closes bpo-39605: Fix some casts to not cast away const. (GH-18453)

gcc -Wcast-qual turns up a number of instances of casting away constness of pointers. Some of these can be safely modified, by either:

Adding the const to the type cast, as in:

-    return _PyUnicode_FromUCS1((unsigned char*)s, size);
+    return _PyUnicode_FromUCS1((const unsigned char*)s, size);

or, Removing the cast entirely, because it's not necessary (but probably was at one time), as in:

-    PyDTrace_FUNCTION_ENTRY((char *)filename, (char *)funcname, lineno);
+    PyDTrace_FUNCTION_ENTRY(filename, funcname, lineno);

These changes will not change code, but they will make it much easier to check for errors in consts
(cherry picked from commit e6be9b59a911626d6597fe148c32f0342bd2bd24)

Co-authored-by: Andy Lester <andy@petdance.com>

diff --git a/Modules/_io/textio.c b/Modules/_io/textio.c
index 2f4a5245681f0b2685d8ac59a4ad6c605c80d6ce..a4bf7cd1076707e9e1602f2824d3dda93ac73dec 100644 (file)
--- a/Modules/_io/textio.c
+++ b/Modules/_io/textio.c
@@ -1985,7 +1985,7 @@ find_control_char(int kind, const char *s, const char *end, Py_UCS4 ch)
 {
     if (kind == PyUnicode_1BYTE_KIND) {
         assert(ch < 256);
-        return (char *) memchr((void *) s, (char) ch, end - s);
+        return (char *) memchr((const void *) s, (char) ch, end - s);
     }
     for (;;) {
         while (PyUnicode_READ(kind, s, 0) > ch)
@@ -2003,7 +2003,7 @@ _PyIO_find_line_ending(
     int translated, int universal, PyObject *readnl,
     int kind, const char *start, const char *end, Py_ssize_t *consumed)
 {
-    Py_ssize_t len = ((char*)end - (char*)start)/kind;
+    Py_ssize_t len = (end - start)/kind;
 
     if (translated) {
         /* Newlines are already translated, only search for \n */

diff --git a/Objects/bytes_methods.c b/Objects/bytes_methods.c
index 7d131842059228598cd5cb952cec002978c7acb1..db030be4fe7561600e8c968759e073981bc8d776 100644 (file)
--- a/Objects/bytes_methods.c
+++ b/Objects/bytes_methods.c
@@ -12,7 +12,7 @@ PyObject*
 _Py_bytes_isspace(const char *cptr, Py_ssize_t len)
 {
     const unsigned char *p
-        = (unsigned char *) cptr;
+        = (const unsigned char *) cptr;
     const unsigned char *e;
 
     /* Shortcut for single character strings */
@@ -42,7 +42,7 @@ PyObject*
 _Py_bytes_isalpha(const char *cptr, Py_ssize_t len)
 {
     const unsigned char *p
-        = (unsigned char *) cptr;
+        = (const unsigned char *) cptr;
     const unsigned char *e;
 
     /* Shortcut for single character strings */
@@ -72,7 +72,7 @@ PyObject*
 _Py_bytes_isalnum(const char *cptr, Py_ssize_t len)
 {
     const unsigned char *p
-        = (unsigned char *) cptr;
+        = (const unsigned char *) cptr;
     const unsigned char *e;
 
     /* Shortcut for single character strings */
@@ -123,7 +123,7 @@ _Py_bytes_isascii(const char *cptr, Py_ssize_t len)
             /* Help allocation */
             const char *_p = p;
             while (_p < aligned_end) {
-                unsigned long value = *(unsigned long *) _p;
+                unsigned long value = *(const unsigned long *) _p;
                 if (value & ASCII_CHAR_MASK) {
                     Py_RETURN_FALSE;
                 }
@@ -154,7 +154,7 @@ PyObject*
 _Py_bytes_isdigit(const char *cptr, Py_ssize_t len)
 {
     const unsigned char *p
-        = (unsigned char *) cptr;
+        = (const unsigned char *) cptr;
     const unsigned char *e;
 
     /* Shortcut for single character strings */
@@ -184,7 +184,7 @@ PyObject*
 _Py_bytes_islower(const char *cptr, Py_ssize_t len)
 {
     const unsigned char *p
-        = (unsigned char *) cptr;
+        = (const unsigned char *) cptr;
     const unsigned char *e;
     int cased;
 
@@ -218,7 +218,7 @@ PyObject*
 _Py_bytes_isupper(const char *cptr, Py_ssize_t len)
 {
     const unsigned char *p
-        = (unsigned char *) cptr;
+        = (const unsigned char *) cptr;
     const unsigned char *e;
     int cased;
 
@@ -254,7 +254,7 @@ PyObject*
 _Py_bytes_istitle(const char *cptr, Py_ssize_t len)
 {
     const unsigned char *p
-        = (unsigned char *) cptr;
+        = (const unsigned char *) cptr;
     const unsigned char *e;
     int cased, previous_is_cased;
 

diff --git a/Objects/memoryobject.c b/Objects/memoryobject.c
index a873ac1ec1ea5f325ba27654158aaf254b9099ff..92050791b5d1674aa138ba74de9a0c24d24a520c 100644 (file)
--- a/Objects/memoryobject.c
+++ b/Objects/memoryobject.c
@@ -1681,8 +1681,8 @@ unpack_single(const char *ptr, const char *fmt)
     switch (fmt[0]) {
 
     /* signed integers and fast path for 'B' */
-    case 'B': uc = *((unsigned char *)ptr); goto convert_uc;
-    case 'b': ld =   *((signed char *)ptr); goto convert_ld;
+    case 'B': uc = *((const unsigned char *)ptr); goto convert_uc;
+    case 'b': ld =   *((const signed char *)ptr); goto convert_ld;
     case 'h': UNPACK_SINGLE(ld, ptr, short); goto convert_ld;
     case 'i': UNPACK_SINGLE(ld, ptr, int); goto convert_ld;
     case 'l': UNPACK_SINGLE(ld, ptr, long); goto convert_ld;
@@ -2683,8 +2683,8 @@ unpack_cmp(const char *p, const char *q, char fmt,
     switch (fmt) {
 
     /* signed integers and fast path for 'B' */
-    case 'B': return *((unsigned char *)p) == *((unsigned char *)q);
-    case 'b': return *((signed char *)p) == *((signed char *)q);
+    case 'B': return *((const unsigned char *)p) == *((const unsigned char *)q);
+    case 'b': return *((const signed char *)p) == *((const signed char *)q);
     case 'h': CMP_SINGLE(p, q, short); return equal;
     case 'i': CMP_SINGLE(p, q, int); return equal;
     case 'l': CMP_SINGLE(p, q, long); return equal;

diff --git a/Objects/stringlib/asciilib.h b/Objects/stringlib/asciilib.h
index d0fc18d22fa5f36b59176fcf1f47569b9a44e2fa..8f83614488e844c0d830b739456fe9870fb730d8 100644 (file)
--- a/Objects/stringlib/asciilib.h
+++ b/Objects/stringlib/asciilib.h
@@ -18,7 +18,7 @@
 #define STRINGLIB_TODECIMAL      Py_UNICODE_TODECIMAL
 #define STRINGLIB_STR            PyUnicode_1BYTE_DATA
 #define STRINGLIB_LEN            PyUnicode_GET_LENGTH
-#define STRINGLIB_NEW(STR,LEN)   _PyUnicode_FromASCII((char*)(STR),(LEN))
+#define STRINGLIB_NEW(STR,LEN)   _PyUnicode_FromASCII((const char*)(STR),(LEN))
 #define STRINGLIB_CHECK          PyUnicode_Check
 #define STRINGLIB_CHECK_EXACT    PyUnicode_CheckExact
 

diff --git a/Objects/stringlib/codecs.h b/Objects/stringlib/codecs.h
index d6f2b98f2b30a33f9e69243f720cc8ce3db9d015..269a5581f70055261bbdf753a85a33fa4a9e85c4 100644 (file)
--- a/Objects/stringlib/codecs.h
+++ b/Objects/stringlib/codecs.h
@@ -46,7 +46,7 @@ STRINGLIB(utf8_decode)(const char **inptr, const char *end,
                     /* Read a whole long at a time (either 4 or 8 bytes),
                        and do a fast unrolled copy if it only contains ASCII
                        characters. */
-                    unsigned long value = *(unsigned long *) _s;
+                    unsigned long value = *(const unsigned long *) _s;
                     if (value & ASCII_CHAR_MASK)
                         break;
 #if PY_LITTLE_ENDIAN
@@ -515,7 +515,7 @@ STRINGLIB(utf16_decode)(const unsigned char **inptr, const unsigned char *e,
             /* Fast path for runs of in-range non-surrogate chars. */
             const unsigned char *_q = q;
             while (_q < aligned_end) {
-                unsigned long block = * (unsigned long *) _q;
+                unsigned long block = * (const unsigned long *) _q;
                 if (native_ordering) {
                     /* Can use buffer directly */
                     if (block & FAST_CHAR_MASK)

diff --git a/Objects/stringlib/find_max_char.h b/Objects/stringlib/find_max_char.h
index 8ccbc3094463df3123873d44f32f7dc2fbf58ee6..f4e0a7761d31198bf426dc8b7cbcb70b940ffc0b 100644 (file)
--- a/Objects/stringlib/find_max_char.h
+++ b/Objects/stringlib/find_max_char.h
@@ -28,7 +28,7 @@ STRINGLIB(find_max_char)(const STRINGLIB_CHAR *begin, const STRINGLIB_CHAR *end)
             /* Help register allocation */
             const unsigned char *_p = p;
             while (_p < aligned_end) {
-                unsigned long value = *(unsigned long *) _p;
+                unsigned long value = *(const unsigned long *) _p;
                 if (value & UCS1_ASCII_CHAR_MASK)
                     return 255;
                 _p += SIZEOF_LONG;

diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
index 0556eff8d07132f703b5f510663b8c9d0f023602..8ba379e8887e98f5f9a49fee1be19f62af2ffca5 100644 (file)
--- a/Objects/unicodeobject.c
+++ b/Objects/unicodeobject.c
@@ -171,8 +171,8 @@ extern "C" {
 #define _PyUnicode_CONVERT_BYTES(from_type, to_type, begin, end, to) \
     do {                                                \
         to_type *_to = (to_type *)(to);                \
-        const from_type *_iter = (from_type *)(begin);  \
-        const from_type *_end = (from_type *)(end);     \
+        const from_type *_iter = (const from_type *)(begin);\
+        const from_type *_end = (const from_type *)(end);\
         Py_ssize_t n = (_end) - (_iter);                \
         const from_type *_unrolled_end =                \
             _iter + _Py_SIZE_ROUND_DOWN(n, 4);          \
@@ -919,21 +919,21 @@ findchar(const void *s, int kind,
         if ((Py_UCS1) ch != ch)
             return -1;
         if (direction > 0)
-            return ucs1lib_find_char((Py_UCS1 *) s, size, (Py_UCS1) ch);
+            return ucs1lib_find_char((const Py_UCS1 *) s, size, (Py_UCS1) ch);
         else
-            return ucs1lib_rfind_char((Py_UCS1 *) s, size, (Py_UCS1) ch);
+            return ucs1lib_rfind_char((const Py_UCS1 *) s, size, (Py_UCS1) ch);
     case PyUnicode_2BYTE_KIND:
         if ((Py_UCS2) ch != ch)
             return -1;
         if (direction > 0)
-            return ucs2lib_find_char((Py_UCS2 *) s, size, (Py_UCS2) ch);
+            return ucs2lib_find_char((const Py_UCS2 *) s, size, (Py_UCS2) ch);
         else
-            return ucs2lib_rfind_char((Py_UCS2 *) s, size, (Py_UCS2) ch);
+            return ucs2lib_rfind_char((const Py_UCS2 *) s, size, (Py_UCS2) ch);
     case PyUnicode_4BYTE_KIND:
         if (direction > 0)
-            return ucs4lib_find_char((Py_UCS4 *) s, size, ch);
+            return ucs4lib_find_char((const Py_UCS4 *) s, size, ch);
         else
-            return ucs4lib_rfind_char((Py_UCS4 *) s, size, ch);
+            return ucs4lib_rfind_char((const Py_UCS4 *) s, size, ch);
     default:
         Py_UNREACHABLE();
     }
@@ -3353,7 +3353,7 @@ PyUnicode_Decode(const char *s,
 
     /* Decode via the codec registry */
     buffer = NULL;
-    if (PyBuffer_FillInfo(&info, NULL, (void *)s, size, 1, PyBUF_FULL_RO) < 0)
+    if (PyBuffer_FillInfo(&info, NULL, (const void *)s, size, 1, PyBUF_FULL_RO) < 0)
         goto onError;
     buffer = PyMemoryView_FromBuffer(&info);
     if (buffer == NULL)
@@ -4861,7 +4861,7 @@ ascii_decode(const char *start, const char *end, Py_UCS1 *dest)
             /* Help allocation */
             const char *_p = p;
             while (_p < aligned_end) {
-                unsigned long value = *(unsigned long *) _p;
+                unsigned long value = *(const unsigned long *) _p;
                 if (value & ASCII_CHAR_MASK)
                     break;
                 _p += SIZEOF_LONG;
@@ -5406,7 +5406,7 @@ PyUnicode_DecodeUTF32Stateful(const char *s,
     PyObject *errorHandler = NULL;
     PyObject *exc = NULL;
 
-    q = (unsigned char *)s;
+    q = (const unsigned char *)s;
     e = q + size;
 
     if (byteorder)
@@ -5731,7 +5731,7 @@ PyUnicode_DecodeUTF16Stateful(const char *s,
     PyObject *exc = NULL;
     const char *encoding;
 
-    q = (unsigned char *)s;
+    q = (const unsigned char *)s;
     e = q + size;
 
     if (byteorder)
@@ -6660,7 +6660,7 @@ PyUnicode_DecodeLatin1(const char *s,
                        const char *errors)
 {
     /* Latin-1 is equivalent to the first 256 ordinals in Unicode. */
-    return _PyUnicode_FromUCS1((unsigned char*)s, size);
+    return _PyUnicode_FromUCS1((const unsigned char*)s, size);
 }
 
 /* create or adjust a UnicodeEncodeError */
@@ -13708,7 +13708,7 @@ _PyUnicodeWriter_WriteASCIIString(_PyUnicodeWriter *writer,
     if (len == -1)
         len = strlen(ascii);
 
-    assert(ucs1lib_find_max_char((Py_UCS1*)ascii, (Py_UCS1*)ascii + len) < 128);
+    assert(ucs1lib_find_max_char((const Py_UCS1*)ascii, (const Py_UCS1*)ascii + len) < 128);
 
     if (writer->buffer == NULL && !writer->overallocate) {
         PyObject *str;
@@ -13767,7 +13767,7 @@ _PyUnicodeWriter_WriteLatin1String(_PyUnicodeWriter *writer,
 {
     Py_UCS4 maxchar;
 
-    maxchar = ucs1lib_find_max_char((Py_UCS1*)str, (Py_UCS1*)str + len);
+    maxchar = ucs1lib_find_max_char((const Py_UCS1*)str, (const Py_UCS1*)str + len);
     if (_PyUnicodeWriter_Prepare(writer, len, maxchar) == -1)
         return -1;
     unicode_write_cstr(writer->buffer, writer->pos, str, len);

diff --git a/Python/marshal.c b/Python/marshal.c
index c6a06e8c239764a8688ac107d3474be61a591fd1..d3fee32380b58b814cdf69bad9211d4c252b1d39 100644 (file)
--- a/Python/marshal.c
+++ b/Python/marshal.c
@@ -734,7 +734,7 @@ r_byte(RFILE *p)
     else {
         const char *ptr = r_string(1, p);
         if (ptr != NULL)
-            c = *(unsigned char *) ptr;
+            c = *(const unsigned char *) ptr;
     }
     return c;
 }

diff --git a/Python/pyhash.c b/Python/pyhash.c
index 4c0b929586fc16e59f6917dfe6c9c1c858299275..ba224ee373631d3722044d81a620c64eef98b221 100644 (file)
--- a/Python/pyhash.c
+++ b/Python/pyhash.c
@@ -366,7 +366,7 @@ static PyHash_FuncDef PyHash_Func = {fnv, "fnv", 8 * SIZEOF_PY_HASH_T,
 static uint64_t
 siphash24(uint64_t k0, uint64_t k1, const void *src, Py_ssize_t src_sz) {
     uint64_t b = (uint64_t)src_sz << 56;
-    const uint8_t *in = (uint8_t*)src;
+    const uint8_t *in = (const uint8_t*)src;
 
     uint64_t v0 = k0 ^ 0x736f6d6570736575ULL;
     uint64_t v1 = k1 ^ 0x646f72616e646f6dULL;