bpo-39943: Add the const qualifier to pointers on non-mutable PyUnicode data. (GH-19345)

16 files changed, 57 insertions, 57 deletions

diff --git a/Modules/_csv.c b/Modules/_csv.c
index 9c49715..950b0d7 100644
--- a/Modules/_csv.c
+++ b/Modules/_csv.c
@@ -789,7 +789,7 @@ Reader_iternext(ReaderObj *self)
     Py_UCS4 c;
     Py_ssize_t pos, linelen;
     unsigned int kind;
-    void *data;
+    const void *data;
     PyObject *lineobj;
 
     if (parse_reset(self) < 0)
@@ -996,7 +996,7 @@ join_reset(WriterObj *self)
  * record length.
  */
 static Py_ssize_t
-join_append_data(WriterObj *self, unsigned int field_kind, void *field_data,
+join_append_data(WriterObj *self, unsigned int field_kind, const void *field_data,
                  Py_ssize_t field_len, int *quoted,
                  int copy_phase)
 {
@@ -1107,7 +1107,7 @@ static int
 join_append(WriterObj *self, PyObject *field, int quoted)
 {
     unsigned int field_kind = -1;
-    void *field_data = NULL;
+    const void *field_data = NULL;
     Py_ssize_t field_len = 0;
     Py_ssize_t rec_len;
 
@@ -1139,7 +1139,7 @@ join_append_lineterminator(WriterObj *self)
 {
     Py_ssize_t terminator_len, i;
     unsigned int term_kind;
-    void *term_data;
+    const void *term_data;
 
     terminator_len = PyUnicode_GET_LENGTH(self->dialect->lineterminator);
     if (terminator_len == -1)
diff --git a/Modules/_decimal/_decimal.c b/Modules/_decimal/_decimal.c
index b36e309..cdc942f 100644
--- a/Modules/_decimal/_decimal.c
+++ b/Modules/_decimal/_decimal.c
@@ -1878,7 +1878,7 @@ dec_dealloc(PyObject *dec)
 /******************************************************************************/
 
 Py_LOCAL_INLINE(int)
-is_space(enum PyUnicode_Kind kind, void *data, Py_ssize_t pos)
+is_space(enum PyUnicode_Kind kind, const void *data, Py_ssize_t pos)
 {
     Py_UCS4 ch = PyUnicode_READ(kind, data, pos);
     return Py_UNICODE_ISSPACE(ch);
@@ -1896,7 +1896,7 @@ static char *
 numeric_as_ascii(const PyObject *u, int strip_ws, int ignore_underscores)
 {
     enum PyUnicode_Kind kind;
-    void *data;
+    const void *data;
     Py_UCS4 ch;
     char *res, *cp;
     Py_ssize_t j, len;
diff --git a/Modules/_elementtree.c b/Modules/_elementtree.c
index 49c372d..c0c741e 100644
--- a/Modules/_elementtree.c
+++ b/Modules/_elementtree.c
@@ -1132,7 +1132,7 @@ checkpath(PyObject* tag)
 
     if (PyUnicode_Check(tag)) {
         const Py_ssize_t len = PyUnicode_GET_LENGTH(tag);
-        void *data = PyUnicode_DATA(tag);
+        const void *data = PyUnicode_DATA(tag);
         unsigned int kind = PyUnicode_KIND(tag);
         if (len >= 3 && PyUnicode_READ(kind, data, 0) == '{' && (
                 PyUnicode_READ(kind, data, 1) == '}' || (
diff --git a/Modules/_io/textio.c b/Modules/_io/textio.c
index dedbefe..12dba38 100644
--- a/Modules/_io/textio.c
+++ b/Modules/_io/textio.c
@@ -340,7 +340,7 @@ _PyIncrementalNewlineDecoder_decode(PyObject *myself,
             goto error;
         kind = PyUnicode_KIND(modified);
         out = PyUnicode_DATA(modified);
-        PyUnicode_WRITE(kind, PyUnicode_DATA(modified), 0, '\r');
+        PyUnicode_WRITE(kind, out, 0, '\r');
         memcpy(out + kind, PyUnicode_DATA(output), kind * output_len);
         Py_DECREF(output);
         output = modified; /* output remains ready */
@@ -367,7 +367,7 @@ _PyIncrementalNewlineDecoder_decode(PyObject *myself,
     /* Record which newlines are read and do newline translation if desired,
        all in one pass. */
     {
-        void *in_str;
+        const void *in_str;
         Py_ssize_t len;
         int seennl = self->seennl;
         int only_lf = 0;
@@ -447,7 +447,7 @@ _PyIncrementalNewlineDecoder_decode(PyObject *myself,
         else {
             void *translated;
             int kind = PyUnicode_KIND(output);
-            void *in_str = PyUnicode_DATA(output);
+            const void *in_str = PyUnicode_DATA(output);
             Py_ssize_t in, out;
             /* XXX: Previous in-place translation here is disabled as
                resizing is not possible anymore */
@@ -2085,7 +2085,7 @@ _PyIO_find_line_ending(
     else {
         /* Non-universal mode. */
         Py_ssize_t readnl_len = PyUnicode_GET_LENGTH(readnl);
-        Py_UCS1 *nl = PyUnicode_1BYTE_DATA(readnl);
+        const Py_UCS1 *nl = PyUnicode_1BYTE_DATA(readnl);
         /* Assume that readnl is an ASCII character. */
         assert(PyUnicode_KIND(readnl) == PyUnicode_1BYTE_KIND);
         if (readnl_len == 1) {
@@ -2139,7 +2139,7 @@ _textiowrapper_readline(textio *self, Py_ssize_t limit)
     chunked = 0;
 
     while (1) {
-        char *ptr;
+        const char *ptr;
         Py_ssize_t line_len;
         int kind;
         Py_ssize_t consumed = 0;
diff --git a/Modules/_json.c b/Modules/_json.c
index 3ab1cb3..1754416 100644
--- a/Modules/_json.c
+++ b/Modules/_json.c
@@ -159,8 +159,8 @@ ascii_escape_unicode(PyObject *pystr)
     Py_ssize_t output_size;
     Py_ssize_t chars;
     PyObject *rval;
-    void *input;
-    unsigned char *output;
+    const void *input;
+    Py_UCS1 *output;
     int kind;
 
     if (PyUnicode_READY(pystr) == -1)
@@ -225,7 +225,7 @@ escape_unicode(PyObject *pystr)
     Py_ssize_t output_size;
     Py_ssize_t chars;
     PyObject *rval;
-    void *input;
+    const void *input;
     int kind;
     Py_UCS4 maxchar;
 
@@ -678,7 +678,7 @@ _parse_object_unicode(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ss
 
     Returns a new PyObject (usually a dict, but object_hook can change that)
     */
-    void *str;
+    const void *str;
     int kind;
     Py_ssize_t end_idx;
     PyObject *val = NULL;
@@ -808,7 +808,7 @@ _parse_array_unicode(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssi
 
     Returns a new PyList
     */
-    void *str;
+    const void *str;
     int kind;
     Py_ssize_t end_idx;
     PyObject *val = NULL;
@@ -911,7 +911,7 @@ _match_number_unicode(PyScannerObject *s, PyObject *pystr, Py_ssize_t start, Py_
         PyLong, or PyFloat.
         May return other types if parse_int or parse_float are set
     */
-    void *str;
+    const void *str;
     int kind;
     Py_ssize_t end_idx;
     Py_ssize_t idx = start;
@@ -1028,7 +1028,7 @@ scan_once_unicode(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssize_
     Returns a new PyObject representation of the term.
     */
     PyObject *res;
-    void *str;
+    const void *str;
     int kind;
     Py_ssize_t length;
 
diff --git a/Modules/_operator.c b/Modules/_operator.c
index 007c21b..19026b6 100644
--- a/Modules/_operator.c
+++ b/Modules/_operator.c
@@ -1170,7 +1170,7 @@ attrgetter_new(PyTypeObject *type, PyObject *args, PyObject *kwds)
     for (idx = 0; idx < nattrs; ++idx) {
         PyObject *item = PyTuple_GET_ITEM(args, idx);
         Py_ssize_t item_len;
-        void *data;
+        const void *data;
         unsigned int kind;
         int dot_count;
 
diff --git a/Modules/_pickle.c b/Modules/_pickle.c
index c3385ad..4b46c1f 100644
--- a/Modules/_pickle.c
+++ b/Modules/_pickle.c
@@ -2581,7 +2581,7 @@ raw_unicode_escape(PyObject *obj)
 {
     char *p;
     Py_ssize_t i, size;
-    void *data;
+    const void *data;
     unsigned int kind;
     _PyBytesWriter writer;
 
diff --git a/Modules/_sqlite/connection.c b/Modules/_sqlite/connection.c
index 697295d..92bdfe3 100644
--- a/Modules/_sqlite/connection.c
+++ b/Modules/_sqlite/connection.c
@@ -1644,7 +1644,7 @@ pysqlite_connection_create_collation(pysqlite_Connection* self, PyObject* args)
     const char *uppercase_name_str;
     int rc;
     unsigned int kind;
-    void *data;
+    const void *data;
 
     if (!pysqlite_check_thread(self) || !pysqlite_check_connection(self)) {
         goto finally;
diff --git a/Modules/_sre.c b/Modules/_sre.c
index bee2e12..836d796 100644
--- a/Modules/_sre.c
+++ b/Modules/_sre.c
@@ -351,7 +351,7 @@ state_reset(SRE_STATE* state)
     data_stack_dealloc(state);
 }
 
-static void*
+static const void*
 getstring(PyObject* string, Py_ssize_t* p_length,
           int* p_isbytes, int* p_charsize,
           Py_buffer *view)
@@ -398,11 +398,11 @@ state_init(SRE_STATE* state, PatternObject* pattern, PyObject* string,
 
     Py_ssize_t length;
     int isbytes, charsize;
-    void* ptr;
+    const void* ptr;
 
     memset(state, 0, sizeof(SRE_STATE));
 
-    state->mark = PyMem_New(void *, pattern->groups * 2);
+    state->mark = PyMem_New(const void *, pattern->groups * 2);
     if (!state->mark) {
         PyErr_NoMemory();
         goto err;
@@ -891,7 +891,7 @@ _sre_SRE_Pattern_split_impl(PatternObject *self, PyObject *string,
     Py_ssize_t status;
     Py_ssize_t n;
     Py_ssize_t i;
-    void* last;
+    const void* last;
 
     assert(self->codesize != 0);
 
@@ -984,7 +984,7 @@ pattern_subx(PatternObject* self, PyObject* ptemplate, PyObject* string,
     PyObject* item;
     PyObject* filter;
     PyObject* match;
-    void* ptr;
+    const void* ptr;
     Py_ssize_t status;
     Py_ssize_t n;
     Py_ssize_t i, b, e;
@@ -1895,7 +1895,7 @@ match_getslice_by_index(MatchObject* self, Py_ssize_t index, PyObject* def)
     int isbytes, charsize;
     Py_buffer view;
     PyObject *result;
-    void* ptr;
+    const void* ptr;
     Py_ssize_t i, j;
 
     assert(0 <= index && index < self->groups);
diff --git a/Modules/cjkcodecs/cjkcodecs.h b/Modules/cjkcodecs/cjkcodecs.h
index 8f6f880..e41755b 100644
--- a/Modules/cjkcodecs/cjkcodecs.h
+++ b/Modules/cjkcodecs/cjkcodecs.h
@@ -72,7 +72,7 @@ static const struct dbcs_map *mapping_list;
 #define ENCODER(encoding)                                               \
     static Py_ssize_t encoding##_encode(                                \
         MultibyteCodec_State *state, const void *config,                \
-        int kind, void *data,                          \
+        int kind, const void *data,                                     \
         Py_ssize_t *inpos, Py_ssize_t inlen,                            \
         unsigned char **outbuf, Py_ssize_t outleft, int flags)
 #define ENCODER_RESET(encoding)                                         \
diff --git a/Modules/cjkcodecs/multibytecodec.c b/Modules/cjkcodecs/multibytecodec.c
index a09c75d..9f9fbeb 100644
--- a/Modules/cjkcodecs/multibytecodec.c
+++ b/Modules/cjkcodecs/multibytecodec.c
@@ -228,7 +228,7 @@ multibytecodec_encerror(MultibyteCodec *codec,
         Py_ssize_t r;
         Py_ssize_t inpos;
         int kind;
-        void *data;
+        const void *data;
 
         replchar = PyUnicode_FromOrdinal('?');
         if (replchar == NULL)
@@ -457,7 +457,7 @@ multibytecodec_encode(MultibyteCodec *codec,
     Py_ssize_t finalsize, r = 0;
     Py_ssize_t datalen;
     int kind;
-    void *data;
+    const void *data;
 
     if (PyUnicode_READY(text) < 0)
         return NULL;
diff --git a/Modules/cjkcodecs/multibytecodec.h b/Modules/cjkcodecs/multibytecodec.h
index 4d2b355..5946821 100644
--- a/Modules/cjkcodecs/multibytecodec.h
+++ b/Modules/cjkcodecs/multibytecodec.h
@@ -30,7 +30,7 @@ typedef struct {
 typedef int (*mbcodec_init)(const void *config);
 typedef Py_ssize_t (*mbencode_func)(MultibyteCodec_State *state,
                         const void *config,
-                        int kind, void *data,
+                        int kind, const void *data,
                         Py_ssize_t *inpos, Py_ssize_t inlen,
                         unsigned char **outbuf, Py_ssize_t outleft,
                         int flags);
diff --git a/Modules/pyexpat.c b/Modules/pyexpat.c
index a7f8b50..d930e3e 100644
--- a/Modules/pyexpat.c
+++ b/Modules/pyexpat.c
@@ -1060,7 +1060,7 @@ PyUnknownEncodingHandler(void *encodingHandlerData,
     static unsigned char template_buffer[256] = {0};
     PyObject* u;
     int i;
-    void *data;
+    const void *data;
     unsigned int kind;
 
     if (PyErr_Occurred())
diff --git a/Modules/sre.h b/Modules/sre.h
index a728488..9b0d8b1 100644
--- a/Modules/sre.h
+++ b/Modules/sre.h
@@ -54,17 +54,17 @@ typedef struct {
 
 typedef struct SRE_REPEAT_T {
     Py_ssize_t count;
-    SRE_CODE* pattern; /* points to REPEAT operator arguments */
-    void* last_ptr; /* helper to check for infinite loops */
+    const SRE_CODE* pattern; /* points to REPEAT operator arguments */
+    const void* last_ptr; /* helper to check for infinite loops */
     struct SRE_REPEAT_T *prev; /* points to previous repeat context */
 } SRE_REPEAT;
 
 typedef struct {
     /* string pointers */
-    void* ptr; /* current position (also end of current slice) */
-    void* beginning; /* start of original string */
-    void* start; /* start of current slice */
-    void* end; /* end of original string */
+    const void* ptr; /* current position (also end of current slice) */
+    const void* beginning; /* start of original string */
+    const void* start; /* start of current slice */
+    const void* end; /* end of original string */
     /* attributes for the match object */
     PyObject* string;
     Py_buffer buffer;
@@ -74,7 +74,7 @@ typedef struct {
     /* registers */
     Py_ssize_t lastindex;
     Py_ssize_t lastmark;
-    void** mark;
+    const void** mark;
     int match_all;
     int must_advance;
     /* dynamically allocated stuff */
diff --git a/Modules/sre_lib.h b/Modules/sre_lib.h
index 437ab43..9cc7863 100644
--- a/Modules/sre_lib.h
+++ b/Modules/sre_lib.h
@@ -13,7 +13,7 @@
 /* This file is included three times, with different character settings */
 
 LOCAL(int)
-SRE(at)(SRE_STATE* state, SRE_CHAR* ptr, SRE_CODE at)
+SRE(at)(SRE_STATE* state, const SRE_CHAR* ptr, SRE_CODE at)
 {
     /* check if pointer is at given position */
 
@@ -101,7 +101,7 @@ SRE(at)(SRE_STATE* state, SRE_CHAR* ptr, SRE_CODE at)
 }
 
 LOCAL(int)
-SRE(charset)(SRE_STATE* state, SRE_CODE* set, SRE_CODE ch)
+SRE(charset)(SRE_STATE* state, const SRE_CODE* set, SRE_CODE ch)
 {
     /* check if character is a member of the given set */
 
@@ -188,7 +188,7 @@ SRE(charset)(SRE_STATE* state, SRE_CODE* set, SRE_CODE ch)
 }
 
 LOCAL(int)
-SRE(charset_loc_ignore)(SRE_STATE* state, SRE_CODE* set, SRE_CODE ch)
+SRE(charset_loc_ignore)(SRE_STATE* state, const SRE_CODE* set, SRE_CODE ch)
 {
     SRE_CODE lo, up;
     lo = sre_lower_locale(ch);
@@ -199,15 +199,15 @@ SRE(charset_loc_ignore)(SRE_STATE* state, SRE_CODE* set, SRE_CODE ch)
     return up != lo && SRE(charset)(state, set, up);
 }
 
-LOCAL(Py_ssize_t) SRE(match)(SRE_STATE* state, SRE_CODE* pattern, int toplevel);
+LOCAL(Py_ssize_t) SRE(match)(SRE_STATE* state, const SRE_CODE* pattern, int toplevel);
 
 LOCAL(Py_ssize_t)
-SRE(count)(SRE_STATE* state, SRE_CODE* pattern, Py_ssize_t maxcount)
+SRE(count)(SRE_STATE* state, const SRE_CODE* pattern, Py_ssize_t maxcount)
 {
     SRE_CODE chr;
     SRE_CHAR c;
-    SRE_CHAR* ptr = (SRE_CHAR *)state->ptr;
-    SRE_CHAR* end = (SRE_CHAR *)state->end;
+    const SRE_CHAR* ptr = (const SRE_CHAR *)state->ptr;
+    const SRE_CHAR* end = (const SRE_CHAR *)state->end;
     Py_ssize_t i;
 
     /* adjust end */
@@ -335,14 +335,14 @@ SRE(count)(SRE_STATE* state, SRE_CODE* pattern, Py_ssize_t maxcount)
 
 #if 0 /* not used in this release */
 LOCAL(int)
-SRE(info)(SRE_STATE* state, SRE_CODE* pattern)
+SRE(info)(SRE_STATE* state, const SRE_CODE* pattern)
 {
     /* check if an SRE_OP_INFO block matches at the current position.
        returns the number of SRE_CODE objects to skip if successful, 0
        if no match */
 
-    SRE_CHAR* end = (SRE_CHAR*) state->end;
-    SRE_CHAR* ptr = (SRE_CHAR*) state->ptr;
+    const SRE_CHAR* end = (const SRE_CHAR*) state->end;
+    const SRE_CHAR* ptr = (const SRE_CHAR*) state->ptr;
     Py_ssize_t i;
 
     /* check minimal length */
@@ -531,8 +531,8 @@ do { \
 typedef struct {
     Py_ssize_t last_ctx_pos;
     Py_ssize_t jump;
-    SRE_CHAR* ptr;
-    SRE_CODE* pattern;
+    const SRE_CHAR* ptr;
+    const SRE_CODE* pattern;
     Py_ssize_t count;
     Py_ssize_t lastmark;
     Py_ssize_t lastindex;
@@ -546,9 +546,9 @@ typedef struct {
 /* check if string matches the given pattern.  returns <0 for
    error, 0 for failure, and 1 for success */
 LOCAL(Py_ssize_t)
-SRE(match)(SRE_STATE* state, SRE_CODE* pattern, int toplevel)
+SRE(match)(SRE_STATE* state, const SRE_CODE* pattern, int toplevel)
 {
-    SRE_CHAR* end = (SRE_CHAR *)state->end;
+    const SRE_CHAR* end = (const SRE_CHAR *)state->end;
     Py_ssize_t alloc_pos, ctx_pos = -1;
     Py_ssize_t i, ret = 0;
     Py_ssize_t jump;
diff --git a/Modules/unicodedata.c b/Modules/unicodedata.c
index 1a9e1c0..569e785 100644
--- a/Modules/unicodedata.c
+++ b/Modules/unicodedata.c
@@ -496,7 +496,7 @@ nfd_nfkd(PyObject *self, PyObject *input, int k)
     Py_UCS4 *output;
     Py_ssize_t i, o, osize;
     int kind;
-    void *data;
+    const void *data;
     /* Longest decomposition in Unicode 3.2: U+FDFA */
     Py_UCS4 stack[20];
     Py_ssize_t space, isize;
@@ -643,7 +643,7 @@ nfc_nfkc(PyObject *self, PyObject *input, int k)
 {
     PyObject *result;
     int kind;
-    void *data;
+    const void *data;
     Py_UCS4 *output;
     Py_ssize_t i, i1, o, len;
     int f,l,index,index1,comb;
@@ -804,7 +804,7 @@ is_normalized_quickcheck(PyObject *self, PyObject *input,
 
     Py_ssize_t i, len;
     int kind;
-    void *data;
+    const void *data;
     unsigned char prev_combining = 0;
 
     /* The two quickcheck bits at this shift have type QuickcheckResult. */