dont use stored event_id for repeated timers

1 files changed, 14 insertions, 11 deletions

diff --git a/RecordTimer.py b/RecordTimer.py
index 73e7eefe..1c707c36 100644
--- a/RecordTimer.py
+++ b/RecordTimer.py
@@ -164,23 +164,26 @@ class RecordTimerEntry(timer.TimerEntry, object):
 				self.log(1, "'record service' failed")
 				return False
 
-			event_id = self.eit
-			if event_id is None:
-				event_id = -1
-				
-			prep_res=self.record_service.prepare(self.Filename + ".ts", self.begin, self.end, event_id)
-			if prep_res:
-				self.log(2, "'prepare' failed: error %d" % prep_res)
-				NavigationInstance.instance.stopRecordService(self.record_service)
-				self.record_service = None
-				return False
-				
 			if self.repeated:
 				epgcache = eEPGCache.getInstance()
 				queryTime=self.begin+(self.end-self.begin)/2
 				evt = epgcache.lookupEventTime(rec_ref, queryTime)
 				if evt:
 					self.description = evt.getShortDescription()
+					event_id = evt.getEventId()
+				else:
+					event_id = -1
+			else:
+				event_id = self.eit
+				if event_id is None:
+					event_id = -1
+
+			prep_res=self.record_service.prepare(self.Filename + ".ts", self.begin, self.end, event_id)
+			if prep_res:
+				self.log(2, "'prepare' failed: error %d" % prep_res)
+				NavigationInstance.instance.stopRecordService(self.record_service)
+				self.record_service = None
+				return False
 
 			self.log(3, "prepare ok, writing meta information to %s" % self.Filename)
 			try: