Re: [boost] [gil][io_new] Problems with templates

Subject: Re: [boost] [gil][io_new] Problems with templates
From: Christian Henning (chhenning_at_[hidden])
Date: 2010-10-25 10:25:28

Hi Mateusz, thanks for helping. I have fixed the issues you noted in
your last email. The problem still remains.

I'll give it another shoot again later today.

Christian

On 10/24/10, Mateusz Loskot <mateusz_at_[hidden]> wrote:
> On 24/10/10 20:06, Christian Henning wrote:
>> Hi there, I don't see the woods because of all the trees. Could
>> someone get the latest code from here:
>>
>> http://code.google.com/p/gil-contributions/source/browse/#svn/trunk
>>
>> and try to compile the following code:
>>
>> #include <boost/gil/extension/io_new/tiff_all.hpp>
>>
>> using namespace std;
>> using namespace boost;
>> using namespace gil;
>>
>> int main()
>> {
>>     typedef mpl::vector< gray8_image_t
>>                        , gray16_image_t
>>                        , rgb8_image_t
>>                        , gil::detail::gray1_image_t
>>                        > my_img_types;
>>
>>
>>     any_image< my_img_types > runtime_image;
>>
>>     read_image( "C:/gil_contributions/test_images/tiff/test.tif"
>>               , runtime_image
>>               , tiff_tag()
>>               );
>>
>>     return 0;
>> }
>>
>> Why do I have this static assertion in read.hpp[662]? I don't
>> understand why the two buffer types are different.
>>
>> Any help is very welcome,
>
>
> Christian,
>
> I'm not sure myself, but quick look suggests me there are some related
> or different problems too:
>
> In gil/extension/io_new/formats/tiff/is_allowed.hpp,
> missing template keyword around line 65
>
> p->template read_data< row_buffer_helper_view< View > >( plane, K );
>
> and missing qualifier around line 800 in:
>
>             parent_t::init_image( images
>                        , this->_info
>                        );
>
> In gil/extension/io_new/formats/tiff/is_allowed.hpp, around line 126
>
> compare_channel_sizes_fn fn( &channel_sizes.front() );
>
> does not seem to match any ctor of compare_channel_sizes_fn.
>
> There is also mismatch of unsigned short* vs unsigned int*.
>
> I will try to continue checking tomorrow.
>
> Best regards,
> --
> Mateusz Loskot, http://mateusz.loskot.net
> Charter Member of OSGeo, http://osgeo.org
> Member of ACCU, http://accu.org
>