Stowage/cpython
2
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-11-01 14:11:41 +00:00
Code Issues Projects Releases Packages Wiki Activity

gh-139156: Optimize the UTF-7 encoder (#139253)

Browse source 
Remove base64SetO and base64WhiteSpace parameters.
This commit is contained in:
Victor Stinner 2025-09-24 17:57:29 +02:00 • committed by GitHub
parent c7b11b7546
commit 8d83b7df3f
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 6 additions and 13 deletions
Download patch file Download diff file Expand all files Collapse all files

15
Objects/unicodeobject.c
View file

@ -4670,15 +4670,12 @@ char utf7_category[128] = {
/* ENCODE_DIRECT: this character should be encoded as itself. The
* answer depends on whether we are encoding set O as itself, and also
* on whether we are encoding whitespace as itself. RFC2152 makes it
* on whether we are encoding whitespace as itself. RFC 2152 makes it
* clear that the answers to these questions vary between
* applications, so this code needs to be flexible. */
#define ENCODE_DIRECT(c, directO, directWS) \
((c) < 128 && (c) > 0 && \
((utf7_category[(c)] == 0) || \
(directWS && (utf7_category[(c)] == 2)) || \
(directO && (utf7_category[(c)] == 1))))
#define ENCODE_DIRECT(c) \
((c) < 128 && (c) > 0 && ((utf7_category[(c)] != 3)))
PyObject *
PyUnicode_DecodeUTF7(const char *s,
@ -4895,8 +4892,6 @@ PyUnicode_DecodeUTF7Stateful(const char *s,
PyObject *
_PyUnicode_EncodeUTF7(PyObject *str,
int base64SetO,
int base64WhiteSpace,
const char *errors)
{
Py_ssize_t len = PyUnicode_GET_LENGTH(str);
@ -4923,7 +4918,7 @@ _PyUnicode_EncodeUTF7(PyObject *str,
Py_UCS4 ch = PyUnicode_READ(kind, data, i);
if (inShift) {
if (ENCODE_DIRECT(ch, !base64SetO, !base64WhiteSpace)) {
if (ENCODE_DIRECT(ch)) {
/* shifting out */
if (base64bits) { /* output remaining bits */
*out++ = TO_BASE64(base64buffer << (6-base64bits));
@ -4947,7 +4942,7 @@ _PyUnicode_EncodeUTF7(PyObject *str,
*out++ = '+';
*out++ = '-';
}
else if (ENCODE_DIRECT(ch, !base64SetO, !base64WhiteSpace)) {
else if (ENCODE_DIRECT(ch)) {
*out++ = (char) ch;
}
else {