IMDB TMDB

[part timer]

So I just went looking through one of my folders where I'm keeping my newly renamed movies and noticed something strange. I have 3 folders for the same movie. 2 of them I was not SHOCKED to see, but disappointed. I've seen this a few times now where it will add an extra space in one spot. I renamed some of my directories manually to correct it, but I don't know where the space is coming from. Maybe when when I've copy and pasted my formats back and forth (another reason, saving them is better I guess).

The other folder though I'm guessing is IMDB and TMDB having a difference of opinion on what to call the movie. I have clicked on tmdb and imdb to match at different times, because I trust imdb more but it seems to take longer, so I've gone back and forth. I guess this would be a reason not to do that, I had no idea they might have slightly different titles. Sheesh, no matter switching I guess.

Here's what I ended up with:

Code: Select all

28 Days Later...  [2002]  [Rated  R]  [Voted 8.2]
28 Days Later  [2003]  [Rated  R]  [Voted 8.2]
28 Days Later  [2003]  [Rated R]  [Voted 8.2]

The one with the extra space between rated and r had only the srt file of the video that was in the folder that didn't have the extra space. Strange.

[rednoah]

The extra space is wierd. I also suspect that there were 2 spaces in the format. Data I get from TMDB should be fine, and I think I even trim spaces in my code automatically.

As for movie names that can't be helped, but you can do some cleanup in the format, e.g. {n.replaceAll(/[.]+$/), But that's all special cases that would add to your format over time.

Personally I prefer TMDB. By now equal to or better than IMDB for all my test cases. Also with TMDB it respects your language setting which the imdb page scraper can't.

[part timer]

Well I do find it seems to me that TMDB seems quicker to match to. The program I used to use, "therenamer" used imbd akas to get the english match. For example it added /releaseinfo#akas to the imdb link, so instead of "Ching Se" I got "Green Snake".

Check it out:
http://www.imdb.com/title/tt0106559
http://www.imdb.com/title/tt0106559/releaseinfo#akas

In the settings you set up your preferred language, when it's available.

[rednoah]

FileBot just parses the main title from that same page:
http://akas.imdb.com/title/tt0106559/releaseinfo

Not sure if I care enough about imdb to add more features when tmdb is better anyway.

[part timer]

We have a difference of opinion on these two websites I see. I love IMDB and you say TMDB is better. Matching to IMDB seemed to frustrate me and would get a connection reset fairly often after waiting so long (again, I wish at that point it would show me the ones it DID match in that time and let me rename them before going on to the rest that it didn't get to yet, instead of having to start over). TMDB always seemed to match faster to me, but sometimes the titles are slightly different, so now I try to stick to IMDB again instead.

One of the biggest things IMDB has over TMDB for my movie names though is the $rating. The movie Changing Lanes (imdb tt0264472) is a good example to me. I see this in my list and see it's voted 10/10 on the TMDB. Holy crap that must be a good movie to get a 10! I go look and 1 person voted. I go to imdb and after 38,271 people voted it gets a 6.5. I think the latter is a pretty good average and a better indication of the movie. Some movies show a vote of 0. Great movies, just no one voted on them. It makes it deceiving when scrolling through a list of folders and you don't know the movies.

In my mind IMDB $rating would be the way to go here

We had talked once about the votes($rating) in my folder names coming from TMDB instead of IMDB and you had said there is no API for IMDB to get that info. The program I used to use before filebot, therenamer apparently used an API to get its rating from IMDB. I think it may have been unofficial though and right now when I go to search that specific site to add the link I read it on I can't find the reference right now.

[rednoah]

IMDb only has one API with all the data, the website. Everything that uses IMDb is doing the same as you would do, fetch webpages and lookup the data from the html.

You'll notice that therenamer regularly breaks, that's whenever they change the website. I dropped IMDb episode list support last time they changed that part of the website 'cause I can't be bothered fixing things all the time, just waiting for it to break again at some point without any warning. Luckily they're lazy bastards and don't change things alot, but it's still shit work to maintain a scraper, so I keep mine as simple as possible.

I definitely won't bother with maintaining a scraper for all that extra data. Luckily some really nice guy is maintaining an imdb scraper and feeds that data into a nice and simple API:
http://www.imdbapi.com
http://www.imdbapi.com/?i=tt0379786&r=xml&tomatoes=true

I'll give you some bindings for all that data, but that API might not be very stable/fast, it's a free service with no business model after all.

[part timer]

That would be awesome, until it breaks that is .

Then also I'm going to be asking how I rename just my folders and leave my data alone for that too of course, to replace "[Voted x.y]" w/the imdb votes instead.

[rednoah]

With r1133 you can use {imdbapi} to get all that stuff from the xml, or specifically {imdbapi.rating} for imdb rating.

[part timer]

That would be awesome.
Except, like I said that means in the near future I'll be asking how I could rename the 1000+ folders I already renamed, swapping out the old tmdb rating for the new imdb rating

[part timer]

So how much would be involved in renaming folders only and leaving files alone inside that folder to take care of the imdb rating and possibly imdb ids added?

[rednoah]

That's really a special case, doesn't make much sense to add that to main application. Don't really have time to build an extra script for that lately though, but that's were special usecases would go.