[mythtv-users] Holy porno coverart batman!

Preston Crow pc-mythtv08a at crowcastle.net
Wed Oct 9 17:55:22 UTC 2013


On 10/09/2013 01:39 PM, Thomas Mashos wrote:
> On Wed, Oct 9, 2013 at 10:26 AM, Preston Crow wrote:
>> On 10/09/2013 01:11 PM, Thomas Mashos wrote:
>>> On Wed, Oct 9, 2013 at 9:58 AM, Rob Jensen <bertaboy at gmail.com> wrote:
>>>> Is there any reason why we can't populate these when we choose the
>>>> appropriate item in the query for metadata?  It might require tweaking
>>>> the initial metadata search query, but there should be no reason why
>>>> we can't indicate whether an entry comes from ttvdb or tmdb in the
>>>> select prompt.  Selecting from one metadata source then assigns the
>>>> video as one of the options above and that assignment then becomes a
>>>> dependency for the logic between importing data from ttvdb or tmdb.
>>> That doesn't work very well for mass metadata lookups (eg. when you
>>>
>>> hit the scan button and it does a metadatalookup for all new items).
>> No, that works perfectly.  You do the lookup the way it's done now (or some
>> other way, it doesn't really matter for this point).  When you get an index
>> number from the external database, you save both the index number and a tag
>> saying which database.  Then any future use of that index number will be
>> restricted to only use that database, so you won't get images from the wrong
>> database.  One option would be to prepend a tag for the database to the
>> numbers. Another would be to have a separate field.  I can see advantages to
>> both approaches, and it doesn't really matter how it's done, as long as it
>> solves the problem.
>>
> No, that doesn't work perfectly. When I do a scan for metadata, I'm
> not prompted with a list to pick the correct item, precisely because
> my file names and the metadata lookup provide the correct result
> without user intervention (ideally, as it should be). Sure there are
> one offs, but requiring that for every item would be IMO a bad idea.
>
> And according to a previous poster, there already is a separate field
> "the CONTENTTYPE column in mythconverg:videometadata"
I think we agree that the scan should proceed without user input, 
magically getting most videos right and needing some help on ones that 
are named strangely or are just too generic.
>>> Why can't we just name files appropriately with a SxxExx and that
>>> means it's a TV show? (even accounting for specials and such that
>>> aren't necessarily episodic, we could just use S00E00)
>> Relying on how people name their files to avoid mixing up between databases
>> is crazy.  Sure, there are advantages to having files named correctly, but
>> that doesn't excuse the bug.
>>
> You're right. It's a good thing we don't tell people to do exactly that.
> http://www.mythtv.org/wiki/MythVideo_File_Parsing
>
> I thought the bug in question was where you give it an inetref number
> and it gets confused if it's a TV show or video since both databases
> use similar numbers.
>
I thought the bug in question was where the number was found by MythTV 
at one point, and then at another point it used that number to retrieve 
images from a different database from where it had found the number.  If 
you are correct, and this is an issue of a user having entered the 
number, then I would say that this isn't a bug. In that case, the only 
thing that needs to be corrected is to tweak the UI when entering a 
number to make it obvious how to specify what type of number it is.


More information about the mythtv-users mailing list