Raise codec errors when casting to std::string (#2903)

* Raise codec errors when casting to std::string

Allow the codec's exception to be raised instead of RuntimeError when
casting from py::str to std::string.

PY2 allows ucs surrogates in UTF-8 conversion

Signed-off-by: Shane Loretz <sloretz@openrobotics.org>
Signed-off-by: Shane Loretz <sloretz@osrfoundation.org>

* Attempt to fix py2 error

* Revert all unicode literals

* Fixed

Co-authored-by: Aaron Gokaslan <skylion.aaron@gmail.com>

include/pybind11/pytypes.h

@@ -1015,7 +1015,7 @@ public:
         if (PyUnicode_Check(m_ptr)) {
             temp = reinterpret_steal<object>(PyUnicode_AsUTF8String(m_ptr));
             if (!temp)
-                pybind11_fail("Unable to extract string contents! (encoding issue)");
+                throw error_already_set();
         }
         char *buffer = nullptr;
         ssize_t length = 0;

tests/test_pytypes.cpp

@@ -76,6 +76,9 @@ TEST_SUBMODULE(pytypes, m) {
     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_from_string_from_str", [](const py::str& obj) {
+        return py::str(static_cast<std::string>(obj));
+    });
 
     m.def("str_format", []() {
         auto s1 = "{} + {} = {}"_s.format(1, 2, 3);

tests/test_pytypes.py

@@ -133,6 +133,14 @@ def test_str(doc):
     else:
         assert m.str_from_handle(malformed_utf8) == "b'\\x80'"
 
+    assert m.str_from_string_from_str("this is a str") == "this is a str"
+    ucs_surrogates_str = u"\udcc3"
+    if env.PY2:
+        assert u"\udcc3" == m.str_from_string_from_str(ucs_surrogates_str)
+    else:
+        with pytest.raises(UnicodeEncodeError):
+            m.str_from_string_from_str(ucs_surrogates_str)
+
 
 def test_bytes(doc):
     assert m.bytes_from_string().decode() == "foo"