|
|
Ublas : |
Subject: Re: [ublas] Support for more general ranges
From: Nasos Iliopoulos (nasos_i_at_[hidden])
Date: 2013-06-03 11:04:40
Giorgo,
this will probably help you:
http://boost.2283326.n4.nabble.com/indirect-array-strikes-back-td2711579.html
I don't recall this functionality being integrated though. We should
probably mark this as a feature request.
-Nasos
On 06/03/2013 10:10 AM, gsermaid wrote:
> Hello again,
>
> Further to the recommendation by oswin krause for trying out
> vector_indirect, I would like to ask the following.
> The documentation
> (http://www.boost.org/doc/libs/1_53_0/libs/numeric/ublas/doc/html/classboost_1_1numeric_1_1ublas_1_1vector__indirect.html#_details)
> states that a vector_indirect can be constructed by a broad range of
> vectors of indices, such as
> std::vector, ublas::vector or indirect_array.
>
> The problem is I seem to be able to construct an indirect vector using
> indirect_array types _only_. When for example I use a
> ublas::vector<size_t> or a std::vector<size_t> as a vector of indices
> I get a compilation error from gcc 4.7.2 that there is a preprocess
> member missing.
>
> Is the documentation simply out of date and only the use of
> indirect_arrays is allowed, or is it a bug? Thanks in advance.
>
> Regards,
> Giorgos
>
>
> ------------------------------------------------------------------------
>
> Message: 1
> Date: Mon, 20 May 2013 17:49:30 +0100 (BST)
> From: gsermaid <linuxfever_at_[hidden] <mailto:linuxfever_at_[hidden]>>
> To: "ublas_at_[hidden] <mailto:ublas_at_[hidden]>"
> <ublas_at_[hidden] <mailto:ublas_at_[hidden]>>
> Subject: [ublas] Support for more general ranges
> Message-ID:
> ??? <1369068570.74879.YahooMailNeo_at_[hidden]
> <mailto:1369068570.74879.YahooMailNeo_at_[hidden]>>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Hi all,
>
> I would like to ask if there is any way to create a vector proxy based
> on a general range of indices. For example, if I have
> a vector
>
> vector<unsigned> x(4);
> x <<= 1,2,3,4;
>
> and a vector of indices?
>
> vector<size_t> k(2);
> k <<= 0,3;
>
> I would like to be able to create a vector proxy as follows
>
> vector_proxy<unsigned> mysubvector(x, k);
>
> with the property that
>
> mysubvector[i] = x[k[i]].
>
> I know I can create a simple new class which provides this
> functionality but ideally I also want to have
> all the useful iterator functionality and mathematical operations
> supported by a typical ublas vector. Is this currently supported? If
> not, is there an easy way to?
> implement it and have all the functionality that ublas offer? Finally,
> do you think something like this could be included in future versions?
>
> Regards,
> Giorgos
> -------------- next part --------------
> HTML attachment scrubbed and removed
>
> ------------------------------
>
> Message: 2
> Date: 20 May 2013 18:51:54 +0200
> From: "oswin krause" <oswin.krause_at_[hidden]
> <mailto:oswin.krause_at_[hidden]>>
> To: ublas_at_[hidden] <mailto:ublas_at_[hidden]>
> Subject: Re: [ublas] Support for more general ranges
> Message-ID: <519A54AA.9020907_at_[hidden]
> <mailto:519A54AA.9020907_at_[hidden]>>
> Content-Type: text/plain; charset="iso-8859-1"; Format="flowed"
>
> Hi,
>
> take a look at vector_indirect.
>
> On 20.05.2013 18:49, gsermaid wrote:
> > Hi all,
> >
> > I would like to ask if there is any way to create a vector proxy based
> > on a general range of indices. For example, if I have
> > a vector
> >
> > vector<unsigned> x(4);
> > x <<= 1,2,3,4;
> >
> > and a vector of indices
> >
> > vector<size_t> k(2);
> > k <<= 0,3;
> >
> > I would like to be able to create a vector proxy as follows
> >
> > vector_proxy<unsigned> mysubvector(x, k);
> >
> > with the property that
> >
> > mysubvector[i] = x[k[i]].
> >
> > I know I can create a simple new class which provides this
> > functionality but ideally I also want to have
> > all the useful iterator functionality and mathematical operations
> > supported by a typical ublas vector. Is this currently supported? If
> > not, is there an easy way to
> > implement it and have all the functionality that ublas offer? Finally,
> > do you think something like this could be included in future versions?
> >
> > Regards,
> > Giorgos
> >
> >
> > _______________________________________________
> > ublas mailing list
> > ublas_at_[hidden] <mailto:ublas_at_[hidden]>
> > http://lists.boost.org/mailman/listinfo.cgi/ublas
> > Sent to: Oswin.Krause_at_[hidden]
> <mailto:Oswin.Krause_at_[hidden]>
>
> -------------- next part --------------
> HTML attachment scrubbed and removed
>
> ------------------------------
>
> Subject: Digest Footer
>
> _______________________________________________
> ublas mailing list
> ublas_at_[hidden] <mailto:ublas_at_[hidden]>
> http://lists.boost.org/mailman/listinfo.cgi/ublas
>
>
> ------------------------------
>
> End of ublas Digest, Vol 102, Issue 11
> **************************************
> -------------- next part --------------
> HTML attachment scrubbed and removed
>
> ------------------------------
>
> Subject: Digest Footer
>
> _______________________________________________
> ublas mailing list
> ublas_at_[hidden] <mailto:ublas_at_[hidden]>
> http://lists.boost.org/mailman/listinfo.cgi/ublas
>
>
> ------------------------------
>
> End of ublas Digest, Vol 102, Issue 12
> **************************************
>
>
>
>
> _______________________________________________
> ublas mailing list
> ublas_at_[hidden]
> http://lists.boost.org/mailman/listinfo.cgi/ublas
> Sent to: athanasios.iliopoulos.ctr.gr_at_[hidden]