Great!
Then I don't do anything, the fixes are on their way?

/Peder




On 26 okt 2011, at 11:16, Mateusz Łoskot wrote:

> 2011/10/26 Peder Axensten <[email protected]>:
>>> I presented bug and fix here: https://gist.github.com/1313176
>> 
>> Should not the other two be fixed too:
>>        void Point::SetClassification(Classification::bitset_type const& 
>> flags) and
>>        void Point::SetClassification(boost::uint8_t const& flags)?
> 
> Yes, they should.
> 
>> Maybe a few unit tests using the various SetClassification(...) to check for 
>> this bug?
> 
> Patches are welcome, I guess.
> 
>> The only way to retrieve the uint8_t (m_data[15]) as a uint8_t is through 
>> conversion to unsigned long and back?
> 
> Yes. This is because I decided to use std::bitset users convenience to
> be able to play with bit flags easier.
> LAS classification is not 1 byte, but it's 4 different kinds of
> information encoded in space of 1 byte
> in order to save some space. Main purpose of liblas::Classification is
> to represent the classification concept
> and attributes, but not its low-level implementation.
> 
>> (By static_cast<boost::uint8_t>(bits.to_ulong()).)
> 
> or syntax sugar of functional notation which is equivalent:
> 
> boost::uint8_t(bits.to_ulong()))
> 
> Best regards,
> -- 
> Mateusz Loskot, http://mateusz.loskot.net
> Charter Member of OSGeo, http://osgeo.org
> Member of ACCU, http://accu.org

_______________________________________________
Liblas-devel mailing list
[email protected]
http://lists.osgeo.org/mailman/listinfo/liblas-devel

Reply via email to