David Abrahams wrote:

Anthony Williams <anthony_w.geo@yahoo.com> writes:

...

I understand things differently. I thought the point of the report was to flag files that hadn't been thought about. Since these files aren't under the BSL, and can't be made to be, the boostinspect::nolicense was added precisely to address this issue, so files like these didn't clutter the report.

If that's not the intention, I can revert the changes.

I guess I don't know which intention we want to pursue. I know that our users would like to know about any Boost source files that aren't under BSL.

Do we need two reports? One (posted regularly to the list, and nagged about) which reports files which have neither a boost license nor boostinspect::nolicense; the other which also includes those files which have boostinspect::nolicense (and hence can be assumed to be under a non-boost license). The latter report would be useful to users cautious about license issues; the former would be useful to developers who need to clean things up. If not, what is the point of boostinspect::nolicense? -- Martin Bonner Martin.Bonner@Pitechnology.com Pi Technology, Milton Hall, Ely Road, Milton, Cambridge, CB4 6WZ, ENGLAND Tel: +44 (0)1223 203894