I am trying to argue that the bug does not warrant closure, wine-devel does not seem like an appropriate venue for such a discussion.  I am quite familiar with how much effort goes into properly implementing a full-fledged library for working with HTTP, I have done it several times in several languages.  I initiated that bug to serve as a starting point to say "we need a minimal implementation", which I think was successfully illustrated.  I was not aware that someone was already working on something, otherwise I would have contacted them in order to get a head start.

Honestly, I believe that this problem illustrates the need for these kind of bugs - if I had found a bug for WinHttp in bugzilla assigned to Zac then I could easily contact him.  I do not have time to monitor wine-devel in order keep track of what everyone is working on so that I can keep from stepping on people's toes.  Wine is not my full time job, and I think it is unreasonable to expect people to keep track of this kind of off-the-tree activity without some sort of database like bugzilla to keep track of who is doing what.  As I'm sure other people do, I only have time to do work on Wine when I'm on my vacation.

Erich Hoover
ehoover@mines.edu

On Sun, Aug 10, 2008 at 5:52 PM, James Hawkins <truiken@gmail.com> wrote:
On Sun, Aug 10, 2008 at 6:43 PM,  <wine-bugs@winehq.org> wrote:
> http://bugs.winehq.org/show_bug.cgi?id=14810
>
>
> --- Comment #4 from Erich Hoover <ehoover@mines.edu>  2008-08-10 18:43:16 ---
> This whole DLL has a total of 27 functions, 14 of which are currently exposed,
> and only 1 is "implemented".  I put some work into implementing enough of the
> DLL to get RA3 to authenticate and found that it needs 9 of the functions to
> operate properly.  I think it is reasonable to say that implementing the rest
> of the functions for WinHttp would not be too difficult.  I do not think that
> this particular issue is anywhere close to a broad-spectrum "Win32 API needs to
> be implemented".
>

Of course it's not the same scale as all of the Win32 API.  It was an
analogy illustrating why the bug is superfluous.  On the other hand,
you're vastly underestimating the amount of work needed to properly
implement winhttp.  You might want to chat with Zac Brown who is
working on implementing winhttp right now.  Also, please don't post in
closed bugs.

--
James Hawkins