Fixes to bytes_to_utf8-type functions

khwilliamson · khwilliamson · commit 9de1a4943557 · 2025-01-20T18:21:47.000-07:00
This commit turns bytes_to_utf8() back into an (inline) function, and changes the type of a parameter in bytes_to_utf8_free_me() to void*, which is a more accurate type for it. Fixes #22902
diff --git a/embed.fnc b/embed.fnc
@@ -794,12 +794,12 @@ Adp	|int	|bytes_cmp_utf8 |NN const U8 *b 			\
 Adp	|U8 *	|bytes_from_utf8|NN const U8 *s 			\
 				|NN STRLEN *lenp			\
 				|NN bool *is_utf8p
-Admp	|U8 *	|bytes_to_utf8	|NN const U8 *s 			\
+Adip	|U8 *	|bytes_to_utf8	|NN const U8 *s 			\
 				|NN STRLEN *lenp
 Adp	|U8 *	|bytes_to_utf8_free_me					\
 				|NN const U8 *s 			\
 				|NN STRLEN *lenp			\
-				|NULLOK const U8 **free_me
+				|NULLOK void **free_me
 AOdp	|SSize_t|call_argv	|NN const char *sub_name		\
 				|I32 flags				\
 				|NN char **argv
diff --git a/embed.h b/embed.h
@@ -155,7 +155,7 @@
 # define block_start(a)                         Perl_block_start(aTHX_ a)
 # define bytes_cmp_utf8(a,b,c,d)                Perl_bytes_cmp_utf8(aTHX_ a,b,c,d)
 # define bytes_from_utf8(a,b,c)                 Perl_bytes_from_utf8(aTHX_ a,b,c)
-# define bytes_to_utf8(a,b)                     Perl_bytes_to_utf8(aTHX,a,b)
+# define bytes_to_utf8(a,b)                     Perl_bytes_to_utf8(aTHX_ a,b)
 # define bytes_to_utf8_free_me(a,b,c)           Perl_bytes_to_utf8_free_me(aTHX_ a,b,c)
 # define c9strict_utf8_to_uv                    Perl_c9strict_utf8_to_uv
 # define call_argv(a,b,c)                       Perl_call_argv(aTHX_ a,b,c)
diff --git a/inline.h b/inline.h
@@ -1132,7 +1132,6 @@ in lvalue context.
 =cut
 */
 
-
 PERL_STATIC_INLINE bool
 Perl_rpp_is_lone(pTHX_ SV *sv)
 {
@@ -1231,6 +1230,12 @@ Perl_append_utf8_from_native_byte(const U8 byte, U8** dest)
     }
 }
 
+PERL_STATIC_INLINE U8 *
+Perl_bytes_to_utf8(pTHX_ const U8 *s, STRLEN *lenp)
+{
+    return bytes_to_utf8_free_me(s, lenp, NULL);
+}
+
 /*
 =for apidoc valid_utf8_to_uvchr
 Like C<L<perlapi/utf8_to_uvchr_buf>>, but should only be called when it is
diff --git a/proto.h b/proto.h
diff --git a/utf8.c b/utf8.c
@@ -3269,7 +3269,7 @@ But when it is a non-NULL pointer, C<bytes_to_utf8_free_me> stores into it
 either NULL if no memory was allocated; or a pointer to that new memory.  This
 allows the following convenient paradigm:
 
- U8 * free_me;
+ void * free_me;
  U8 converted = bytes_to_utf8_free_me(string, &len, &free_me);
 
  ...
@@ -3292,7 +3292,7 @@ EBCDIC), see L</sv_recode_to_utf8>().
 
 U8*
 Perl_bytes_to_utf8_free_me(pTHX_ const U8 *s, Size_t *lenp,
-                                 const U8 ** free_me_ptr)
+                                 void ** free_me_ptr)
 {
     PERL_ARGS_ASSERT_BYTES_TO_UTF8_FREE_ME;
     PERL_UNUSED_CONTEXT;
diff --git a/utf8.h b/utf8.h
@@ -1330,7 +1330,6 @@ point's representation.
 
 #define Perl_is_utf8_char_buf(buf, buf_end) isUTF8_CHAR(buf, buf_end)
 
-#define Perl_bytes_to_utf8(mTHX, s, lenp)  Perl_bytes_to_utf8_free_me(aTHX_ s, lenp, NULL)
 typedef enum {
     PL_utf8_to_bytes_overwrite = 0,
     PL_utf8_to_bytes_new_memory,

Original file line number	Diff line number	Diff line change
`@@ -1132,7 +1132,6 @@ in lvalue context.`
`1132`	`1132`	`=cut`
`1133`	`1133`	`*/`
`1134`	`1134`
`1135`		`-`
`1136`	`1135`	`PERL_STATIC_INLINE bool`
`1137`	`1136`	`Perl_rpp_is_lone(pTHX_ SV *sv)`
`1138`	`1137`	`{`
`@@ -1231,6 +1230,12 @@ Perl_append_utf8_from_native_byte(const U8 byte, U8** dest)`
`1231`	`1230`	`}`
`1232`	`1231`	`}`
`1233`	`1232`
	`1233`	`+PERL_STATIC_INLINE U8 *`
	`1234`	`+Perl_bytes_to_utf8(pTHX_ const U8 s, STRLEN lenp)`
	`1235`	`+{`
	`1236`	`+ return bytes_to_utf8_free_me(s, lenp, NULL);`
	`1237`	`+}`
	`1238`	`+`
`1234`	`1239`	`/*`
`1235`	`1240`	`=for apidoc valid_utf8_to_uvchr`
`1236`	`1241`	`Like C<L<perlapi/utf8_to_uvchr_buf>>, but should only be called when it is`