FazBrowse GitHub Viewer | Trending |
URL:
| Home
Tools: [Download Repo ZIP]   [Original HTTPS Page]

bpo-40521: Make Unicode latin1 singletons per interpreter by vstinner · Pull Request #21101 · python/cpython · GitHub

  1. python
  2. cpython

Repository navigation

bpo-40521: Make Unicode latin1 singletons per interpreter #21101

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
vstinner merged 1 commit into from
Jun 24, 2020
Conversation 0 Commits 1 Checks 0 Files changed
Merged

bpo-40521: Make Unicode latin1 singletons per interpreter #21101

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension .c  (1) .h  (1) .rst  (1) All 3 file types selected
Viewed files
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Unified
Split
Hide whitespace
Diff view
Unified
Split
Hide whitespace
3 changes: 3 additions & 0 deletions Include/internal/pycore_interp.h
Show comments
View file
Open in desktop
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode characters
Show hidden characters
Original file line number Diff line number Diff line change
Expand Up @@ -73,6 +73,9 @@ struct _Py_bytes_state {
struct _Py_unicode_state {
// The empty Unicode object is a singleton to improve performance.
PyObject *empty;
/* Single character Unicode strings in the Latin-1 range are being
shared as well. */
PyObject *latin1[256];
struct _Py_unicode_fs_codec fs_codec;
};

Expand Down
2 changes: 1 addition & 1 deletion Misc/NEWS.d/next/Core and Builtins/2020-05-20-01-17-34.bpo-40521.wvAehI.rst
Show comments
View file
Open in desktop
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode characters
Show hidden characters
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@ Each interpreter now its has own free lists, singletons and caches:
* Free lists: float, tuple, list, dict, frame, context,
asynchronous generator, MemoryError.
* Singletons: empty tuple, empty bytes string, empty Unicode string,
single byte character.
single byte character, single Unicode (latin1) character.
* Slice cache.

They are no longer shared by all interpreters.
74 changes: 32 additions & 42 deletions Objects/unicodeobject.c
Show comments
View file
Open in desktop
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode characters
Show hidden characters
Original file line number Diff line number Diff line change
Expand Up @@ -303,17 +303,6 @@ unicode_decode_utf8(const char *s, Py_ssize_t size,
/* List of static strings. */
static _Py_Identifier *static_strings = NULL;

/* bpo-40521: Latin1 singletons are shared by all interpreters. */
#ifndef EXPERIMENTAL_ISOLATED_SUBINTERPRETERS
# define LATIN1_SINGLETONS
#endif

#ifdef LATIN1_SINGLETONS
/* Single character Unicode strings in the Latin-1 range are being
shared as well. */
static PyObject *unicode_latin1[256] = {NULL};
#endif

/* Fast detection of the most frequent whitespace characters */
const unsigned char _Py_ascii_whitespace[] = {
0, 0, 0, 0, 0, 0, 0, 0,
Expand Down Expand Up @@ -657,9 +646,8 @@ unicode_result_wchar(PyObject *unicode)
if (len == 1) {
wchar_t ch = _PyUnicode_WSTR(unicode)[0];
if ((Py_UCS4)ch < 256) {
PyObject *latin1_char = get_latin1_char((unsigned char)ch);
Py_DECREF(unicode);
return latin1_char;
return get_latin1_char((unsigned char)ch);
}
}

Expand Down Expand Up @@ -692,13 +680,13 @@ unicode_result_ready(PyObject *unicode)
return empty;
}

#ifdef LATIN1_SINGLETONS
if (length == 1) {
const void *data = PyUnicode_DATA(unicode);
int kind = PyUnicode_KIND(unicode);
Py_UCS4 ch = PyUnicode_READ(kind, data, 0);
if (ch < 256) {
PyObject *latin1_char = unicode_latin1[ch];
if (kind == PyUnicode_1BYTE_KIND) {
Py_UCS1 *data = PyUnicode_1BYTE_DATA(unicode);
Py_UCS1 ch = data[0];
struct _Py_unicode_state *state = get_unicode_state();
PyObject *latin1_char = state->latin1[ch];
if (latin1_char != NULL) {
if (unicode != latin1_char) {
Py_INCREF(latin1_char);
Expand All @@ -709,12 +697,14 @@ unicode_result_ready(PyObject *unicode)
else {
assert(_PyUnicode_CheckConsistency(unicode, 1));
Py_INCREF(unicode);
unicode_latin1[ch] = unicode;
state->latin1[ch] = unicode;
return unicode;
}
}
else {
assert(PyUnicode_READ_CHAR(unicode, 0) >= 256);
}
}
#endif

assert(_PyUnicode_CheckConsistency(unicode, 1));
return unicode;
Expand Down Expand Up @@ -1981,18 +1971,18 @@ unicode_dealloc(PyObject *unicode)
static int
unicode_is_singleton(PyObject *unicode)
{
if (unicode == unicode_get_empty()) {
struct _Py_unicode_state *state = get_unicode_state();
if (unicode == state->empty) {
return 1;
}
#ifdef LATIN1_SINGLETONS
PyASCIIObject *ascii = (PyASCIIObject *)unicode;
if (ascii->state.kind != PyUnicode_WCHAR_KIND && ascii->length == 1)
{
Py_UCS4 ch = PyUnicode_READ_CHAR(unicode, 0);
if (ch < 256 && unicode_latin1[ch] == unicode)
if (ch < 256 && state->latin1[ch] == unicode) {
return 1;
}
}
#endif
return 0;
}
#endif
Expand Down Expand Up @@ -2130,17 +2120,15 @@ unicode_write_cstr(PyObject *unicode, Py_ssize_t index,
}

static PyObject*
get_latin1_char(unsigned char ch)
get_latin1_char(Py_UCS1 ch)
{
PyObject *unicode;
struct _Py_unicode_state *state = get_unicode_state();

#ifdef LATIN1_SINGLETONS
unicode = unicode_latin1[ch];
PyObject *unicode = state->latin1[ch];
if (unicode) {
Py_INCREF(unicode);
return unicode;
}
#endif

unicode = PyUnicode_New(1, ch);
if (!unicode) {
Expand All @@ -2150,10 +2138,8 @@ get_latin1_char(unsigned char ch)
PyUnicode_1BYTE_DATA(unicode)[0] = ch;
assert(_PyUnicode_CheckConsistency(unicode, 1));

#ifdef LATIN1_SINGLETONS
Py_INCREF(unicode);
unicode_latin1[ch] = unicode;
#endif
state->latin1[ch] = unicode;
return unicode;
}

Expand All @@ -2164,8 +2150,9 @@ unicode_char(Py_UCS4 ch)

assert(ch <= MAX_UNICODE);

if (ch < 256)
if (ch < 256) {
return get_latin1_char(ch);
}

unicode = PyUnicode_New(1, ch);
if (unicode == NULL)
Expand Down Expand Up @@ -2367,11 +2354,13 @@ _PyUnicode_FromUCS1(const Py_UCS1* u, Py_ssize_t size)
PyObject *res;
unsigned char max_char;

if (size == 0)
if (size == 0) {
_Py_RETURN_UNICODE_EMPTY();
}
assert(size > 0);
if (size == 1)
if (size == 1) {
return get_latin1_char(u[0]);
}

max_char = ucs1lib_find_max_char(u, u + size);
res = PyUnicode_New(size, max_char);
Expand Down Expand Up @@ -5008,8 +4997,9 @@ unicode_decode_utf8(const char *s, Py_ssize_t size,

/* ASCII is equivalent to the first 128 ordinals in Unicode. */
if (size == 1 && (unsigned char)s[0] < 128) {
if (consumed)
if (consumed) {
*consumed = 1;
}
return get_latin1_char((unsigned char)s[0]);
}

Expand Down Expand Up @@ -7176,8 +7166,9 @@ PyUnicode_DecodeASCII(const char *s,
_Py_RETURN_UNICODE_EMPTY();

/* ASCII is equivalent to the first 128 ordinals in Unicode. */
if (size == 1 && (unsigned char)s[0] < 128)
if (size == 1 && (unsigned char)s[0] < 128) {
return get_latin1_char((unsigned char)s[0]);
}

// Shortcut for simple case
PyObject *u = PyUnicode_New(size, 127);
Expand Down Expand Up @@ -16234,12 +16225,11 @@ _PyUnicode_Fini(PyThreadState *tstate)

Py_CLEAR(state->empty);

for (Py_ssize_t i = 0; i < 256; i++) {
Py_CLEAR(state->latin1[i]);
}

if (is_main_interp) {
#ifdef LATIN1_SINGLETONS
for (Py_ssize_t i = 0; i < 256; i++) {
Py_CLEAR(unicode_latin1[i]);
}
#endif
unicode_clear_static_strings();
}

Expand Down
Back | FazBrowse Home | New Git URL