Re: Quick check on signed promotion of bytes
- From: Eric Sosman <Eric.Sosman@xxxxxxx>
- Date: Thu, 31 Jan 2008 11:22:56 -0500
toe@xxxxxxxxxxx wrote:
I have a byte called "x". I want byte "y" to be the complement of
"x" (i.e. all the bits flipped).
Initially I wrote:
char unsigned x, y;
...
x = 72;
y = ~x;
But then I thought that the following might happen on your average
system (CHAR_BIT == 8, sizeof(int) == 4):
1) x is promoted to signed int.
2) The complement is take of this signed int.
3) This signed int is then converted to an unsigned char
That's right, under your assumptions. On "exotic" machines
where UCHAR_MAX > INT_MAX (for example, on hardware where all
of char, short, and int are 16 bits wide), then x promotes to
an unsigned int instead of to an int in step 1.
Also see vippstar's response: It's wrong.
Am I right in thinking that this is what will happen? If so, would I
be wise to do the following instead:
y = ~(unsigned)x;
Yes, this will work on both "exotic" and "humdrum" machines.
--
Eric.Sosman@xxxxxxx
.
- Follow-Ups:
- Re: Quick check on signed promotion of bytes
- From: vippstar
- Re: Quick check on signed promotion of bytes
- From: Tomás Ó hÉilidhe
- Re: Quick check on signed promotion of bytes
- References:
- Quick check on signed promotion of bytes
- From: toe
- Quick check on signed promotion of bytes
- Prev by Date: translating oo features to C
- Next by Date: Re: translating oo features to C
- Previous by thread: Re: Quick check on signed promotion of bytes
- Next by thread: Re: Quick check on signed promotion of bytes
- Index(es):
Relevant Pages
|