greg/cpython
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

gh-55531: Implement normalize_encoding in C (#136643)

Browse Source 
Closes gh-55531
This commit is contained in:
Stan Ulbrych
2025-10-30 14:31:47 +00:00
committed by GitHub
parent 6826166280
commit a3ce2f77f0
6 changed files with 123 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
Objects/unicodeobject.c
View File

@@ -3449,13 +3449,14 @@ PyUnicode_FromEncodedObject(PyObject *obj,
return v;
}
/* Normalize an encoding name: similar to encodings.normalize_encoding(), but
also convert to lowercase. Return 1 on success, or 0 on error (encoding is
longer than lower_len-1). */
/* Normalize an encoding name like encodings.normalize_encoding()
but allow to convert to lowercase if *to_lower* is true.
Return 1 on success, or 0 on error (encoding is longer than lower_len-1). */
int
_Py_normalize_encoding(const char *encoding,
char *lower,
size_t lower_len)
size_t lower_len,
int to_lower)
{
const char *e;
char *l;
@@ -3486,7 +3487,7 @@ _Py_normalize_encoding(const char *encoding,
if (l == l_end) {
return 0;
}
*l++ = Py_TOLOWER(c);
*l++ = to_lower ? Py_TOLOWER(c) : c;
}
else {
punct = 1;
@@ -3521,7 +3522,7 @@ PyUnicode_Decode(const char *s,
}
/* Shortcuts for common default encodings */
if (_Py_normalize_encoding(encoding, buflower, sizeof(buflower))) {
if (_Py_normalize_encoding(encoding, buflower, sizeof(buflower), 1)) {
char *lower = buflower;
/* Fast paths */
@@ -3778,7 +3779,7 @@ PyUnicode_AsEncodedString(PyObject *unicode,
}
/* Shortcuts for common default encodings */
if (_Py_normalize_encoding(encoding, buflower, sizeof(buflower))) {
if (_Py_normalize_encoding(encoding, buflower, sizeof(buflower), 1)) {
char *lower = buflower;
/* Fast paths */