FIX: Binary data padding + tests

mssql_python/cursor.py

@@ -389,9 +389,9 @@ def _map_sql_type(self, param, parameters_list, i):
                     False,
                 )
             return (
-                ddbc_sql_const.SQL_BINARY.value,
+                ddbc_sql_const.SQL_VARBINARY.value,  # Use VARBINARY instead of BINARY
                 ddbc_sql_const.SQL_C_BINARY.value,
-                len(param),
+                max(len(param), 1),  # Ensure minimum column size of 1
                 0,
                 False,
             )
@@ -406,9 +406,9 @@ def _map_sql_type(self, param, parameters_list, i):
                     True,
                 )
             return (
-                ddbc_sql_const.SQL_BINARY.value,
+                ddbc_sql_const.SQL_VARBINARY.value,  # Use VARBINARY instead of BINARY
                 ddbc_sql_const.SQL_C_BINARY.value,
-                len(param),
+                max(len(param), 1),  # Ensure minimum column size of 1
                 0,
                 False,
             )
@@ -849,6 +849,8 @@ def _select_best_sample_value(column):
             return max(non_nulls, key=lambda s: len(str(s)))
         if all(isinstance(v, datetime.datetime) for v in non_nulls):
             return datetime.datetime.now()
+        if all(isinstance(v, (bytes, bytearray)) for v in non_nulls):
+            return max(non_nulls, key=lambda b: len(b))
         if all(isinstance(v, datetime.date) for v in non_nulls):
             return datetime.date.today()
         return non_nulls[0]  # fallback

mssql_python/pybind/ddbc_bindings.cpp

@@ -1356,7 +1356,13 @@
                                 std::string offending = WideToUTF8(wstr);
                                 ThrowStdException("Input string exceeds allowed column size at parameter index " + std::to_string(paramIndex));
                             }
+#if defined(__APPLE__) || defined(__linux__)
+                            auto utf16Buf = WStringToSQLWCHAR(wstr);
+                            size_t copySize = std::min(utf16Buf.size(), static_cast<size_t>(info.columnSize + 1));
+                            std::memcpy(wcharArray + i * (info.columnSize + 1), utf16Buf.data(), copySize * sizeof(SQLWCHAR));
+#else
                             std::memcpy(wcharArray + i * (info.columnSize + 1), wstr.c_str(), (wstr.length() + 1) * sizeof(SQLWCHAR));
+#endif
                             strLenOrIndArray[i] = SQL_NTS;
                         }
                     }
@@ -1420,7 +1426,12 @@
                             std::string str = columnValues[i].cast<std::string>();
                             if (str.size() > info.columnSize)
                                 ThrowStdException("Input exceeds column size at index " + std::to_string(i));
-                            std::memcpy(charArray + i * (info.columnSize + 1), str.c_str(), str.size());
+                            // Clear the entire buffer slot first
+                            std::memset(charArray + i * (info.columnSize + 1), 0, info.columnSize + 1);
+                            // Then copy the actual data
+                            if (str.size() > 0) {
+                                std::memcpy(charArray + i * (info.columnSize + 1), str.c_str(), str.size());
+                            }
                             strLenOrIndArray[i] = static_cast<SQLLEN>(str.size());
                         }
                     }
@@ -1778,6 +1789,9 @@
                         }
 				    } else if (dataLen == SQL_NULL_DATA) {
 					    row.append(py::none());
+                    } else if (dataLen == 0) {
+                        // Empty string
+                        row.append(std::string(""));
                     } else {
                         assert(dataLen == SQL_NO_TOTAL);
                         LOG("SQLGetData couldn't determine the length of the data. "
@@ -1834,6 +1848,9 @@
                         }
 				    } else if (dataLen == SQL_NULL_DATA) {
 					    row.append(py::none());
+                    } else if (dataLen == 0) {
+                        // Empty string
+                        row.append(py::str(""));
                     } else {
                         assert(dataLen == SQL_NO_TOTAL);
                         LOG("SQLGetData couldn't determine the length of the data. "
@@ -2030,6 +2047,9 @@
                         }
 				    } else if (dataLen == SQL_NULL_DATA) {
 					    row.append(py::none());
+                    } else if (dataLen == 0) {
+                        // Empty bytes
+                        row.append(py::bytes(""));
                     } else {
                         assert(dataLen == SQL_NO_TOTAL);
                         LOG("SQLGetData couldn't determine the length of the data. "
@@ -2318,7 +2338,7 @@
                 row.append(py::none());
                 continue;
             }
-            assert(dataLen > 0 && "Must be > 0 since SQL_NULL_DATA & SQL_NO_DATA is already handled");
+            assert(dataLen >= 0 && "Data length must be >= 0");
 
             switch (dataType) {
                 case SQL_CHAR: