Re: MinGW port usable
- From
Johannes Schindelin <johannes.schindelin@gmx.de>
- Date
- Jan 30, 2007, 17:36 UTC
- Message-ID
- <Pine.LNX.4.63.0701301835340.22628@wbgn013.biozentrum.uni-wuerzburg.de>
- In-Reply-To
- <Pine.LNX.4.64.0701301204400.20138@iabervon.org>
Hi,
On Tue, 30 Jan 2007, Daniel Barkalow wrote:
Show 24 quoted lines
> On Tue, 30 Jan 2007, Johannes Sixt wrote: > > > Daniel Barkalow wrote: > > > > > > On Mon, 29 Jan 2007, Johannes Sixt wrote: > > > > > > > (*) The reason is that on Windows read() and write() cannot operate on > > > > descriptors created by socket(). A work-around is to implement a (threaded) > > > > proxy, but that's almost the same as if netcat were used as > > > > GIT_PROXY_COMMAND. > > > > > > Can you do > > > > > > #define read(fd, buffer, len) recv(fd, buffer, len, 0) > > > #define write(fd, buffer, len) send(fd, buffer, len, 0) > > > > > > in the appropriate file? > > > > I doubt that recv and send can operate on regular file descriptors, as > > opened by _pipe(), open(), can they? > > I don't think so, but I think it should be possible to make > packet_write/packet_read always get a socket, by calling receive-pack and > upload-pack with a socket pair instead of a pair of pipes.
As I already mentioned in this thread, that would break inetd support.
Ciao, Dscho