On 03/04/2011 03:49 AM, Joel Falcou wrote:
The premises are OK. We have some large integer type that has proper value semantic and can be used in a variety of operations and functions. So far, so good. The main culprit is that the current library design is rather hardwired to work with a choice of implementation the author thought to be OK at some point but which still have to be proven better than existing solution.
The overall design lacks separation between the data and the algorithm that makes the library difficult to extend
To the extent that Boost is a proving ground for things suitable for possible inclusion in a future C++ standard, it seems to me that a review of something like XInt (in particular due to its general utility) ought to focus on the quality of the interface rather than a specific implementation. If the interface really does prescribe the implementation with a great deal of specificity, then in my view it should not be adopted as a library.
: what if i want to take an arbitrary range of integers and look at them as a huge series of bits ? What if i don't want the extra memory overhead so i can mmap some files directly into memory (cause reading a 10k bits integer is a no-no) ? Can I pass the integer type to a STD algorithm if I want to iterate over the bytes it contains ? or copy them using boost::range copy ?
Or stream the bits lazily, generating them as they are needed?
- use expression template to handle efficient combination of operations and potentially detect optimisable pattern. Proto is a shoe-in for this.
Is that not prescribing an implementation? A bigint facility that strained the limits of my compiler and pushed compile times through the roof is not useful to me.
- the algorithm should be more generic and not be tied to the concrete type of large integer used.
What about just working out a bigint concept with a straightforward implementation that permitted, but did not require, expression templates? XInt could be adapted to serve as a reference implementation and a frontend with selectable back-ends. - Marsh