2006-06-26, 09:42 PM
Sorry, but I'm still not seeing anything that helps me work out what is wrong. Just from the quick look i've had, those insert statements look fine to me.
2006-06-26, 09:42 PM
Sorry, but I'm still not seeing anything that helps me work out what is wrong. Just from the quick look i've had, those insert statements look fine to me.
2006-06-26, 10:03 PM
sub Wrote:Sorry, but I'm still not seeing anything that helps me work out what is wrong. Just from the quick look i've had, those insert statements look fine to me.Yes I agree those statements look fine. What entries are supposed to exist in the database for the various recording types. (all episodes this channel) (all in this weakly timeslot, this channel) (all in this daily timeslot, this channel) My guess is this statement: SELECT rs.oid as rsoid, * FROM RECORDING_SCHEDULE rs left join PROGRAMME p on rs.programme_oid = p.oid order by manual_start_time But what parameter is the reoccuring?
2006-06-26, 10:07 PM
Sorry, but I dont have much time available for this. The best thing you can do is run the same tests with GB-PVR configured to use MS Access or SQLite, then look in the RECORDING_SCHEDULE table for an understanding of what data should be created.
2006-06-26, 10:14 PM
I should hopefully have some more time a available in a few days.
2006-06-26, 10:21 PM
sub Wrote:Sorry, but I dont have much time available for this. The best thing you can do is run the same tests with GB-PVR configured to use MS Access or SQLite, then look in the RECORDING_SCHEDULE table for an understanding of what data should be created.Figured that one out. in recording_schedule the recording_type are supposed to be 5. The insert statement that are supposed to acomplish this never enter the SQL server, so the error must exist somewhere in the code itself. If the Insert statement that sets the recording_type to 5 never hits the server the problem must be elsevere.
2006-06-26, 10:24 PM
I dont have any special case logic for SQLite or MS Access, and those work just fine. I treat all databases the same. The exact same statements should be executed for MSDE. Something must be failing some with one of the database insert/update/select statements with MSDE.
2006-06-28, 07:34 AM
sub Wrote:I dont have any special case logic for SQLite or MS Access, and those work just fine. I treat all databases the same. The exact same statements should be executed for MSDE. Something must be failing some with one of the database insert/update/select statements with MSDE.I must say I realy, realy, realy appreciate SUBs time and effort trying to help me. Using the info in this link i tried to catch the error in a another stage in the process, http://msdn.microsoft.com/library/defaul...access.asp with no success. I only get the same error message.I was hoping in a result similar to this: " errorMessage='Syntax error converting the varchar value '172-32-1176' to a column of data type int.' procedure='' lineNumber=1"" Does any one know if there are any way to trace the SQL queries sent to SQLlite/access and the responses sent from SQLlite/access? Is there any way to get a better output from GBPVR logs ?
2006-06-28, 05:51 PM
Quote:Does any one know if there are any way to trace the SQL queries sent to SQLlite/access and the responses sent from SQLlite/access?Sorry, I'm not sure. You could try asking here: http://sqlite.phxsoftware.com/. These guys are the authors of the SQLite Provider. Quote:Is there any way to get a better output from GBPVR logs ?There is no way to log the SQL to GB-PVR's logs.
2006-06-29, 09:25 PM
Urbansk Wrote:I must say I realy, realy, realy appreciate SUBs time and effort trying to help me. I haven't been around lately to mess around with this issue, but I would definitely like to thank Sub too for the time he has spent on helping to debug this issue. Urbansk Wrote:Does any one know if there are any way to trace the SQL queries sent to SQLlite/access and the responses sent from SQLlite/access? I noticed the source code for the ADO.NET provider is available for download. We could put in some logging statements and then recompile. I could look into this next week if you think it's a worth while effort. |
|