[Xmltv-devel] _eu_epgdata: can we raise the default of 5 days?

[Karl Dietz]

Hi Jan,

every now and then someone asks why the EPGDATA grabber defaults to
only 5 days. (Usually in the context of MythTV only grabbing 5 days
instead of all available)

Can we raise that to default to "grab all days"?
That would be the correct way to do it, but its well hidden in the
wiki :-(

from http://wiki.xmltv.org/index.php/XmltvCapabilities#baseline

The default number of days is 'as many as possible'.

The root cause is that MythTV uses the "allatonce" capability and just
calls the grabber once without specifying --days and --offset.
(I remember some limit on the number of times you can request the data
each day)

btw, can we update the manpage and remove the part with "timezone are
not handled correctly" (was fixed) and also the part with "a day does
not start at midnight" (not a bug, many grabbers work like that)? (I can
remove/rewrite it if you like)

Regards,
Karl

[Ben Bucksch]

btw, can we update the manpage and remove ... the part with "a day does
not start at midnight" (not a bug, many grabbers work like that)? (I can
remove/rewrite it if you like)

Please keep this note. The way it works matches what TV stations and
magazines publish, but it's simply wrong, and it's highly confusing. It
makes sense for display to humans, but not for computer formats.

In fact, it's surprising even me. Are you saying that 2012-06-12 23:00
comes before 2012-06-12 02:00 ? That's just totally wrong. If many
grabbers work like that, they should all be fixed. It would totally kill
any logic I ever wrote and any Date class in existence.

Ben

[Karl Dietz]

Post by Ben Bucksch

btw, can we update the manpage and remove ... the part with "a day does
not start at midnight" (not a bug, many grabbers work like that)? (I can
remove/rewrite it if you like)

Please keep this note. The way it works matches what TV stations and
magazines publish, but it's simply wrong, and it's highly confusing. It
makes sense for display to humans, but not for computer formats.

Yes, its confusing. The trick is to use the date only for the first
time and properly roll over at midnight. Most of the time it works quite
well.

Post by Ben Bucksch
In fact, it's surprising even me. Are you saying that 2012-06-12 23:00
comes before 2012-06-12 02:00 ? That's just totally wrong. If many
grabbers work like that, they should all be fixed. It would totally kill
any logic I ever wrote and any Date class in existence.

I totally agree. But I thought the grabber properly outputs days so that
2012-06-12 simply starts at 2012-06-1*2*T06:00 and ends in the moment
just before 2012-06-1*3*T06:00. (taking a day that starts at 06:00 as an
example, might be anything)

Regards,
Karl

[Jan Schneider]

Post by Karl Dietz

Post by Ben Bucksch

btw, can we update the manpage and remove ... the part with "a day does
not start at midnight" (not a bug, many grabbers work like that)? (I can
remove/rewrite it if you like)

Please keep this note. The way it works matches what TV stations and
magazines publish, but it's simply wrong, and it's highly confusing. It
makes sense for display to humans, but not for computer formats.

Yes, its confusing. The trick is to use the date only for the first
time and properly roll over at midnight. Most of the time it works quite
well.

So, what's the consensus? Should I keep this note in the manpage?

Post by Karl Dietz

Post by Ben Bucksch
In fact, it's surprising even me. Are you saying that 2012-06-12 23:00
comes before 2012-06-12 02:00 ? That's just totally wrong. If many
grabbers work like that, they should all be fixed. It would totally kill
any logic I ever wrote and any Date class in existence.

I totally agree. But I thought the grabber properly outputs days so that
2012-06-12 simply starts at 2012-06-1*2*T06:00 and ends in the moment
just before 2012-06-1*3*T06:00. (taking a day that starts at 06:00 as an
example, might be anything)
Regards,
Karl
------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
xmltv-devel mailing list
https://lists.sourceforge.net/lists/listinfo/xmltv-devel