Hi Neal,
I am really excited to see program_options progress. From my limited testing so far it looks really good, and I have been looking forward to using it in my code. I hope we can add it to boost soon, because I'm reluctant to rely on unreleased boost code.
Thanks for your encouragement! I, too, hope that the library will be added to CVS soon -- the only issue that was specifically requested as needed before inclusion is Unicode support. I've already made up my mind how to best add it, but it might take a couple of weeks to discuss the approach and really implement it.
The doc needs work. It was hard for me to find the doc for the function parse_environment for example. There needs to be links from the top-level html page.
I though there should be annotated list of symbols: a table with a name of symbols (classes/functions) and a brief description. Maybe, it can be a full table for all symbols, or maybe several tables for each component. What do you think?
Also, how can I build the html docs myself from the distributed xml? It looks like I need the "boost v2" build stuff, but I don't know what that means.
In fact, "boost build v2" is the simplest thing of all, because it's already in boost. You'd need to get XSLT processor and Docbook and configure them as described in http://www.boost.org/regression-logs/cs-win32_metacomm/doc/html/boostbook.ge... After that, running "bjam --v2" in the "doc" directory should produce a bunch of html files. - Volodya