My code compiles fine with gcc-4.1.1, but chokes on gcc-3.4.6.  I'm not sure what's going on here:

/usr/local/src/stlsoft/include/stlsoft/containers/fixed_array.hpp: In copy constructor `stlsoft::fixed_array_3d<T, A, P, R>::fixed_array_3d(const stlsoft::fixed_array_3d<T, A, P, R>&) [with T = double, A = std::allocator<double>, P = stlsoft::do_construction<double>, bool R =  false]':
/usr/local/src/stlsoft/include/stlsoft/containers/fixed_array.hpp:1838:
  instantiated from `typename stlsoft::fixed_array_4d<T, A, P, R>::dimension_element_type stlsoft::fixed_array_4d<T, A, P, R>::operator[](stlsoft::ss_size_t) [with T = double, A = std::allocator<double>, P = stlsoft::do_construction<double>, bool R =  true]'
idma/turbo_dec_1_10_2.H:387:
  instantiated from `void turbo_dec_1_10<flt_t>::compute_gamma(const apriori_t&, const rec_enc_parity_t&, const rec_tail_t&, gamma_t&, gamma2_t&) [with apriori_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_initialisation_never,  true>, rec_enc_parity_t = stlsoft::fixed_array_2d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, rec_tail_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, gamma_t = stlsoft::fixed_array_3d<double, std::allocator<double>, stlsoft::do_construction<double>,  true>, gamma2_t = stlsoft::fixed_array_4d<double, std::allocator<double>, stlsoft::do_construction<double>,  true>, flt_t = double]'
idma/turbo_dec_1_10_2.H:342:
  instantiated from `void turbo_dec_1_10<flt_t>::turbo_constituent_dec(const apriori_t&, const rec_enc_parity_t&, const rec_tail_t&, info_extrinsic_t&, parity_extrinsic_t&, tail_extrinsic_t&) [with apriori_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_initialisation_never,  true>, rec_enc_parity_t = stlsoft::fixed_array_2d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, rec_tail_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, info_extrinsic_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_construction<double>,  true>, parity_extrinsic_t = stlsoft::fixed_array_2d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, tail_extrinsic_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, flt_t = double]'
idma/turbo_dec_1_10_2.H:250:
   instantiated from `void turbo_dec_1_10<flt_t>::iter(const in_t&, flt_t) [with in_t = boost::numeric::ublas::vector<double, boost::numeric::ublas::unbounded_array<double, std::allocator<double> > >, flt_t = double]'
idma/turbo_dec_1_10.cc:192:
   instantiated from here
/usr/local/src/stlsoft/include/stlsoft/containers/fixed_array.hpp:1441: error: creating array with size zero (`0')

The line that triggered the error:
        gamma2[0][k][m][0] = gamma[k][m][1];

I'm sorry, but I can't post the source for this.  Perhaps someone has encountered a similar problem?

"Neal Becker" <ndbecker2@gmail.com> wrote in message news:ep3233$11ba$1@digitaldaemon.com...
> My code compiles fine with gcc-4.1.1, but chokes on gcc-3.4.6.  I'm not
sure what's going on here:
>
> /usr/local/src/stlsoft/include/stlsoft/containers/fixed_array.hpp: In copy
constructor `stlsoft::fixed_array_3d<T, A, P, R>::fixed_array_3d(const stlsoft::fixed_array_3d<T, A, P, R>&) [with T = double, A = std::allocator<double>, P = stlsoft::do_construction<double>, bool R = false]':
> /usr/local/src/stlsoft/include/stlsoft/containers/fixed_array.hpp:1838:
>   instantiated from `typename stlsoft::fixed_array_4d<T, A, P,
R>::dimension_element_type stlsoft::fixed_array_4d<T, A, P, R>::operator[](stlsoft::ss_size_t) [with T = double, A = std::allocator<double>, P = stlsoft::do_construction<double>, bool R = true]'
> idma/turbo_dec_1_10_2.H:387:
>   instantiated from `void turbo_dec_1_10<flt_t>::compute_gamma(const
apriori_t&, const rec_enc_parity_t&, const rec_tail_t&, gamma_t&, gamma2_t&) [with apriori_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_initialisation_never,  true>, rec_enc_parity_t = stlsoft::fixed_array_2d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, rec_tail_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, gamma_t = stlsoft::fixed_array_3d<double, std::allocator<double>, stlsoft::do_construction<double>,  true>, gamma2_t = stlsoft::fixed_array_4d<double, std::allocator<double>, stlsoft::do_construction<double>,  true>, flt_t = double]'
> idma/turbo_dec_1_10_2.H:342:
>   instantiated from `void
turbo_dec_1_10<flt_t>::turbo_constituent_dec(const apriori_t&, const rec_enc_parity_t&, const rec_tail_t&, info_extrinsic_t&, parity_extrinsic_t&, tail_extrinsic_t&) [with apriori_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_initialisation_never,  true>, rec_enc_parity_t = stlsoft::fixed_array_2d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, rec_tail_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, info_extrinsic_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_construction<double>,  true>, parity_extrinsic_t = stlsoft::fixed_array_2d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, tail_extrinsic_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, flt_t = double]'
> idma/turbo_dec_1_10_2.H:250:
>    instantiated from `void turbo_dec_1_10<flt_t>::iter(const in_t&, flt_t)
[with in_t = boost::numeric::ublas::vector<double, boost::numeric::ublas::unbounded_array<double, std::allocator<double> > >, flt_t = double]'
> idma/turbo_dec_1_10.cc:192:
>    instantiated from here
> /usr/local/src/stlsoft/include/stlsoft/containers/fixed_array.hpp:1441:
error: creating array with size zero (`0')
>
> The line that triggered the error:
>         gamma2[0][k][m][0] = gamma[k][m][1];
>
> I'm sorry, but I can't post the source for this.  Perhaps someone has
encountered a similar problem?

It's a bit hard to tell from the report.

Can you post a chopped down version, or a little sample that brings out the same bug?

btw, did the fe.N mods fix your 64-bit build problem? If so, I'll try and get a new Pantheios release out asap.

Cheers

Matthew

Matthew Wilson wrote:

>
> btw, did the fe.N mods fix your 64-bit build problem? If so, I'll try and get a new Pantheios release out asap.
> 

Looks like it gets further.  Now stops here:
g++   -c -Wall -pedantic -Wno-long-long -Wno-unused-value -Werror  -O1   -D_REENTRANT  -D_DEBUG -UNDEBUG  -Dunix  -I../../include -I"/usr/local/src/stlsoft/include"    -o test_api.debug.o ../../test/test_api/test_api.cpp
../../test/test_api/test_api.cpp: In function ‘stlsoft::basic_shim_string<char, 64ul, false, std::allocator<char>, stlsoft::stlsoft_char_traits<char> > stlsoft::c_str_data_a(const Thingy&)’:
../../test/test_api/test_api.cpp:142: error: cast from ‘const Thingy*’ to ‘unsigned int’ loses precision
../../test/test_api/test_api.cpp: In function ‘std::string Thingy_to_string(const Thingy*)’:
../../test/test_api/test_api.cpp:168: error: cast from ‘const Thingy**’ to ‘unsigned int’ loses precision
make: *** [test_api.debug.o] Error 1

> The line that triggered the error:
>         gamma2[0][k][m][0] = gamma[k][m][1];
> 
> I'm sorry, but I can't post the source for this.  Perhaps someone has encountered a similar problem?

I'm trying to have a go at this, but without knowing what gamma and gamm2 are, it's pretty near impossible. Can you tell me how they're defined?

Thanks

Matthew

Try this patch.

No guarantees ... ;-)

If it works, I'll release a beta 44.

Cheers

Matthew


"Neal Becker" <ndbecker2@gmail.com> wrote in message news:ep3233$11ba$1@digitaldaemon.com...
> My code compiles fine with gcc-4.1.1, but chokes on gcc-3.4.6.  I'm not
sure what's going on here:
>
> /usr/local/src/stlsoft/include/stlsoft/containers/fixed_array.hpp: In copy
constructor `stlsoft::fixed_array_3d<T, A, P, R>::fixed_array_3d(const stlsoft::fixed_array_3d<T, A, P, R>&) [with T = double, A = std::allocator<double>, P = stlsoft::do_construction<double>, bool R = false]':
> /usr/local/src/stlsoft/include/stlsoft/containers/fixed_array.hpp:1838:
>   instantiated from `typename stlsoft::fixed_array_4d<T, A, P,
R>::dimension_element_type stlsoft::fixed_array_4d<T, A, P, R>::operator[](stlsoft::ss_size_t) [with T = double, A = std::allocator<double>, P = stlsoft::do_construction<double>, bool R = true]'
> idma/turbo_dec_1_10_2.H:387:
>   instantiated from `void turbo_dec_1_10<flt_t>::compute_gamma(const
apriori_t&, const rec_enc_parity_t&, const rec_tail_t&, gamma_t&, gamma2_t&) [with apriori_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_initialisation_never,  true>, rec_enc_parity_t = stlsoft::fixed_array_2d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, rec_tail_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, gamma_t = stlsoft::fixed_array_3d<double, std::allocator<double>, stlsoft::do_construction<double>,  true>, gamma2_t = stlsoft::fixed_array_4d<double, std::allocator<double>, stlsoft::do_construction<double>,  true>, flt_t = double]'
> idma/turbo_dec_1_10_2.H:342:
>   instantiated from `void
turbo_dec_1_10<flt_t>::turbo_constituent_dec(const apriori_t&, const rec_enc_parity_t&, const rec_tail_t&, info_extrinsic_t&, parity_extrinsic_t&, tail_extrinsic_t&) [with apriori_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_initialisation_never,  true>, rec_enc_parity_t = stlsoft::fixed_array_2d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, rec_tail_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, info_extrinsic_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_construction<double>,  true>, parity_extrinsic_t = stlsoft::fixed_array_2d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, tail_extrinsic_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, flt_t = double]'
> idma/turbo_dec_1_10_2.H:250:
>    instantiated from `void turbo_dec_1_10<flt_t>::iter(const in_t&, flt_t)
[with in_t = boost::numeric::ublas::vector<double, boost::numeric::ublas::unbounded_array<double, std::allocator<double> > >, flt_t = double]'
> idma/turbo_dec_1_10.cc:192:
>    instantiated from here
> /usr/local/src/stlsoft/include/stlsoft/containers/fixed_array.hpp:1441:
error: creating array with size zero (`0')
>
> The line that triggered the error:
>         gamma2[0][k][m][0] = gamma[k][m][1];
>
> I'm sorry, but I can't post the source for this.  Perhaps someone has
encountered a similar problem?

"Neal Becker" <ndbecker2@gmail.com> wrote in message news:ep3iho$1mp7$1@digitaldaemon.com...
> Matthew Wilson wrote:
>
> >
> > btw, did the fe.N mods fix your 64-bit build problem? If so, I'll try
and
> > get a new Pantheios release out asap.
> >
>
> Looks like it gets further.  Now stops here:
>
+   -c -Wall -pedantic -Wno-long-long -Wno-unused-value -Werror  -O1   -D_RE ENTRANT  -D_DEBUG -UNDEBUG  -Dunix  -I../../include -I"/usr/local/src/stlsof t/include"    -o test_api.debug.o ../../test/test_api/test_api.cpp
> ../../test/test_api/test_api.cpp: In function
'stlsoft::basic_shim_string<char, 64ul, false, std::allocator<char>, stlsoft::stlsoft_char_traits<char> > stlsoft::c_str_data_a(const Thingy&)':
> ../../test/test_api/test_api.cpp:142: error: cast from 'const Thingy*' to
'unsigned int' loses precision
> ../../test/test_api/test_api.cpp: In function 'std::string
Thingy_to_string(const Thingy*)':
> ../../test/test_api/test_api.cpp:168: error: cast from 'const Thingy**' to
'unsigned int' loses precision
> make: *** [test_api.debug.o] Error 1

Well, the good news is that we seem to be out of the libraries, and into the test programs (about which I'm less bothered).

I'll check into it.

A favour, if I may. Could you execute the following build targets on your
64-bit system:
    "make -f makefile.unix build.libs.core"
    "make -f makefile.unix build.libs.fe"
    "make -f makefile.unix build.libs.be"
    "make -f makefile.unix build.examples"

and let me know which, if any, have any problems.

Thanks

Matthew

Matthew Wilson wrote:

> 
> "Neal Becker" <ndbecker2@gmail.com> wrote in message news:ep3iho$1mp7$1@digitaldaemon.com...
>> Matthew Wilson wrote:
>>
>> >
>> > btw, did the fe.N mods fix your 64-bit build problem? If so, I'll try
> and
>> > get a new Pantheios release out asap.
>> >
>>
>> Looks like it gets further.  Now stops here:
>>
> +   -c -Wall -pedantic -Wno-long-long -Wno-unused-value -Werror  -O1
> -D_RE
> ENTRANT  -D_DEBUG -UNDEBUG  -Dunix  -I../../include
> -I"/usr/local/src/stlsof
> t/include"    -o test_api.debug.o ../../test/test_api/test_api.cpp
>> ../../test/test_api/test_api.cpp: In function
> 'stlsoft::basic_shim_string<char, 64ul, false, std::allocator<char>, stlsoft::stlsoft_char_traits<char> > stlsoft::c_str_data_a(const Thingy&)':
>> ../../test/test_api/test_api.cpp:142: error: cast from 'const Thingy*' to
> 'unsigned int' loses precision
>> ../../test/test_api/test_api.cpp: In function 'std::string
> Thingy_to_string(const Thingy*)':
>> ../../test/test_api/test_api.cpp:168: error: cast from 'const Thingy**' to
> 'unsigned int' loses precision
>> make: *** [test_api.debug.o] Error 1
> 
> Well, the good news is that we seem to be out of the libraries, and into the test programs (about which I'm less bothered).
> 
> I'll check into it.
> 
> A favour, if I may. Could you execute the following build targets on your
> 64-bit system:
>     "make -f makefile.unix build.libs.core"
>     "make -f makefile.unix build.libs.fe"
>     "make -f makefile.unix build.libs.be"
>     "make -f makefile.unix build.examples"
> 
> and let me know which, if any, have any problems.
> 

No problems.

Matthew Wilson wrote:

> Try this patch.
> 
> No guarantees ... ;-)
> 
> If it works, I'll release a beta 44.

Yes, it seems to compile now (at least).

In response to your other question,
  stlsoft::fixed_array_3d<flt_t> gamma (MSG_BITS + CONST_LEN, DEC_STATES, 2);
  stlsoft::fixed_array_4d<flt_t> gamma2 (E, MSG_BITS + CONST_LEN, DEC_STATES, 2);

> 
> Cheers
> 
> Matthew
> 
> 
> "Neal Becker" <ndbecker2@gmail.com> wrote in message news:ep3233$11ba$1@digitaldaemon.com...
>> My code compiles fine with gcc-4.1.1, but chokes on gcc-3.4.6.  I'm not
> sure what's going on here:
>>
>> /usr/local/src/stlsoft/include/stlsoft/containers/fixed_array.hpp: In copy
> constructor `stlsoft::fixed_array_3d<T, A, P, R>::fixed_array_3d(const stlsoft::fixed_array_3d<T, A, P, R>&) [with T = double, A = std::allocator<double>, P = stlsoft::do_construction<double>, bool R = false]':
>> /usr/local/src/stlsoft/include/stlsoft/containers/fixed_array.hpp:1838:
>>   instantiated from `typename stlsoft::fixed_array_4d<T, A, P,
> R>::dimension_element_type stlsoft::fixed_array_4d<T, A, P, R>::operator[](stlsoft::ss_size_t) [with T = double, A = std::allocator<double>, P = stlsoft::do_construction<double>, bool R = true]'
>> idma/turbo_dec_1_10_2.H:387:
>>   instantiated from `void turbo_dec_1_10<flt_t>::compute_gamma(const
> apriori_t&, const rec_enc_parity_t&, const rec_tail_t&, gamma_t&,
> gamma2_t&)
> [with apriori_t = stlsoft::fixed_array_1d<double, std::allocator<double>,
> stlsoft::do_initialisation_never,  true>, rec_enc_parity_t =
> stlsoft::fixed_array_2d<double, std::allocator<double>,
> stlsoft::do_construction<double>,  false>, rec_tail_t =
> stlsoft::fixed_array_1d<double, std::allocator<double>,
> stlsoft::do_construction<double>,  false>, gamma_t =
> stlsoft::fixed_array_3d<double, std::allocator<double>,
> stlsoft::do_construction<double>,  true>, gamma2_t =
> stlsoft::fixed_array_4d<double, std::allocator<double>,
> stlsoft::do_construction<double>,  true>, flt_t = double]'
>> idma/turbo_dec_1_10_2.H:342:
>>   instantiated from `void
> turbo_dec_1_10<flt_t>::turbo_constituent_dec(const apriori_t&, const rec_enc_parity_t&, const rec_tail_t&, info_extrinsic_t&, parity_extrinsic_t&, tail_extrinsic_t&) [with apriori_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_initialisation_never,  true>, rec_enc_parity_t = stlsoft::fixed_array_2d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, rec_tail_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, info_extrinsic_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_construction<double>,  true>, parity_extrinsic_t = stlsoft::fixed_array_2d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, tail_extrinsic_t = stlsoft::fixed_array_1d<double, std::allocator<double>, stlsoft::do_construction<double>,  false>, flt_t = double]'
>> idma/turbo_dec_1_10_2.H:250:
>>    instantiated from `void turbo_dec_1_10<flt_t>::iter(const in_t&,
>>    flt_t)
> [with in_t = boost::numeric::ublas::vector<double, boost::numeric::ublas::unbounded_array<double, std::allocator<double> > >, flt_t = double]'
>> idma/turbo_dec_1_10.cc:192:
>>    instantiated from here
>> /usr/local/src/stlsoft/include/stlsoft/containers/fixed_array.hpp:1441:
> error: creating array with size zero (`0')
>>
>> The line that triggered the error:
>>         gamma2[0][k][m][0] = gamma[k][m][1];
>>
>> I'm sorry, but I can't post the source for this.  Perhaps someone has
> encountered a similar problem?

"Neal Becker" <ndbecker2@gmail.com> wrote in message news:ep4vdu$end$1@digitaldaemon.com...
> Matthew Wilson wrote:
>
> > Try this patch.
> >
> > No guarantees ... ;-)
> >
> > If it works, I'll release a beta 44.
>
> Yes, it seems to compile now (at least).

Good.

Don't forget to let me know if it run.  :-)

"Neal Becker" <ndbecker2@gmail.com> wrote in message news:ep4v24$e9v$1@digitaldaemon.com...
> Matthew Wilson wrote:
>
> >
> > "Neal Becker" <ndbecker2@gmail.com> wrote in message news:ep3iho$1mp7$1@digitaldaemon.com...
> >> Matthew Wilson wrote:
> >>
> >> >
> >> > btw, did the fe.N mods fix your 64-bit build problem? If so, I'll try
> > and
> >> > get a new Pantheios release out asap.
> >> >
> >>
> >> Looks like it gets further.  Now stops here:
> >>
> > +   -c -Wall -pedantic -Wno-long-long -Wno-unused-value -Werror  -O1
> > -D_RE
> > ENTRANT  -D_DEBUG -UNDEBUG  -Dunix  -I../../include
> > -I"/usr/local/src/stlsof
> > t/include"    -o test_api.debug.o ../../test/test_api/test_api.cpp
> >> ../../test/test_api/test_api.cpp: In function
> > 'stlsoft::basic_shim_string<char, 64ul, false, std::allocator<char>, stlsoft::stlsoft_char_traits<char> > stlsoft::c_str_data_a(const Thingy&)':
> >> ../../test/test_api/test_api.cpp:142: error: cast from 'const Thingy*'
to
> > 'unsigned int' loses precision
> >> ../../test/test_api/test_api.cpp: In function 'std::string
> > Thingy_to_string(const Thingy*)':
> >> ../../test/test_api/test_api.cpp:168: error: cast from 'const Thingy**' to
> > 'unsigned int' loses precision
> >> make: *** [test_api.debug.o] Error 1
> >
> > Well, the good news is that we seem to be out of the libraries, and into the test programs (about which I'm less bothered).
> >
> > I'll check into it.
> >
> > A favour, if I may. Could you execute the following build targets on
your
> > 64-bit system:
> >     "make -f makefile.unix build.libs.core"
> >     "make -f makefile.unix build.libs.fe"
> >     "make -f makefile.unix build.libs.be"
> >     "make -f makefile.unix build.examples"
> >
> > and let me know which, if any, have any problems.
> >
>
> No problems.

Cool. Thanks