On Oct 15, 2011, at 10:29 PM, Don wrote: > > std.windows is a complete joke. I don't think it's usable. There are some > really simple functions it doesn't include. Last time I used it, I even found > cases where it has a GetXXX function but not the corresponding SetXXX > function. As far as I can tell, it is the set of functions required to > compile "Empire", and then a few more functions added. > > It's possible to make a case for including only a small subset of the Windows > API in the download (ie, that part of the headers which is used by Phobos > itself). But in that case, I think it should be marketed differently, and not > called "windows". > > I would say, though, that it's pretty hard to make that case if we include > headers for SQLlite, as seems to be happening. > > The bindings Win32 project was originally set up to replace std.windows. It > was based on the MingW headers to avoid license issues. Now, I don't know > what recent contributors have done; it could be that some people have grabbed > stuff from the Windows headers (I suspect that may be true of the DirectX > stuff). > If it's true that headers aren't copyrightable, then I don't know why the > MingW headers exist. Is that really true?
Seems my memory was a bit off. From http://www.boost.org/users/license.html: "Does the copyright and license cover interfaces too? The conceptual interface to a library isn't covered. The particular representation expressed in the header is covered, as is the documentation, examples, test programs, and all the other material that goes with the library. A different implementation is free to use the same logical interface, however. Interface issues have been fought out in court several times; ask a lawyer for details." > Here's where the MingW guys ask the same question, and I think the answer is > "nobody is certain". > http://old.nabble.com/Google%27s-Android-Bionic-header-files,-and-relevance-to-w32api-headers-td31271566.html > > But, more interesting is Digital Mars's agreement with Microsoft WRT header > file distribution. It may be possible to make it a DMD download rather than > Phobos per se. Good point. Well in that case, I'd be fine with dropping in the other headers.
