[expr.unary.op] Cleanup of {ones',two's} complement, and of "operator" #5998
Conversation
|
@opensdh could you please have a look? |
source/expressions.tex
Outdated
| @@ -4662,6 +4662,9 @@ | |||
| of an unsigned quantity is computed by subtracting its value from $2^n$, | |||
| where $n$ is the number of bits in the promoted operand. The type of the | |||
| result is the type of the promoted operand. | |||
| \begin{note} | |||
| The result is the two's complement function of the operand. | |||
There was a problem hiding this comment.
This should differ in only the operative word: the "considered as unsigned" (or however we choose to phrase that) still applies when we're thinking of this as a complement operation (rather than merely commenting that negative numbers are represented by the two's complement in this system), and the result is not itself a function (although I see that [expr.bit.and] and friends say "of" when they really mean "applied to").
There was a problem hiding this comment.
"The sine of 0 is 0" vs "the sine applied to 0 is 0"... both work, don't they? Yes, "of" has two meanings (function declaration vs evaluation), but it's not wrong to use for evaluation.
I don't quite understand the first part; could you suggest a complete note?
There was a problem hiding this comment.
That said, I'm happy to lose the notes entirely now that we got rid of "unary"!
There was a problem hiding this comment.
Yes, but no one would say "the error function of 0.5" but rather "the error function{,'s value, evaluated} at 0.5" because "function of" is confusing. (It is funny, though, that even English has declaration-follows-use here.)
I just meant
The result is the two's complement of the operand (both considered as unsigned).
There was a problem hiding this comment.
Ah yes, that's what you suggested above. PTAL.
There was a problem hiding this comment.
It looks good to me; you might want to check that no one has qualms about "considering" things as unsigned in a note (even though the actual bit behavior should be pretty obvious to all).
There was a problem hiding this comment.
@jensmaurer, @burblebee, @jfbastien: any qualms about "considering" things as unsigned?
There was a problem hiding this comment.
We still haven't resolved whether there are qualms!
There was a problem hiding this comment.
I'm fine with the phrasing. It's a note.
|
We can keep the notes if you find them useful! |
tkoeppe
changed the title
We are currently using the term "unary complement operator" to refer to the "~" operator without ever giving it that name. This change removes the use of that undefined term.
The terms "one's complement" and "two's complement" are well-known, and it is useful to associate them with these operators.
Very much so. |
We are currently using the term "unary complement" to refer to the "~" operator without ever giving it that name.
This change removes unnecessary uses of the terms "unary" and "complement", but adds notes that make referenes to "ones'/two's complement".