Skip to content

gh-140328: Use interned versions of string constants if they're already present #140688

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
wants to merge 3 commits into
base: main
Choose a base branch
from
Draft

gh-140328: Use interned versions of string constants if they're already present #140688

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
db85987
this reuses SOME interned strings, but not `utf-8` and friends
albertedwardson Oct 27, 2025
e448a20
Update codeobject.c: - unnecessary check
albertedwardson Oct 27, 2025
ba291d3
correct error handling, refcount interned_dict
albertedwardson Oct 28, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
44 changes: 33 additions & 11 deletions Objects/codeobject.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -113,7 +113,7 @@ PyCode_ClearWatcher(int watcher_id)

#define _PyCodeObject_CAST(op) (assert(PyCode_Check(op)), (PyCodeObject *)(op))

static int
static inline int
should_intern_string(PyObject *o)
{
#ifdef Py_GIL_DISABLED
Expand Down Expand Up @@ -196,17 +196,34 @@ intern_strings(PyObject *tuple)
return 0;
}


/* Intern constants. In the default build, this interns selected string
constants. In the free-threaded build, this also interns non-string
constants. */
static int
intern_constants(PyObject *tuple, int *modified)
{
PyInterpreterState *interp = _PyInterpreterState_GET();
PyObject *interned_dict = _Py_INTERP_CACHED_OBJECT(interp, interned_strings);
Py_INCREF(interned_dict);
for (Py_ssize_t i = PyTuple_GET_SIZE(tuple); --i >= 0; ) {
PyObject *v = PyTuple_GET_ITEM(tuple, i);
if (PyUnicode_CheckExact(v)) {
if (should_intern_string(v)) {
if (PyUnicode_CheckExact(v) && PyUnicode_GET_LENGTH(v) > 1) {
if (PyUnicode_CHECK_INTERNED(v) != 0) {
continue;
}
PyObject *interned = PyDict_GetItemWithError(interned_dict, v);
if (interned == NULL && PyErr_Occurred()) {
goto error;
}
if (interned != NULL && interned != v) {
Py_INCREF(interned);
PyTuple_SET_ITEM(tuple, i, interned);
Py_DECREF(v);
if (modified) {
*modified = 1;
}
} else if (should_intern_string(v)) {
PyObject *w = v;
_PyUnicode_InternMortal(interp, &v);
if (w != v) {
Expand All @@ -219,25 +236,25 @@ intern_constants(PyObject *tuple, int *modified)
}
else if (PyTuple_CheckExact(v)) {
if (intern_constants(v, NULL) < 0) {
return -1;
goto error;
}
}
else if (PyFrozenSet_CheckExact(v)) {
PyObject *w = v;
PyObject *tmp = PySequence_Tuple(v);
if (tmp == NULL) {
return -1;
goto error;
}
int tmp_modified = 0;
if (intern_constants(tmp, &tmp_modified) < 0) {
Py_DECREF(tmp);
return -1;
goto error;
}
if (tmp_modified) {
v = PyFrozenSet_New(tmp);
if (v == NULL) {
Py_DECREF(tmp);
return -1;
goto error;
}

PyTuple_SET_ITEM(tuple, i, v);
Expand All @@ -253,23 +270,23 @@ intern_constants(PyObject *tuple, int *modified)
PySliceObject *slice = (PySliceObject *)v;
PyObject *tmp = PyTuple_New(3);
if (tmp == NULL) {
return -1;
goto error;
}
PyTuple_SET_ITEM(tmp, 0, Py_NewRef(slice->start));
PyTuple_SET_ITEM(tmp, 1, Py_NewRef(slice->stop));
PyTuple_SET_ITEM(tmp, 2, Py_NewRef(slice->step));
int tmp_modified = 0;
if (intern_constants(tmp, &tmp_modified) < 0) {
Py_DECREF(tmp);
return -1;
goto error;
}
if (tmp_modified) {
v = PySlice_New(PyTuple_GET_ITEM(tmp, 0),
PyTuple_GET_ITEM(tmp, 1),
PyTuple_GET_ITEM(tmp, 2));
if (v == NULL) {
Py_DECREF(tmp);
return -1;
goto error;
}
PyTuple_SET_ITEM(tuple, i, v);
Py_DECREF(slice);
Expand All @@ -288,7 +305,7 @@ intern_constants(PyObject *tuple, int *modified)
{
PyObject *interned = intern_one_constant(v);
if (interned == NULL) {
return -1;
goto error;
}
else if (interned != v) {
PyTuple_SET_ITEM(tuple, i, interned);
Expand All @@ -300,7 +317,12 @@ intern_constants(PyObject *tuple, int *modified)
}
#endif
}
Py_DECREF(interned_dict);
return 0;

error:
Py_DECREF(interned_dict);
return -1;
}

/* Return a shallow copy of a tuple that is
Expand Down
Loading