Robert Ramey wrote:
Soooooo - from my perspective, I would like to see.
a) templated constructors added to boost.iterators
I'm not sure what you mean there.
b) a bunch of iterator adaptors - some of which are like "dataflow iterators".
My library provides that kind of thing, albeit it works element per element; maybe you were imagining a buffered system when talking of "dataflow iterators"?
c) a class for construction a codecvt facet out any iterator adaptor.
That's planned, but not done yet. Integrating with iostreams hasn't been a priority.
Previous efforts at a unicode library have seemed to bog down in codepoints and whole lot of other stuff I don't understand.
My opinion on the whole Unicode "problem" is that ultimately we're just dealing with ranges of data, and what we need to have are tools that convert and segment (i.e. iterate over specific subsequences) that data. Hence the utmost importance of ranges and iterators in the library, but those are still a hot topic (new Boost.RangeEx library, Alexandrescu talk at last Boostcon...)