Path: eternal-september.org!news.eternal-september.org!feeder3.eternal-september.org!i2pn.org!i2pn2.org!.POSTED!not-for-mail
From: fir <fir@grunge.pl>
Newsgroups: comp.lang.c
Subject: Re: if(!(i%16))
Date: Sun, 31 Mar 2024 16:49:49 +0200
Organization: i2pn2 (i2pn.org)
Message-ID: <uubt68$3krq7$1@i2pn2.org>
References: <uua137$3in1a$1@i2pn2.org> <uua45d$18naj$1@dont-email.me> <uua4uh$3irjh$1@i2pn2.org> <uua7hj$19ih8$1@dont-email.me> <uubrel$3kpld$1@i2pn2.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Injection-Date: Sun, 31 Mar 2024 14:49:46 -0000 (UTC)
Injection-Info: i2pn2.org;
	logging-data="3829575"; mail-complaints-to="usenet@i2pn2.org";
	posting-account="+ydHcGjgSeBt3Wz3WTfKefUptpAWaXduqfw5xdfsuS0";
User-Agent: Mozilla/5.0 (Windows NT 5.1; rv:27.0) Gecko/20100101 Firefox/27.0 SeaMonkey/2.24
X-Spam-Checker-Version: SpamAssassin 4.0.0
In-Reply-To: <uubrel$3kpld$1@i2pn2.org>

fir wrote:
> bart wrote:
>> On 30/03/2024 22:49, fir wrote:
>>> bart wrote:
>>>> On 30/03/2024 21:44, fir wrote:
>>>>>   void bytes_dump_in_hex()
>>>>>    {
>>>>>      for(int i=0; i<bytes_size; i++)
>>>>>      {
>>>>>        if(!(i%16)) printf("\n");
>>>>>
>>>>>        printf("%02x ", bytes[i]);
>>>>>      }
>>>>>    }
>>>>>
>>>>> in the code above seem that those inner () in if(!(i%16)) are needed
>>>>>
>>>>> why is that so?
>>>>
>>>> Without them !i%16 will be parsed as (!i)%16.
>>>>
>>> and does it have sense?
>>
>> It's a unary operator. Usually they are applied before binary ones. So
>> '++ p + n' means '(++p)+n' not '++(p+n)' which wouldn't work anyway.
>>
>> And -2-3 means (-2)-3 or -5, not -(2-3) which would be +1.
>>
>> However whether it makes sense is beside the point. It's how C has
>> always worked.
>>
>>> if ! is boolean operator and % is arithmetic then converting things to
>>> bolean and then do arithmetoc on it seems  not much reasonable...
>>
>> Some people think that ! should work like that, so that here:
>>
>>     not a<b and b<c
>>
>> the 'not' applies to the entire expression: not(a<b and b<c), rather
>> than (not a)<b and b<c.
>>
>> But honestly, there are lots of things that are worse about C's set of
>> precedences.
>>
> i remember i answered then go to sleep now i dont see my answer
>
> never thinked on this operator precedences but as i said for sure
> thise "relative" (< > ==) and logical (and or not) should be last
> after bitwise  (and or xor not) and arithmetic (+-*/%)
> and this is as i say no matter of they do like that but the fact
> thet the output of relatives and logical is boleean and boolean is not
> to much use (if any) for arithmetic of bitwise - so this is for sure
>
> relative should be before logical becouse for example
>
> a<10 & a>-10  //say % is logical ias for me there is damn error in c and
> & should be logical and && eventually could be bitwise
>
> is common usage  (so fron this 4 types mentioned logical is 4.
> and relative is 3. )
>
> as to arythmetic vs bitwice
>
> 7*9 &  78+90
>
> im not sure
>
> bitwise generally not fit here to much at all (to all 3 arithmetic,
> logical, and relative..so possiby as this is a bit outside here
> it ebventually could go first to beoutside it
>
> like
>
> a&0xff+b&0xff00+c&0xff0000
>
> common usage to work
>
> so
>
> 1. bitwise
> 2. arithmetic
> 3. relative
> 4. logical
>
>
>
there are yet assigment ones

a=b+c
for sure it should go after arithmetic and bitwise
it also should before relative becouse the same reason - assigment of 
boolean is small use

so

  1. bitwise
  2. arithmetic
  3. assigment
  4. relative
  5. logical

what else those lik . for member or * & for pointers should be first 
(0.) and ?:  should be lowest i guess



and how it look like in c:


arithmetic seem be higher then  relational then bitwise then logical and 
assigment but with some exceptions

2.(ar)4(rel).1(bit).5(logg).3.(assign)

2.4.5. is in order so one should ship bitwise and be aware of assigment


quite funny if i treat bitwisa & | as logical it still work
becouse at least

arithm < rel < log

assigment also shoudl work partially (as to arithmetic and new logical)

i would need to check it in practice what work as here above its written 
  much approximately