Re: [boost] License of endian and limits in Boost detail

On Tue, Mar 19, 2013 at 9:26 AM, Andrey Semashev wrote:

On Tue, Mar 19, 2013 at 6:00 PM, Rene Rivera wrote:

...

On Tue, Mar 19, 2013 at 3:59 AM, Marc Glisse wrote:

...

BOOST_BIG_ENDIAN or BOOST_LITTLE_ENDIAN, BOOST_BYTE_ORDER which can be 4321 or 1234, etc.

If you read the documentation for the Predef library you would see that I'm following a particular naming convention that defines categories (BOOST_category_subcat_subsubcat). I do have some questions regarding the BYTE_ORDER macros..

Are the macros defining the 4321/1234 numbers actually needed? Are there situations where the on/off indicator macros aren't enough?

According to Wikipedia [1], there are cases of mixed endianness, so the BOOST_BYTE_ORDER macro could have e.g. value 3412. Also, FP endianness may differ from the one of integers, so probably there have to be separate sets of macros for these.

That, 3412, would be word-swapped big-endian. And hence already handled by BOOST_ENDIAN_BIG_WORD. And so is 2143 word-swapped little-endian as BOOST_ENDIAN_LITTLE_WORD. Not that anyone is likely to run into those in modern programming ;-) -- -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim - grafikrobot/yahoo