From: Gennaro Prota (gennaro_prota_at_[hidden])
Date: 2002-08-20 17:05:53
On Tue, 20 Aug 2002 14:59:00 -0500 (EST), Jeremy Siek
>On Tue, 20 Aug 2002, Gennaro Prota wrote:
>gennar> Anyhow these are the main issues:
>gennar> a) you prefer making all data members (m_bits, m_num_bits,
>gennar> m_num_blocks) and base classes public for compilers that have
>gennar> BOOST_NO_MEMBER_TEMPLATE_FRIENDS problems. There are 6 (now 5)
>gennar> functions that are involved in this problem: operator==, operator<,
>gennar> operator >, to_block_range and dump_to_string, which need to access
>gennar> the internal representation but are currently not members.
>I respect your opinion, but chose to keep the code as it is now in this
>regard. This is not an important issue and enough time has already been
>spent on it.
I respect your opinion too. My intent was just to suggest what I
consider better for the quality of the code. No problem if you don't
agree, of course.
>gennar> b) many member functions are still extremely slow and now I don't have
>gennar> enough time to reimplement them.
>Do you not have enough time over the next couple days, or do you not have
>enough time over the next few months? There is no rush. It is not the end
>of the world if all the optimizations do not go out this release. I am
>perfectly happy with the optimizations going in slowly as you or others
>have the time to put them in.
In the next two days. Optimizations require a lot of measurements.
Though some of them are quite obvious it takes a little
experimentation to avoid some compilers' idiosyncrasies.
>There is no last useful day. August 22 is just the next release. There
>will be another release a few month later, and another in a few months
>after that. Also, people can always grab the latest code out of CVS.
Yes. I meant for this release, of course. I'll work at the changes I
have in mind and will send you the code. Considering my current
engagements I expect this to happen some day in the next week.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk