Skip to content

gh-95382: Use cache for indentations in the JSON encoder #118636

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

Merged
merged 8 commits into from
Nov 12, 2024
Merged

gh-95382: Use cache for indentations in the JSON encoder #118636

Changes from 1 commit
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
2e915fd
gh-95382: Use cache for indentations in the JSON encoder
serhiy-storchaka May 6, 2024
2a7b274
Apply suggestions from code review
serhiy-storchaka May 6, 2024
86c7797
Merge branch 'main' into json-encode-indent-cache
serhiy-storchaka May 6, 2024
a108c34
Merge branch 'main' into json-encode-indent-cache
serhiy-storchaka May 7, 2024
86586a5
Refactoring.
serhiy-storchaka May 7, 2024
5778e86
Merge branch 'main' into json-encode-indent-cache
serhiy-storchaka Jun 29, 2024
21681c5
Merge branch 'main' into json-encode-indent-cache
serhiy-storchaka Sep 28, 2024
0f835d9
Merge branch 'main' into json-encode-indent-cache
serhiy-storchaka Nov 12, 2024
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
174 changes: 106 additions & 68 deletions Modules/_json.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -85,11 +85,11 @@ encoder_dealloc(PyObject *self);
static int
encoder_clear(PyEncoderObject *self);
static int
encoder_listencode_list(PyEncoderObject *s, _PyUnicodeWriter *writer, PyObject *seq, PyObject *newline_indent);
encoder_listencode_list(PyEncoderObject *s, _PyUnicodeWriter *writer, PyObject *seq, Py_ssize_t indent_level, PyObject *indent_cache);
static int
encoder_listencode_obj(PyEncoderObject *s, _PyUnicodeWriter *writer, PyObject *obj, PyObject *newline_indent);
encoder_listencode_obj(PyEncoderObject *s, _PyUnicodeWriter *writer, PyObject *obj, Py_ssize_t indent_level, PyObject *indent_cache);
static int
encoder_listencode_dict(PyEncoderObject *s, _PyUnicodeWriter *writer, PyObject *dct, PyObject *newline_indent);
encoder_listencode_dict(PyEncoderObject *s, _PyUnicodeWriter *writer, PyObject *dct, Py_ssize_t indent_level, PyObject *indent_cache);
static PyObject *
_encoded_const(PyObject *obj);
static void
Expand Down Expand Up @@ -1252,14 +1252,81 @@ encoder_new(PyTypeObject *type, PyObject *args, PyObject *kwds)
}

static PyObject *
_create_newline_indent(PyObject *indent, Py_ssize_t indent_level)
_create_indent_cache(PyEncoderObject *s, Py_ssize_t indent_level)
{
PyObject *newline_indent = PyUnicode_FromOrdinal('\n');
if (newline_indent != NULL && indent_level) {
PyUnicode_AppendAndDel(&newline_indent,
PySequence_Repeat(indent, indent_level));
PySequence_Repeat(s->indent, indent_level));
}
return newline_indent;
if (newline_indent == NULL) {
return NULL;
}
PyObject *indent_cache = PyList_New(2);
if (indent_cache == NULL) {
Py_XDECREF(newline_indent);
return NULL;
}
PyList_SET_ITEM(indent_cache, 0, newline_indent);
PyList_SET_ITEM(indent_cache, 1, Py_NewRef(Py_None)); // not used
return indent_cache;
}

static int
update_newline_indent(PyEncoderObject *s,
Py_ssize_t indent_level, PyObject *indent_cache)
{
assert(indent_level * 2 == PyList_GET_SIZE(indent_cache));
assert(indent_level > 0);
PyObject *newline_indent = PyList_GET_ITEM(indent_cache, (indent_level - 1)*2);
newline_indent = PyUnicode_Concat(newline_indent, s->indent);
if (newline_indent == NULL) {
return -1;
}
if (PyList_Append(indent_cache, newline_indent) < 0) {
Py_DECREF(newline_indent);
return -1;
}
PyObject *separator_indent = PyUnicode_Concat(s->item_separator, newline_indent);
Py_DECREF(newline_indent);
if (PyList_Append(indent_cache, separator_indent) < 0) {
Py_DECREF(separator_indent);
return -1;
}
Py_DECREF(separator_indent);
return 0;
}

static PyObject *
do_indent(PyEncoderObject *s, _PyUnicodeWriter *writer,
Py_ssize_t indent_level, PyObject *indent_cache)
{
assert(indent_level > 0);
assert(s->indent != Py_None);
PyObject *newline_indent;
if (indent_level * 2 == PyList_GET_SIZE(indent_cache)) {
if (update_newline_indent(s, indent_level, indent_cache) < 0) {
return NULL;
}
}
assert(indent_level * 2 <= PyList_GET_SIZE(indent_cache) - 2);

newline_indent = PyList_GET_ITEM(indent_cache, indent_level * 2);
if (_PyUnicodeWriter_WriteStr(writer, newline_indent) < 0) {
return NULL;
}
return PyList_GET_ITEM(indent_cache, indent_level * 2 + 1);
}

static int
do_dedent(PyEncoderObject *s, _PyUnicodeWriter *writer,
Py_ssize_t indent_level, PyObject *indent_cache)
{
assert(indent_level >= 0);
assert(indent_level * 2 <= PyList_GET_SIZE(indent_cache) - 4);
assert(s->indent != Py_None);
PyObject *newline_indent = PyList_GET_ITEM(indent_cache, indent_level * 2);
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
PyObject *newline_indent = PyList_GET_ITEM(indent_cache, indent_level * 2);
assert(indent_level * 2 < PyList_GET_SIZE(indent_cache));
PyObject *newline_indent = PyList_GET_ITEM(indent_cache, indent_level * 2);

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The assert above is placed after conditional resizing the cache. It is useful to check that we always get the right cache size, independently from conditions. But here we do not change the cache.

return _PyUnicodeWriter_WriteStr(writer, newline_indent);
}

static PyObject *
Expand All @@ -1278,20 +1345,20 @@ encoder_call(PyEncoderObject *self, PyObject *args, PyObject *kwds)
_PyUnicodeWriter_Init(&writer);
writer.overallocate = 1;

PyObject *newline_indent = NULL;
PyObject *indent_cache = NULL;
if (self->indent != Py_None) {
newline_indent = _create_newline_indent(self->indent, indent_level);
if (newline_indent == NULL) {
indent_cache = _create_indent_cache(self, indent_level);
if (indent_cache == NULL) {
_PyUnicodeWriter_Dealloc(&writer);
return NULL;
}
}
if (encoder_listencode_obj(self, &writer, obj, newline_indent)) {
if (encoder_listencode_obj(self, &writer, obj, 0, indent_cache)) {
_PyUnicodeWriter_Dealloc(&writer);
Py_XDECREF(newline_indent);
Py_XDECREF(indent_cache);
return NULL;
}
Py_XDECREF(newline_indent);
Py_XDECREF(indent_cache);

result = PyTuple_New(1);
if (result == NULL ||
Expand Down Expand Up @@ -1379,7 +1446,8 @@ _steal_accumulate(_PyUnicodeWriter *writer, PyObject *stolen)

static int
encoder_listencode_obj(PyEncoderObject *s, _PyUnicodeWriter *writer,
PyObject *obj, PyObject *newline_indent)
PyObject *obj,
Py_ssize_t indent_level, PyObject *indent_cache)
{
/* Encode Python object obj to a JSON term */
PyObject *newobj;
Expand Down Expand Up @@ -1415,14 +1483,14 @@ encoder_listencode_obj(PyEncoderObject *s, _PyUnicodeWriter *writer,
else if (PyList_Check(obj) || PyTuple_Check(obj)) {
if (_Py_EnterRecursiveCall(" while encoding a JSON object"))
return -1;
rv = encoder_listencode_list(s, writer, obj, newline_indent);
rv = encoder_listencode_list(s, writer, obj, indent_level, indent_cache);
_Py_LeaveRecursiveCall();
return rv;
}
else if (PyDict_Check(obj)) {
if (_Py_EnterRecursiveCall(" while encoding a JSON object"))
return -1;
rv = encoder_listencode_dict(s, writer, obj, newline_indent);
rv = encoder_listencode_dict(s, writer, obj, indent_level, indent_cache);
_Py_LeaveRecursiveCall();
return rv;
}
Expand Down Expand Up @@ -1456,7 +1524,7 @@ encoder_listencode_obj(PyEncoderObject *s, _PyUnicodeWriter *writer,
Py_XDECREF(ident);
return -1;
}
rv = encoder_listencode_obj(s, writer, newobj, newline_indent);
rv = encoder_listencode_obj(s, writer, newobj, indent_level, indent_cache);
_Py_LeaveRecursiveCall();

Py_DECREF(newobj);
Expand All @@ -1478,7 +1546,7 @@ encoder_listencode_obj(PyEncoderObject *s, _PyUnicodeWriter *writer,
static int
encoder_encode_key_value(PyEncoderObject *s, _PyUnicodeWriter *writer, bool *first,
PyObject *key, PyObject *value,
PyObject *newline_indent,
Py_ssize_t indent_level, PyObject *indent_cache,
PyObject *item_separator)
{
PyObject *keystr = NULL;
Expand Down Expand Up @@ -1534,23 +1602,22 @@ encoder_encode_key_value(PyEncoderObject *s, _PyUnicodeWriter *writer, bool *fir
if (_PyUnicodeWriter_WriteStr(writer, s->key_separator) < 0) {
return -1;
}
if (encoder_listencode_obj(s, writer, value, newline_indent) < 0) {
if (encoder_listencode_obj(s, writer, value, indent_level, indent_cache) < 0) {
return -1;
}
return 0;
}

static int
encoder_listencode_dict(PyEncoderObject *s, _PyUnicodeWriter *writer,
PyObject *dct, PyObject *newline_indent)
PyObject *dct,
Py_ssize_t indent_level, PyObject *indent_cache)
{
/* Encode Python dict dct a JSON term */
PyObject *ident = NULL;
PyObject *items = NULL;
PyObject *key, *value;
bool first = true;
PyObject *new_newline_indent = NULL;
PyObject *separator_indent = NULL;

if (PyDict_GET_SIZE(dct) == 0) /* Fast path */
return _PyUnicodeWriter_WriteASCIIString(writer, "{}", 2);
Expand All @@ -1574,19 +1641,11 @@ encoder_listencode_dict(PyEncoderObject *s, _PyUnicodeWriter *writer,
if (_PyUnicodeWriter_WriteChar(writer, '{'))
goto bail;

PyObject *current_item_separator = s->item_separator; // borrowed reference
PyObject *separator = s->item_separator; // borrowed reference
if (s->indent != Py_None) {
new_newline_indent = PyUnicode_Concat(newline_indent, s->indent);
if (new_newline_indent == NULL) {
goto bail;
}
separator_indent = PyUnicode_Concat(current_item_separator, new_newline_indent);
if (separator_indent == NULL) {
goto bail;
}
// update item separator with a borrowed reference
current_item_separator = separator_indent;
if (_PyUnicodeWriter_WriteStr(writer, new_newline_indent) < 0) {
indent_level++;
separator = do_indent(s, writer, indent_level, indent_cache);
if (separator == NULL) {
goto bail;
}
}
Expand All @@ -1607,8 +1666,8 @@ encoder_listencode_dict(PyEncoderObject *s, _PyUnicodeWriter *writer,
key = PyTuple_GET_ITEM(item, 0);
value = PyTuple_GET_ITEM(item, 1);
if (encoder_encode_key_value(s, writer, &first, key, value,
new_newline_indent,
current_item_separator) < 0)
indent_level, indent_cache,
separator) < 0)
goto bail;
}
Py_CLEAR(items);
Expand All @@ -1617,8 +1676,8 @@ encoder_listencode_dict(PyEncoderObject *s, _PyUnicodeWriter *writer,
Py_ssize_t pos = 0;
while (PyDict_Next(dct, &pos, &key, &value)) {
if (encoder_encode_key_value(s, writer, &first, key, value,
new_newline_indent,
current_item_separator) < 0)
indent_level, indent_cache,
separator) < 0)
goto bail;
}
}
Expand All @@ -1629,12 +1688,8 @@ encoder_listencode_dict(PyEncoderObject *s, _PyUnicodeWriter *writer,
Py_CLEAR(ident);
}
if (s->indent != Py_None) {
Py_CLEAR(new_newline_indent);
Py_CLEAR(separator_indent);

if (_PyUnicodeWriter_WriteStr(writer, newline_indent) < 0) {
goto bail;
}
indent_level--;
do_dedent(s, writer, indent_level, indent_cache);
}

if (_PyUnicodeWriter_WriteChar(writer, '}'))
Expand All @@ -1644,20 +1699,17 @@ encoder_listencode_dict(PyEncoderObject *s, _PyUnicodeWriter *writer,
bail:
Py_XDECREF(items);
Py_XDECREF(ident);
Py_XDECREF(separator_indent);
Py_XDECREF(new_newline_indent);
return -1;
}

static int
encoder_listencode_list(PyEncoderObject *s, _PyUnicodeWriter *writer,
PyObject *seq, PyObject *newline_indent)
PyObject *seq,
Py_ssize_t indent_level, PyObject *indent_cache)
{
PyObject *ident = NULL;
PyObject *s_fast = NULL;
Py_ssize_t i;
PyObject *new_newline_indent = NULL;
PyObject *separator_indent = NULL;

ident = NULL;
s_fast = PySequence_Fast(seq, "_iterencode_list needs a sequence");
Expand Down Expand Up @@ -1689,28 +1741,19 @@ encoder_listencode_list(PyEncoderObject *s, _PyUnicodeWriter *writer,

PyObject *separator = s->item_separator; // borrowed reference
if (s->indent != Py_None) {
new_newline_indent = PyUnicode_Concat(newline_indent, s->indent);
if (new_newline_indent == NULL) {
goto bail;
}

if (_PyUnicodeWriter_WriteStr(writer, new_newline_indent) < 0) {
indent_level++;
separator = do_indent(s, writer, indent_level, indent_cache);
if (separator == NULL) {
goto bail;
}

separator_indent = PyUnicode_Concat(separator, new_newline_indent);
if (separator_indent == NULL) {
goto bail;
}
separator = separator_indent; // assign separator with borrowed reference
}
for (i = 0; i < PySequence_Fast_GET_SIZE(s_fast); i++) {
PyObject *obj = PySequence_Fast_GET_ITEM(s_fast, i);
if (i) {
if (_PyUnicodeWriter_WriteStr(writer, separator) < 0)
goto bail;
}
if (encoder_listencode_obj(s, writer, obj, new_newline_indent))
if (encoder_listencode_obj(s, writer, obj, indent_level, indent_cache))
goto bail;
}
if (ident != NULL) {
Expand All @@ -1720,11 +1763,8 @@ encoder_listencode_list(PyEncoderObject *s, _PyUnicodeWriter *writer,
}

if (s->indent != Py_None) {
Py_CLEAR(new_newline_indent);
Py_CLEAR(separator_indent);
if (_PyUnicodeWriter_WriteStr(writer, newline_indent) < 0) {
goto bail;
}
indent_level--;
do_dedent(s, writer, indent_level, indent_cache);
}

if (_PyUnicodeWriter_WriteChar(writer, ']'))
Expand All @@ -1735,8 +1775,6 @@ encoder_listencode_list(PyEncoderObject *s, _PyUnicodeWriter *writer,
bail:
Py_XDECREF(ident);
Py_DECREF(s_fast);
Py_XDECREF(separator_indent);
Py_XDECREF(new_newline_indent);
return -1;
}

Expand Down