MinGW - Minimalist GNU for Windows

On 2010-06-29 05:48Z, Vincent Torri wrote:
> 
> http://msdn.microsoft.com/en-us/library/xdb9w69d%28VS.71%29.aspx

That page says "The h, l, and L prefixes are Microsoft extensions when
used with data of type char."

> to get an unsigned char, i should use "%hc" as modifier (i think that 
> "Single-byte" means unsigned char, as "%c" is for char (being signed or 
> unsigned))

Then you're using a nonstandard extension.

> then gcc reports that format "%hc" expects type 'short int *' but argument 
> 3 has type 'unsigned char *'

And gcc warns that the code is nonstandard.

> Am I doing something wrong or is there a bug in header files ?

If you don't want gcc to give this warning, then inhibit it.

On Tue, 29 Jun 2010, Greg Chicares wrote:

> On 2010-06-29 05:48Z, Vincent Torri wrote:
>>
>> http://msdn.microsoft.com/en-us/library/xdb9w69d%28VS.71%29.aspx
>
> That page says "The h, l, and L prefixes are Microsoft extensions when
> used with data of type char."
>
>> to get an unsigned char, i should use "%hc" as modifier (i think that
>> "Single-byte" means unsigned char, as "%c" is for char (being signed or
>> unsigned))
>
> Then you're using a nonstandard extension.
>
>> then gcc reports that format "%hc" expects type 'short int *' but argument
>> 3 has type 'unsigned char *'
>
> And gcc warns that the code is nonstandard.
>
>> Am I doing something wrong or is there a bug in header files ?
>
> If you don't want gcc to give this warning, then inhibit it.

What I want is getting an unsigned char. "%hhu" is what should be done 
on linux. It is not supported by Windows. As I'm porting a linux library 
to Windows, I'm wondering what to use...

Your comments are not quite precise, you even don't answer to my question.

regards

Vincent Torri

Vincent Torri wrote:
>
> Your comments are not quite precise, you even don't answer to my question.
>

Greg's comments were precise to your post.  Your question wasn't what do 
I do on Windows using GCC to get the same result as this on Linux.  Your 
question was "What did I do wrong" and Greg answered that question by 
telling you to disable the warning if you don't want them.

Since I don't have the answer to your real question or the time at the 
moment to research it, I'll let someone else comment on that.

-- 
Earnie
-- http://www.for-my-kids.com

On Tue, 29 Jun 2010, Earnie wrote:

> Vincent Torri wrote:
>>
>> Your comments are not quite precise, you even don't answer to my question.
>>
>
> Greg's comments were precise to your post.

no. Indeed:

"That page says "The h, l, and L prefixes are Microsoft extensions when
used with data of type char.""

So what ? I have given you that web page. I've read it. I have read the 
comment, and more importantly, i have read what is in the table, that is 
how i should use the modifier. Is Greg's comment useful ? no.

"Then you're using a nonstandard extension."

So what ? I'm using Microsoft extension, for a program that will run on 
Windows, compiled with a compiler that should support miscrosoft 
extensions. Again, useless.

"And gcc warns that the code is nonstandard."

So what ? (bis). Same remark

>  Your question wasn't what do
> I do on Windows using GCC to get the same result as this on Linux.  Your
> question was "What did I do wrong" and Greg answered that question by
> telling you to disable the warning if you don't want them.

Did I ask : "What can I do to not have the warning ?" or "that warning is 
very annoying, how to not have it ?".

I asked:

"Am I doing something wrong or is there a bug in header files ?"

A good answer would have been :

No you are not doing something wrong because (put a justification here). 
And (yes/no) about header files

Or

Yes you are doing something wring because (put a justification).

What did I get as answer ? :

"If you don't want gcc to give this warning, then inhibit it."

And you say that he correctly replied to my question ? Are you kidding ?

For me, it was trivial that the whole purpose of my mail is : Why is there 
such warning, while there shouldn't. For me, there shouldn't be any 
warning because sscanf belongs to Microsoft libc, msvcrt*.dll. Hence, the 
mingw compiler should respect that. According to msdn, "%hc" is for 
unsigned char *. But the mingw compiler warns and say that it should be a 
short. As it is VERY important for us to be sure of what sscanf does in 
that precise case, i'm trying to understand that warning.

Vincent Torri

On Tuesday 29 June 2010 15:14:56 Vincent Torri wrote:
> On Tue, 29 Jun 2010, Earnie wrote:
> > Vincent Torri wrote:
> >> Your comments are not quite precise, you even don't answer to my
> >> question.
> >
> > Greg's comments were precise to your post.
>
> no. Indeed:
>
> "That page says "The h, l, and L prefixes are Microsoft extensions
> when used with data of type char.""
>
> So what ?

Vincent,

Please moderate your tone before I take steps to do so for you.

> I have given you that web page. I've read it.

But not particularly diligently, it would seem.

> I have read  
> the comment, and more importantly, i have read what is in the
> table, that is how i should use the modifier.

And you still think it refers to signed vs. unsigned behaviour?
Nowhere on that page do I see the slightest hint that it might serve 
that purpose.  It doesn't.  It refers exclusively to the *size* of 
the type associated with the pointer you pass as scanf argument. 

> Is Greg's comment useful ? no.

Please bear in mind that neither Greg, nor anyone else, is obliged to 
answer you at all; you should be thankful that he took the trouble to 
do so.  If this is the kind of abuse we can expect from you, when you 
don't get an entirely satisfactory (to you) answer, then don't be at 
all surprised when we simply ignore you altogether.

On Tue, 29 Jun 2010, Keith Marshall wrote:

> On Tuesday 29 June 2010 15:14:56 Vincent Torri wrote:
>> On Tue, 29 Jun 2010, Earnie wrote:
>>> Vincent Torri wrote:
>>>> Your comments are not quite precise, you even don't answer to my
>>>> question.
>>>
>>> Greg's comments were precise to your post.
>>
>> no. Indeed:
>>
>> "That page says "The h, l, and L prefixes are Microsoft extensions
>> when used with data of type char.""
>>
>> So what ?
>
> Vincent,
>
> Please moderate your tone before I take steps to do so for you.
>
>> I have given you that web page. I've read it.
>
> But not particularly diligently, it would seem.
>
>> I have read
>> the comment, and more importantly, i have read what is in the
>> table, that is how i should use the modifier.
>
> And you still think it refers to signed vs. unsigned behaviour?
> Nowhere on that page do I see the slightest hint that it might serve
> that purpose.  It doesn't.  It refers exclusively to the *size* of
> the type associated with the pointer you pass as scanf argument.
>
>> Is Greg's comment useful ? no.
>
> Please bear in mind that neither Greg, nor anyone else, is obliged to
> answer you at all; you should be thankful that he took the trouble to
> do so.  If this is the kind of abuse we can expect from you, when you
> don't get an entirely satisfactory (to you) answer, then don't be at
> all surprised when we simply ignore you altogether.

Indeed, noone is obliged to answer, as i'm not obliged to use MinGW. I'm 
fed up with you guys, with the way you treat people. You don't accept 
people to complain, but you feel free to not care about your users.

You've just lost a user

goodbye

Vincent Torri

On Wednesday 30 June 2010 00:26:13 Vincent Torri wrote:
> You've just lost a user

I won't lose any sleep over that.  One as obnoxious as you, we can do 
without.

On Wed, 30 Jun 2010, Keith Marshall wrote:

> On Wednesday 30 June 2010 00:26:13 Vincent Torri wrote:
>> You've just lost a user
>
> I won't lose any sleep over that.  One as obnoxious as you, we can do
> without.

sure, I am. I asked something and a guy basically told me that i can't 
read. Great.

Also you think you are not obnoxious ? You keep threatening 
people to not answer them because just top posting. Earnie often 
threatening too, by moderating them. **Threatening**, guys. And that's 
always users' fault, never yours.

I've subscribed to many list. That's the only one where devs are 
threatening users like some of the msys/mingw devs.

that's my last message on the msys/mingw lists, no need to moderate me.

Vincent Torri