[mythtv] Re: [mythtv-commits] mythtv commit: r7962 by danielk

[Daniel Kristjansson]

On Mon, 2005-11-21 at 17:42 +0000, Stuart Auchterlonie wrote:
> I think a better way to do this would be to setup priorities
> for EIT information.
Sure, there is a lot of improvement to be made in the EIT info.
At the moment we parse only a tiny portion of the EIT data
and we use it even if we only have only a small portion of
a full program description, so it is never a good idea to 
replace the good datadirect/XMLTV data with the EIT data if
it is describing the same program.

> Not all of these are present on all networks / transponders.
> If we add an extra column to the program table in the DB, we can
> save the information about where it came from to determine the
> best source of information.
> User level configuration should only be between xmltv & EPG.
> The specific EPG levels should be automated in the background.
We also shouldn't be overwriting good EIT collected data with
datadirect/xmltv data every night...

> Don't like this bit much...
> // To overwrite datadirect data with eit data use this instead
> //  static const QString timeFmtDB2 = "yyyy-MM-ddThh:mm:00";
> Rather than having to change the timeFmt, why don't we do a
> QRegExp replacement of 'T' to a ' '.
Huh?

Actually, I didn't expect anyone to want to do this with the current
state of EIT parsing; this is more of a "if you would like to debug
EIT by inserting lots of bogus data uncomment this" directive. This
could be made a non-const string that you set according to the global
datadirect/XMLTV vs. EIT setting you suggest; and probably should
once the EIT implementation is more complete.

-- Daniel