Fwd: [GIL] Point concept
---------- Forwarded message ---------- From: Felipe Magno de Almeida <felipe.m.almeida@gmail.com> Date: Thu, Jul 10, 2008 at 3:28 AM Subject: [GIL] Point concept To: boost-users@lists.boost.org Hi, PointNDConcept is defined as having axis_value member functions. But the GIL code in the boost trunk doesn't comply with this requirement. A free axis_value function is just useless, because it can't be found by ADL when template parameters are passed explicitly. Also, member template axis doesn't model Metafunction. Am I missing something? Regards, -- Felipe Magno de Almeida -- Felipe Magno de Almeida
From: Felipe Magno de Almeida <felipe.m.almeida@gmail.com> PointNDConcept is defined as having axis_value member functions. <...skipped...> Also, member template axis doesn't model Metafunction.
Felipe, Thanks for discovering these two discrepancies. axis_value should be a member function, and the axis member template should be changed to be a real metafunction. The documentation need not change. I will add these to the list of fixes to GIL. Lubomir
Lubomir wrote:
From: Felipe Magno de Almeida <felipe.m.almeida@gmail.com> PointNDConcept is defined as having axis_value member functions. <...skipped...> Also, member template axis doesn't model Metafunction.
Felipe,
Thanks for discovering these two discrepancies. axis_value should be a member function, and the axis member template should be changed to be a real metafunction. The documentation need not change. I will add these to the list of fixes to GIL.
Lubomir, Do you think that parhps since Fusion has come out you might consider making your points & pixels fusion vectors or maps?
John Femiani wrote:
Do you think that parhps since Fusion has come out you might consider making your points & pixels fusion vectors or maps?
I support tighter integration of Boost libraries, and I have tried to reuse code wherever possible, with a few exceptions. For example I wrote a separate version of variant, because in my early tests I found that using the Boost version doubled the compile time of the GIL regression tests (things may have changed since then). I would be interested in replacing parts of GIL with Fusion provided it does not decrease the compile time and, of course, run time performance. Unfortunately I won't be able to make any major changes to GIL any time soon, but we encourage people's contributions towards improving GIL. Lubomir
participants (3)
-
Felipe Magno de Almeida -
John Femiani -
Lubomir Bourdev