Cut disused recode_encoding logic in _PyBytes_DecodeEscape. (GH-16013)

All call sites pass NULL for `recode_encoding`, so this path is completely untested. That's been true since before Python 3.0. It adds significant complexity to this logic, so it's best to take it out. All call sites now have a literal NULL, and that's been true since commit 768921c eliminated a conditional (`foo ? bar : NULL`) at the call site in Python/ast.c where we're parsing a bytes literal. But even before then, that condition `foo` had been a constant since unadorned string literals started meaning Unicode, in commit 572dbf8 aka v3.0a1~1035 . The `unicode` parameter is already unused, so mark it as unused too. The code that acted on it was also taken out before Python 3.0, in commit 8d30cc0 aka v3.0a1~1031 . The function (PyBytes_DecodeEscape) is exposed in the API, but it's never been documented.

Author: gnprice

Include/bytesobject.h

@@ -77,9 +77,7 @@ PyAPI_FUNC(PyObject *) PyBytes_DecodeEscape(const char *, Py_ssize_t,
 #ifndef Py_LIMITED_API
 /* Helper for PyBytes_DecodeEscape that detects invalid escape chars. */
 PyAPI_FUNC(PyObject *) _PyBytes_DecodeEscape(const char *, Py_ssize_t,
- const char *, Py_ssize_t,
- const char *,
- const char **);
+ const char *, const char **);
 #endif
 
 /* Macro, trading safety for speed */

Include/longobject.h

@@ -74,7 +74,7 @@ PyAPI_FUNC(int) _PyLong_Size_t_Converter(PyObject *, void *);
 #endif
 
 /* Used by Python/mystrtoul.c, _PyBytes_FromHex(),
- _PyBytes_DecodeEscapeRecode(), etc. */
+ _PyBytes_DecodeEscape(), etc. */
 #ifndef Py_LIMITED_API
 PyAPI_DATA(unsigned char) _PyLong_DigitValue[256];
 #endif

Objects/bytesobject.c

@@ -1077,52 +1077,10 @@ _PyBytes_FormatEx(const char *format, Py_ssize_t format_len,
  return NULL;
 }
 
-/* Unescape a backslash-escaped string. If unicode is non-zero,
- the string is a u-literal. If recode_encoding is non-zero,
- the string is UTF-8 encoded and should be re-encoded in the
- specified encoding. */
-
-static char *
-_PyBytes_DecodeEscapeRecode(const char **s, const char *end,
- const char *errors, const char *recode_encoding,
- _PyBytesWriter *writer, char *p)
-{
- PyObject *u, *w;
- const char* t;
-
- t = *s;
- /* Decode non-ASCII bytes as UTF-8. */
- while (t < end && (*t & 0x80))
- t++;
- u = PyUnicode_DecodeUTF8(*s, t - *s, errors);
- if (u == NULL)
- return NULL;
-
- /* Recode them in target encoding. */
- w = PyUnicode_AsEncodedString(u, recode_encoding, errors);
- Py_DECREF(u);
- if (w == NULL)
- return NULL;
- assert(PyBytes_Check(w));
-
- /* Append bytes to output buffer. */
- writer->min_size--; /* subtract 1 preallocated byte */
- p = _PyBytesWriter_WriteBytes(writer, p,
- PyBytes_AS_STRING(w),
- PyBytes_GET_SIZE(w));
- Py_DECREF(w);
- if (p == NULL)
- return NULL;
-
- *s = t;
- return p;
-}
-
+/* Unescape a backslash-escaped string. */
 PyObject *_PyBytes_DecodeEscape(const char *s,
  Py_ssize_t len,
  const char *errors,
- Py_ssize_t unicode,
- const char *recode_encoding,
  const char **first_invalid_escape)
 {
  int c;
@@ -1142,17 +1100,7 @@ PyObject *_PyBytes_DecodeEscape(const char *s,
  end = s + len;
  while (s < end) {
  if (*s != '\\') {
- if (!(recode_encoding && (*s & 0x80))) {
- *p++ = *s++;
- }
- else {
- /* non-ASCII character and need to recode */
- p = _PyBytes_DecodeEscapeRecode(&s, end,
- errors, recode_encoding,
- &writer, p);
- if (p == NULL)
- goto failed;
- }
+ *p++ = *s++;
  continue;
  }
 
@@ -1241,12 +1189,11 @@ PyObject *_PyBytes_DecodeEscape(const char *s,
 PyObject *PyBytes_DecodeEscape(const char *s,
  Py_ssize_t len,
  const char *errors,
- Py_ssize_t unicode,
- const char *recode_encoding)
+ Py_ssize_t Py_UNUSED(unicode),
+ const char *Py_UNUSED(recode_encoding))
 {
  const char* first_invalid_escape;
- PyObject *result = _PyBytes_DecodeEscape(s, len, errors, unicode,
- recode_encoding,
+ PyObject *result = _PyBytes_DecodeEscape(s, len, errors,
  &first_invalid_escape);
  if (result == NULL)
  return NULL;

Python/ast.c

@@ -4766,7 +4766,7 @@ decode_bytes_with_escapes(struct compiling *c, const node *n, const char *s,
  size_t len)
 {
  const char *first_invalid_escape;
- PyObject *result = _PyBytes_DecodeEscape(s, len, NULL, 0, NULL,
+ PyObject *result = _PyBytes_DecodeEscape(s, len, NULL,
  &first_invalid_escape);
  if (result == NULL)
  return NULL;