On Sat, Nov 16, 2024 at 10:21:47PM +0100, Patrice Dumas wrote:
> Hello,
>
> On cygwin, there are warnings for the isalpha and similar functions from
> ctype.h, like:
>
> ../../info/indices.c: In function 'index_entry_matches':
> ../../info/indices.c:297:30: warning: array subscript has type 'char'
> [-Wchar-subscripts]
> 297 | else if (!isdigit (*p))
> | ^~
> ../../info/indices.c: In function 'report_index_match':
> ../../info/indices.c:433:29: warning: array subscript has type 'char'
> [-Wchar-subscripts]
> 433 | upper = isupper (match[start]) ? 1 : 0;
> | ~~~~~^~~~~~~
>
> Seems to happen when the implementation of isupper/isalpha... is an
> array, and somewhat because is* functions argument is supposed to be
> unsigned char. In the glibc manual it is indeed recommended to cast
> char to (unsigned char) when calling the functions.
>
> Is it important? Should we cast?
I don't remember for certain but we already cast to unsigned char
in the call to isspace in xspara.c.
2022-04-14 Gavin Smith <gavinsmith0...@gmail.com>
unsigned char for isspace
* tp/Texinfo/XS/parsetexi/end_line.c (end_line_misc_line),
* tp/Texinfo/XS/xspara.c (xspara_add_text):
Cast argument to isspace to unsigned char. This avoids varying
test results on Solaris 10 and 11.
The Linux man pages state ("man isalpha"):
The standards require that the argument c for these functions is either
EOF or a value that is representable in the type unsigned char. If the
argument c is of type char, it must be cast to unsigned char, as in the
following example:
char c;
...
res = toupper((unsigned char) c);
This is necessary because char may be the equivalent of signed char, in
which case a byte where the top bit is set would be sign extended when
converting to int, yielding a value that is outside the range of un‐
signed char.
It may not matter much in practice but I expect that it would be ok to
put the casts in.
