[lldb-dap] Refactoring IO handling for the SBDebugger.

This refactors the IO handling of lldb-dap and its SBDebugger. This adjusts the SBDebugger instance to have a pty associated with in/out/err. Using a pty allows us to handle commands with raw input modes, such as `script` or `breakpoint command add`.

Additionally, to better handle output produced by the debugger and evaluate command I added a print helper. This new print helper will inspect the SBCommandReturnObject and store any associated variables in the variable store. This lets users more easily inspect variables directly in the Debug Console in VSCode.

Author: ashgti

lldb/packages/Python/lldbsuite/test/tools/lldb-dap/dap_server.py

@@ -987,17 +987,23 @@ def request_writeMemory(self, memoryReference, data, offset=0, allowPartial=Fals
         }
         return self._send_recv(command_dict)
 
-    def request_evaluate(self, expression, frameIndex=0, threadId=None, context=None):
+    def request_evaluate(
+        self,
+        expression: str,
+        frameIndex=0,
+        threadId: Optional[int] = None,
+        context: Optional[
+            Literal["watch", "repl", "hover", "clipboard", "variables"]
+        ] = None,
+    ) -> Optional[Response]:
         stackFrame = self.get_stackFrame(frameIndex=frameIndex, threadId=threadId)
-        if stackFrame is None:
-            return []
         args_dict = {
             "expression": expression,
             "frameId": stackFrame["id"],
         }
         if context:
             args_dict["context"] = context
-        command_dict = {
+        command_dict: Request = {
             "command": "evaluate",
             "type": "request",
             "arguments": args_dict,

lldb/test/API/tools/lldb-dap/cancel/TestDAP_cancel.py

@@ -2,8 +2,6 @@
 Test lldb-dap cancel request
 """
 
-import time
-
 from lldbsuite.test.decorators import *
 from lldbsuite.test.lldbtest import *
 import lldbdap_testcase

lldb/test/API/tools/lldb-dap/evaluate/TestDAP_evaluate.py

@@ -125,6 +125,7 @@ def run_test_evaluate_expressions(
         self.assertEvaluate("var1", "20", want_type="int")
         # Empty expression should equate to the previous expression.
         if context == "repl":
+            self.assertEvaluate("p var1", "20")
             self.assertEvaluate("", "20")
         else:
             self.assertEvaluateFailure("")