BOOST Archives Google Group

[Boost-users] boost.units converting double tp SI prefix

ope-de...@gmx.de (Olaf Peter) — Fri, 12 мар 2010 18:00:58 UT

Hi,
is there a way to use boost.units to convert a double automatically to a
SI unit prefix, e.g. 10e-3 to 10m or 10e3 to 10k ?? The value isn't know
at compile time, so the decision has to be done on runtime.
Thanks,
Olaf
______________________________ _________________
Boost-users mailing list
Boost-us...@lists.boost.org

Re: [Boost-users] uBlas with Atlas

rut...@terborg.net (Rutger ter Borg) — Fri, 12 мар 2010 17:20:18 UT

Sure, I can help with the one without the -v1 suffix. The regressions tests
should work for 98%, the examples however aren't ported yet to the newer
bindings structure. One of the major changes in the newer bindings is that
if you would like to use atlas, it's still blas, so
bindings::atlas::gemm( ... )

[Boost-users] uBlas with Atlas

philipp.kr...@flashpixx.de (Kraus Philipp) — Fri, 12 мар 2010 17:08:30 UT

Hello,
I try to install the numerical bindings for my boost, but I can't
compile
the examples. I had installed and compilde the Atlas Lib 3.9.23 with
Lapack support and the Boost 1.42 with MPI Support under OS X Leo.
I take a look at [link]
but I had checked out the bindungs with and without the -v1 prefix and

Re: [Boost-users] storing pointers in a std::map

alan2h...@gmail.com (Alan Tennant) — Fri, 12 мар 2010 16:13:39 UT

Thanks. I presume it would be better in implementing a map that can hold
left = std::string, right = "any" but only pointers to objects to have a
boost::map_ptr<std::string> than a std::map<std::string, boost::any>.

[Boost-users] matrix iterator each element

philipp.kr...@flashpixx.de (Kraus Philipp) — Fri, 12 мар 2010 10:23:49 UT

Hello,
I have very large matrices (not sparse), where I have to change
each element. From the documentation I used the example
with which two for-loops.
I use for vectors:
blas::vector<T> l_vec = p_vec;
BOOST_FOREACH( T& i, l_vec)
i = std::pow(i, p_exponent);
Is there any way to process each element in

[Boost-users] check MPI support

philipp.kr...@flashpixx.de (Kraus Philipp) — Fri, 12 мар 2010 10:15:53 UT

Hello,
I work for a long time again with C + + and some numerical methods
must be implemented with the help of the boost, among other things,
with MPI. Does the Booster precompiler flag that I can verify that the
boost is compiled with MPI support like:
Thanks for help
Phil
______________________________ _________________

Re: [Boost-users] [asio] Callbacks not getting called

rut...@terborg.net (Rutger ter Borg) — Fri, 12 мар 2010 14:43:19 UT

As Igor suggested, the io_service running your client runs out of work after
connecting. In addition, you should replace
gotTransfer_;
by
gotTransfer_ = true;
Cheers,
Rutger
______________________________ _________________
Boost-users mailing list
Boost-us...@lists.boost.org
[link]

Re: [Boost-users] [asio] Callbacks not getting called

rut...@terborg.net (Rutger ter Borg) — Fri, 12 мар 2010 14:04:10 UT

IIRC, an io_service object is not permitted to make copies of arguments
supplied to initiating functions outside its own thread. In your example,
you initiate an asynchronous operation from thread A (async_accept), and
io_service runs in thread B (io_service.run()).
At least, that's if I understood [link] correctly, the

Re: [Boost-users] boost::filesystem directory_iterator: how to get the last file of a directory ?

stras...@uni-bremen.de — Fri, 12 мар 2010 13:36:44 UT

Zitat von "Vial, Florent" <fv...@arri.de>:
define "first" and "last".
directory iterators are single-pass input iterators, meaning that they
represent no particular order.
if you iterate over the same directory 2 times, you might retrieve the
entries in a different order.
you could e.g. copy the entries to a std::vector using std::copy and

[Boost-users] boost::filesystem directory_iterator: how to get the last file of a directory ?

fv...@arri.de (Vial, Florent) — Fri, 12 мар 2010 11:54:33 UT

Hi boost-users,
I am using a recursvive function for iterating over a starting path.
When i finally arrive in a directory that contains no subdirectory
(what i call a "final" directory), i want to retrieve
the first and the last file of this directory.
I succeed in retrieving the first file of the final directory (see code

Re: [Boost-users] Nested regex results with xpressive.

suryakiran.gullapa...@gmail.com (Surya Kiran Gullapalli) — Fri, 12 мар 2010 11:44:09 UT

Got it.. thanks for the explanation.
Surya

Re: [Boost-users] [fusion] transform & iterators

mr.chr.schm...@online.de (Christopher Schmidt) — Fri, 12 мар 2010 11:40:25 UT

David A. Greene schrieb:
Change the signature of square's invocation operator to
result_type operator()(int i) const
and the sample should compile fine.
[snip]
That's correct.
Joel referred to fusion::transform taking the sequence cref qualified
only. This should not be a limitation for your particular use-case though.

Re: [Boost-users] Nested regex results with xpressive.

e...@boostpro.com (Eric Niebler) — Fri, 12 мар 2010 11:22:57 UT

Yes.
Yes.
Nothing. The three nested results correspond to the three regexes nested
in combinedReg: numReg, holidayName, and holidayDateDetails. None of
those regexes has other regexes nested within them.
Are you confusing nested regexes with marked sub-expressions? Perhaps if
you change your innermost nested FOREACH statement with something like this:

[Boost-users] Nested regex results with xpressive.

suryakiran.gullapa...@gmail.com (Surya Kiran Gullapalli) — Fri, 12 мар 2010 10:54:48 UT

Hello all,
I'm trying to match a string with xpressive. The match is fine, but the
nested results doesn't seem to be working. The code is
here<[link]>
my regex *combinedReg* is made up of 3 sregex objects. Two of them store the
results using mark_tag objects. When I try to match the string, I get a

Re: [Boost-users] boost::regex on to match in the middle of a string

abruzzoforteegent...@gmail.com (Conoscenza Silente) — Fri, 12 мар 2010 08:47:16 UT

Hi
I thank you all for your kind help!
Everything works fine using boost::regex_search
I am also sorry because I didn't realize that my regex was incorreect
because of \t
Have a nice day to everybody
CS