On Thu, Apr 04, 2002 at 06:47:33PM -0500, Michael Richardson wrote:
>   The ability to send packets is not inherently portable

Well, I suspect that it can be implemented on a number of platforms,
with a portable interface (i.e., hand it a link-layer frame and off it
goes).

That'd require a new API for opening a capture device, though, which
would include an indication of whether to open it for capture only or
for capture and sending (on some platforms, you have to open the
underlying device for reading and writing, but BPF lets you open for
reading only, and a system might be set up to allow more people to read
the BPF devices than write them).

On platforms where it can't be implemented, we'd leave it out -
applications using that API wouldn't compile on those platforms unless
the user wrote code to use some other mechanism, but they'd have to
write that code in any case....

> and is probably a bad idea for tcpdump to have too much of that.

I agree - I'm not sure whether tcpdump should be the program to actually
*use* that API.
-
This is the TCPDUMP workers list. It is archived at
http://www.tcpdump.org/lists/workers/index.html
To unsubscribe use mailto:[EMAIL PROTECTED]?body=unsubscribe

Reply via email to