Migrate TMDB to using native IDs rather than IMDB ones

[anonymous]

This patch strips out the IMDB compatibility from tmdb.pl, now using the TMDB IDs directly. Included is a quick bash script to convert existing video libraries from IMDB inetref's to TMDB ones.

[wstewart@…]

Is there a good reason to do this? When I created the tmdb script, I spoke with Travis at tmdb. I recall he told me the imdb inetref would always be there.

[r.d.vaughan@…]

I have been using the tmdb.pl patch " http://svn.mythtv.org/trac/ticket/6421" as it cleaned up a number of annoying warning messages. This new patch has issues with patch 6421.

Is there any way to resolve this conflict?

[raymond@…]

Replying to wstewart@hgrace.com:

Is there a good reason to do this? When I created the tmdb script, I spoke with Travis at tmdb. I recall he told me the imdb inetref would always be there.

It started as a complaint on #mythtv-users about how the script cannot use Movies that do not have a IMDB reference. The only reason to use IMDB IDs in the first place is to maintain compatibility with the old scraper and video collections. Since IMDB pads their IDs to 7 digits, anything not converted will not conflict with the TMDB grabber until its database runs into the millions of entries.

[raymond@…]

merged changes with 'tmdb.pl_v4.patch' in #6421

[raymond@…]

Replying to r.d.vaughan@rogers.com:

Is there any way to resolve this conflict?

The above patch should merge the two changesets.

[r.d.vaughan@…]

Thanks for the merged change sets.

This may not be acceptable, but I would like the IMDB number added to the meta data (-D) output if it exists on themoviedb.com. The reason is that some scripts like fetch_poster.py still use IMDB numbers.

The script Jamu ( http://www.mythtv.org/wiki/Jamu), allows the user to specify secondary sources for data and graphics to augment the gaps in themoviedb.com wiki. Almost all of the secondary scripts still use IMDB numbers.

I full back the move to use a source's own reference number but I want to keep all options open. As I understand it both of the primary data and graphics TV and movie wiki's intend to keep IMDB numbers part of their data sets and api search options.

[raymond@…]

current with changeset [20606].

[brandon.d.shields@…]

Hi,

Just a comment from the side line. Could the "inetref" be associated with a reference type that holds the url template for interpreting that inetref? That way multiple reference types could be supported. The default reference type would be imbd with value " http://www.imdb.com/title/tt${inetref}/"

BigB.

[Raymond Wagner <raymond@…>]

Updates for changeset [21188].

[robertm]

One way or another we'll need to handle this eventually. Switching to TMDB numbers makes sense as it makes *every* film at TMDB return a result from the API, rather than just those where the user has entered a buried details screen and added an IMDB number to the page at TMDB. This may be handled by Anduin's upcoming grabber reorganization/API rewrite, though.

[robertm]

(In [22865]) Initial commit for tmdb.py, a new TMDB Grabber script for TheMovieDB.

How does tmdb.py differ from tmdb.pl? I'm glad you asked!

tmdb.py uses the new TMDB v2.1 API, which is being expanded with all sorts of new information. When MPAA ratings/Parental rating is added to this new API, we will be able to add them simply and quickly, along with tons of other new info the API will be exposing.

tmdb.py takes IMDB *or* TMDB numbers as an argument, and spits out the TMDB number in the results, which means running a grab with the new grabber will seamlessly move an existing record over to TMDB numbers.

tmdb.py's search takes into account *all* films at TMDB, not just those with an IMDB number set.

tmdb.py is an alpha script. I would prefer it if people didn't use it with .22 as a) it's unsupported and b) .22 doesn't include some key bits of grabber logic to manage the IMDB->TMDB transition, and c) .23 is out in March, take a deep breath. It will work, I would just prefer that I (and Doug, and others on the lists and IRC channels) not have to provide support for an in progress script.

Like the famous Jamu and tvdb.py scripts, tmdb.py is written by Doug Vaughan, whose dedication to supporting his scripts and their users is always really, really appreciated.

Closes #6467.