From: Jeremy Siek (jsiek_at_[hidden])
Date: 2001-11-25 22:28:48
On 11/25/01 9:28 PM, "Beman Dawes" <bdawes_at_[hidden]> wrote:
> At 03:02 PM 11/24/2001, Jeremy Siek wrote:
>> I just uploaded an implementation of a dynamic sized version of
>> std::bitset to the Files (dyn_bitset.tar.gz). Also included is a complete
>> test suite and documentation. Chuck Allison provided the initial
>> implementation, and I boostified the code, and added docs and test.
>> This classes is needed for several things in the graph library, but it
>> should also be useful in a bunch of other places.
>> If people like what they see I'd like to move this to formal review
>> too long. It's a pretty simple class, and very close to the std::bitset,
>> there shouldn't be much controversy :)
> Just out of curiosity, did you and Chuck consider a design which supplied
> bitset with a policy parameter or some other way of determining whether
> allocation was static or dynamic?
> bitset and dyn_bitset seem conceptually so similar it makes me wonder if
> there is really a need for two different classes.
About 6 months ago I did an implementation of dynamic and static bitset
classes where almost all of the functionality was factored into a common
base class. That could be easily converted into a single class with a policy
parameter. However, I did not bring that up when talking with Chuck, or for
submission to Boost, because the C++ standard already has std::bitset, and I
thought it would be simpler to have something that is a pure addition
instead of something that replaces what's already there and then also makes
an extension. Does that make sense?
> In looking at the docs, I noticed (broken) "See also" links at the
> bottom. I'm looking forward to learning more about bit_vector:-)
Oops, that is cruft left over from copying from the SGI docs of std::bitset.
I'll remove that.
-- Jeremy Siek http://www.osl.iu.edu/~jsiek Ph.D. Student, Indiana Univ. B'ton email: jsiek_at_[hidden] C++ Booster (http://www.boost.org) office phone: (812) 855-3608
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk