
     s1, s2 = m.str_format()
     assert s1 == "1 + 2 = 3"
     assert s1 == s2
 
+    malformed_utf8 = b"\x80"
+    if env.PY2:
+        with pytest.raises(UnicodeDecodeError):
+            assert m.str_from_object(malformed_utf8)
+        with pytest.raises(UnicodeDecodeError):
+            assert m.str_from_handle(malformed_utf8)
+    else:
+        assert m.str_from_object(malformed_utf8) == "b'\\x80'"
+        assert m.str_from_handle(malformed_utf8) == "b'\\x80'"
+
 
 def test_bytes(doc):
     assert m.bytes_from_string().decode() == "foo"
@@ -190,11 +201,17 @@ def func(self, x, *args):
 
 def test_constructors():
     """C++ default and converting constructors are equivalent to type calls in Python"""
-    types = [str, bool, int, float, tuple, list, dict, set]
+    types = [bytes, str, bool, int, float, tuple, list, dict, set]
     expected = {t.__name__: t() for t in types}
+    if env.PY2:
+        # Note that bytes.__name__ == 'str' in Python 2.
+        # pybind11::str is unicode even under Python 2.
+        expected["bytes"] = bytes()
+        expected["str"] = u""  # flake8 complains about unicode().
     assert m.default_constructors() == expected
 
     data = {
+        bytes: b'41',  # Currently no supported or working conversions.
         str: 42,
         bool: "Not empty",
         int: "42",
@@ -207,6 +224,11 @@ def test_constructors():
     }
     inputs = {k.__name__: v for k, v in data.items()}
     expected = {k.__name__: k(v) for k, v in data.items()}
+    if env.PY2:  # Similar to the above. See comments above.
+        inputs["bytes"] = b'41'
+        inputs["str"] = 42
+        expected["bytes"] = b'41'
+        expected["str"] = u"42"
 
     assert m.converting_constructors(inputs) == expected
     assert m.cast_functions(inputs) == expected
@@ -245,13 +267,20 @@ def test_pybind11_str_raw_str():
     valid_orig = u"DZ"
     valid_utf8 = valid_orig.encode("utf-8")
     valid_cvt = cvt(valid_utf8)
-    assert type(valid_cvt) == bytes  # Probably surprising.
-    assert valid_cvt == b'\xc7\xb1'
+    assert type(valid_cvt) is unicode if env.PY2 else str  # noqa: F821
+    if env.PY2:
+        assert valid_cvt == valid_orig
+    else:
+        assert valid_cvt == u"b'\\xc7\\xb1'"
 
     malformed_utf8 = b'\x80'
-    malformed_cvt = cvt(malformed_utf8)
-    assert type(malformed_cvt) == bytes  # Probably surprising.
-    assert malformed_cvt == b'\x80'
+    if env.PY2:
+        with pytest.raises(UnicodeDecodeError):
+            cvt(malformed_utf8)
+    else:
+        malformed_cvt = cvt(malformed_utf8)
+        assert type(malformed_cvt) is unicode if env.PY2 else str  # noqa: F821
+        assert malformed_cvt == u"b'\\x80'"
 
 
 def test_implicit_casting():
@@ -379,3 +408,27 @@ def test_memoryview_from_memory():
     assert isinstance(view, memoryview)
     assert view.format == 'B'
     assert bytes(view) == b'\xff\xe1\xab\x37'
+
+
+def test_isinstance_string_types():
+    assert m.isinstance_pybind11_bytes(b"")
+    assert not m.isinstance_pybind11_bytes(u"")
+
+    assert m.isinstance_pybind11_str(u"")
+    assert not m.isinstance_pybind11_str(b"")
+
+
+def test_pass_bytes_or_unicode_to_string_types():
+    assert m.pass_to_pybind11_bytes(b"Bytes") == 5
+    with pytest.raises(TypeError):
+        m.pass_to_pybind11_bytes(u"Str")  # NO implicit encode
+
+    assert m.pass_to_pybind11_str(b"Bytes") == 5  # implicit decode
+    assert m.pass_to_pybind11_str(u"Str") == 3
+
+    assert m.pass_to_std_string(b"Bytes") == 5
+    assert m.pass_to_std_string(u"Str") == 3
+
+    malformed_utf8 = b"\x80"
+    with pytest.raises(UnicodeDecodeError):
+        m.pass_to_pybind11_str(malformed_utf8)