pythonGH-107596: Specialize str[int] (pythonGH-107597)

miss-islington · Aug 8, 2023 · ea72c6f · ea72c6f
1 parent aab6f71
commit ea72c6f
Show file tree

Hide file tree

Showing 10 changed files with 172 additions and 97 deletions.
diff --git a/Include/internal/pycore_opcode.h b/Include/internal/pycore_opcode.h
diff --git a/Include/internal/pycore_opcode_metadata.h b/Include/internal/pycore_opcode_metadata.h
diff --git a/Include/opcode.h b/Include/opcode.h
diff --git a/Lib/_opcode_metadata.py b/Lib/_opcode_metadata.py
diff --git a/Misc/NEWS.d/next/Core and Builtins/2023-08-03-11-13-09.gh-issue-107596.T3yPGI.rst b/Misc/NEWS.d/next/Core and Builtins/2023-08-03-11-13-09.gh-issue-107596.T3yPGI.rst
@@ -0,0 +1 @@
+Specialize subscripting :class:`str` objects by :class:`int` indexes.
diff --git a/Python/bytecodes.c b/Python/bytecodes.c
@@ -509,6 +509,7 @@ dummy_func(
             BINARY_SUBSCR_DICT,
             BINARY_SUBSCR_GETITEM,
             BINARY_SUBSCR_LIST_INT,
+            BINARY_SUBSCR_STR_INT,
             BINARY_SUBSCR_TUPLE_INT,
         };
 
@@ -574,6 +575,21 @@ dummy_func(
             Py_DECREF(list);
         }
 
+        inst(BINARY_SUBSCR_STR_INT, (unused/1, str, sub -- res)) {
+            DEOPT_IF(!PyLong_CheckExact(sub), BINARY_SUBSCR);
+            DEOPT_IF(!PyUnicode_CheckExact(str), BINARY_SUBSCR);
+            DEOPT_IF(!_PyLong_IsNonNegativeCompact((PyLongObject *)sub), BINARY_SUBSCR);
+            Py_ssize_t index = ((PyLongObject*)sub)->long_value.ob_digit[0];
+            DEOPT_IF(PyUnicode_GET_LENGTH(str) <= index, BINARY_SUBSCR);
+            // Specialize for reading an ASCII character from any string:
+            Py_UCS4 c = PyUnicode_READ_CHAR(str, index);
+            DEOPT_IF(Py_ARRAY_LENGTH(_Py_SINGLETON(strings).ascii) <= c, BINARY_SUBSCR);
+            STAT_INC(BINARY_SUBSCR, hit);
+            res = (PyObject*)&_Py_SINGLETON(strings).ascii[c];
+            _Py_DECREF_SPECIALIZED(sub, (destructor)PyObject_Free);
+            Py_DECREF(str);
+        }
+
         inst(BINARY_SUBSCR_TUPLE_INT, (unused/1, tuple, sub -- res)) {
             DEOPT_IF(!PyLong_CheckExact(sub), BINARY_SUBSCR);
             DEOPT_IF(!PyTuple_CheckExact(tuple), BINARY_SUBSCR);

diff --git a/Python/executor_cases.c.h b/Python/executor_cases.c.h
diff --git a/Python/generated_cases.c.h b/Python/generated_cases.c.h