Re: [Mingw-users] scandir and mingw

[Keith M. <kei...@us...>]

On Fri, 2008-01-04 at 17:02 -0500, Jim Marshall wrote:
> Keith Marshall wrote:
> > On Fri, 2008-01-04 at 00:40 -0500, Jim Marshall wrote:
> >> mingw is basically just a way to compile WIN32 apps without using MS 
> >> tools,
> > 
> > If that were true, then none of our mingwPORTs could be successfully
> > deployed; fortunately it isn't. 
>  
> from mingw.org:
> MinGW: A collection of freely available and freely distributable Windows 
> specific header files and import libraries combined with GNU toolsets 
> that allow one to produce native Windows programs that do not rely on 
> any 3rd-party C runtime DLLs.
> 
> That might need to be changed then

Maybe.  It is true, and it certainly doesn't assert the restricted scope
of MinGW's capabilities which some people seem to read from it; since
that impression is held by some, perhaps some further clarification
would be beneficial.

> >>  it does not provide functions which are specific to *nix systems.
> > 
> > In general, the minimalist strategy precludes provision of a full gamut
> > of *nix functionality, but there are some commonly used *nix functions
> > provided in libmingwex.a, and several others are available as optional
> > downloads, in the form of user contributed packages, or as mingwPORTs.
> 
> I was (perhaps incorrectly) under the impression that the OP just wanted 
> something that worked out of the box, hence my redirect to Cygwin.

It may be that Cygwin is the solution the OP will eventually adopt, but
please don't tell him he must do so.  Many users actually prefer to
avoid moving to Cygwin, and are prepared to put in the effort to port
their applications to use native Woe32 APIs; please don't discourage
them, by telling them they should be using Cygwin.

> The OP said "My program currently has its own replacement that gets
> called in the Windows cross-build by means of some "ifdef WIN32"
> statements."  This made it sound to me that he expected MINGW to be a
> compatibility layer.

Which of course, it isn't.

> Sorry if I provided mis-information :(

No problem.  To call it mis-information is probably too strong; however,
I did feel that it conveyed a rather negative impression of MinGW's
capabilities.

Regards,
Keith.