On 11/15/05, Isaac Richards <[EMAIL PROTECTED]> wrote:
On Tuesday 15 November 2005 06:44 pm, Dwight Hubbard wrote:
> I've been playing with the python script that was put forward a while back
> to try and implement a simple front end in python.
>
> I had to do some signifcant tweeking to get things to work due to changes
> in the protocol version. I have some basic stuff working but of course
> I am going to have to re-tweak everything for the new protocol version
> if/when I choose to go to SVN or there is a new version released.
>
> I was curious why the protocol doesn't use an extensible and easily
> parseable format for passing the data between the client and the server?
>
> For example, if I request the program info from the backend. I will
> receive a response that requires me to know what fields the program info
> response has, what order they are in, and how many fields there are
> because all the programs information are returned back to back. If the
> information was returned in a format such as XML, I could ignore new
> fields that I do not understand and continue to work even if additional
> fields where added in a new backend version.
>
> Am I missing something obvious?
Yeah, the fact that simply ignoring new fields will most often cause things to
fail just as badly as the current protocol.
Isaac
_______________________________________________
mythtv-dev mailing list
[email protected]
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev
regardless, would it be unacceptable to modify the protocol to allow clients to request an XML stream instead? So for example they could log in and then send a command to switch formatting to XML?
--
Thanks,
Devan Lippman <devan at lippman dot net>
_______________________________________________ mythtv-dev mailing list [email protected] http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev
