Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.c++ > #12827 > unrolled thread

vector as a char buffer ?

Back to article view | Back to comp.lang.c++

Contents

  vector as a char buffer ? "Asger Joergensen" <Junk@Asger-P.dk> - 2011-12-27 13:26 +0000
    Re: vector as a char buffer ? Victor Bazarov <v.bazarov@comcast.invalid> - 2011-12-27 09:04 -0500
      Re: vector as a char buffer ? "Asger Joergensen" <Junk@Asger-P.dk> - 2011-12-27 20:25 +0000
      Re: vector as a char buffer ? Jorgen Grahn <grahn+nntp@snipabacken.se> - 2011-12-27 21:05 +0000
    Re: vector as a char buffer ? XeCycle <xecycle@gmail.com> - 2011-12-28 12:27 +0800
      Re: vector as a char buffer ? Victor Bazarov <v.bazarov@comcast.invalid> - 2011-12-28 08:56 -0500

#12827 — vector as a char buffer ?

Hi

I'm not that used to the stl so I ask to be on the safe side:

typedef std::vector<char> TMembuf;
 
TMembuf buf(1000);
char* p = &buf[0];

I have tested it and it seem to work fine just like:

char* p = new char[1000];

and without the delete[];

But is there any downside to doing like this ?

Thanks in advance
Best regards
Asger-P

[toc] | [next] | [standalone]

#12828

On 12/27/2011 8:26 AM, Asger Joergensen wrote:
> I'm not that used to the stl so I ask to be on the safe side:
>
> typedef std::vector<char>  TMembuf;
>
> TMembuf buf(1000);
> char* p =&buf[0];
>
> I have tested it and it seem to work fine just like:
>
> char* p = new char[1000];
>
> and without the delete[];
>
> But is there any downside to doing like this ?

Not really.  Just remember that 'p' is not owned by you and that its 
validity is tightly linked to the lifetime of 'buf'.

V
-- 
I do not respond to top-posted replies, please don't ask

[toc] | [prev] | [next] | [standalone]

#12830

Hi Victor

Victor Bazarov wrote:

> On 12/27/2011 8:26 AM, Asger Joergensen wrote:
> > I'm not that used to the stl so I ask to be on the safe side:
> > 
> >typedef std::vector<char>  TMembuf;
> > 
> > TMembuf buf(1000);
> > char* p =&buf[0];
> > 
> > I have tested it and it seem to work fine just like:
> > 
> > char* p = new char[1000];
> > 
> > and without the delete[];
> > 
> > But is there any downside to doing like this ?
> 
> Not really.  Just remember that 'p' is not owned by you and that its validity is 
> tightly linked to the lifetime of 'buf'.

Thanks, and yes I know that I can only use p withing the scoope of buf.

Best regards
Asger-P

[toc] | [prev] | [next] | [standalone]

#12831

On Tue, 2011-12-27, Victor Bazarov wrote:
> On 12/27/2011 8:26 AM, Asger Joergensen wrote:
>> I'm not that used to the stl so I ask to be on the safe side:
>>
>> typedef std::vector<char>  TMembuf;
>>
>> TMembuf buf(1000);
>> char* p =&buf[0];
>>
>> I have tested it and it seem to work fine just like:
>>
>> char* p = new char[1000];
>>
>> and without the delete[];
>>
>> But is there any downside to doing like this ?
>
> Not really.  Just remember that 'p' is not owned by you and that its 
> validity is tightly linked to the lifetime of 'buf'.

And to 'buf' not being resized, swapped etc. But it's fairly easy to
manage such things -- easier than to deal with new[] IMHO.

/Jorgen

-- 
  // Jorgen Grahn <grahn@  Oo  o.   .     .
\X/     snipabacken.se>   O  o   .

[toc] | [prev] | [next] | [standalone]

#12835

"Asger Joergensen" <Junk@Asger-P.dk> writes:

> Hi

Hello,

> I'm not that used to the stl so I ask to be on the safe side:
>
> typedef std::vector<char> TMembuf;
>  
> TMembuf buf(1000);
> char* p = &buf[0];
>
> I have tested it and it seem to work fine just like:
>
> char* p = new char[1000];
>
> and without the delete[];
>
> But is there any downside to doing like this ?

This is not recommended, of course.  You didn't even use any of
the vector facilities, in this case.  Vectors provide the method
reserve().

However if you want a buffer, there's circular buffer in boost
library.

-- 
Carl Lei (XeCycle)
Department of Physics, Shanghai Jiao Tong University
OpenPGP public key: 7795E591
Fingerprint: 1FB6 7F1F D45D F681 C845 27F7 8D71 8EC4 7795 E591

[toc] | [prev] | [next] | [standalone]

#12839

On 12/27/2011 11:27 PM, XeCycle wrote:
> "Asger Joergensen"<Junk@Asger-P.dk>  writes:
>
>> Hi
>
> Hello,
>
>> I'm not that used to the stl so I ask to be on the safe side:
>>
>> typedef std::vector<char>  TMembuf;
>>
>> TMembuf buf(1000);
>> char* p =&buf[0];
>>
>> I have tested it and it seem to work fine just like:
>>
>> char* p = new char[1000];
>>
>> and without the delete[];
>>
>> But is there any downside to doing like this ?
>
> This is not recommended, of course.  You didn't even use any of
> the vector facilities, in this case.  Vectors provide the method
> reserve().

WHAT is not recommended?  And why the hell not?

And what use would Asger have for 'reserve()'?  std::vector has a proper 
constructor that allocates the buffer and makes it ready to use.

> However if you want a buffer, there's circular buffer in boost
> library.

Asger needed something analogous to an array allocated by 'new[]'.  A 
standard vector is just about what the library can offer.  Boost does 
not exist on every platform, while std::vector does.  There is no sense 
in using a non-portable library when the Standard library suffices.

V
-- 
I do not respond to top-posted replies, please don't ask

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.c++

csiph-web