Jeremy Huntwork wrote:
Looks good. :) Except, for the sake of accuracy, you're missing some includes:
I omitted them on purpose. It seems to work just fine without them. I'm guessing Glibc is coded in such a way that those includes are automatically pulled in? Or maybe GCC is smart enough to figure it out?
But yeah we'd do things right of course when we actually do anything. It was just a Proof Of Concept :)
As far as executing, I suppose you could just have the server pass whatever text it receives from the client (except if it's in a special format) to a system call. Of course, eventually we'd want to figure out some authentication.
There are a few ways that come to mind. Client can send one command at a time but that might become problematic when it comes to environment variables that will get lost. It's probably better for client to send an entire shell script, send a special "end of script" marker so the server knows when it can execute that script.
Or, maybe client will send XML to server and we'll have to add some XML parsing. There are benefits to doing the XML parsing on the client side too so the server can remain dumb. Then again, a dumb client is easier to code so anybody can write an ALFS client app and we'll have the intelligent server figure things out. There's pros and cons to everything as usual. I'm trying not to get to that point yet.
-- Gerard Beekmans /* If Linux doesn't have the solution, you have the wrong problem */ -- http://linuxfromscratch.org/mailman/listinfo/alfs-discuss FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
