Solved Scraping: Number of Actors ...
#1
would it be possible to configure the number of actors scraped?
I am spending quite some time to delete the ones above let's say the first five ...
normally I would not mind, but in Kodi you can sort movies according to actors which gives me thousands and thousands of actors I do not care about;

Peter
#2
mhm.. yes.. might be an interesting feature.
I'll put it on our list...
Sadly, no internet site has something like "main actors"...
tinyMediaManager - THE media manager of your choice :)
Wanna help translate TMM ?
Image
#3
What about not scraping if no picture is available?
#4
(2016-07-02, 14:05)Edworld Wrote: What about not scraping if no picture is available?

good idea: would already reduce number considerably ...

(2016-06-27, 21:55)myron Wrote: ...
Sadly, no internet site has something like "main actors"...

don't know how others do but themoviedb.org sorts them according to "importance" - so main actors first!
i.e. I could easily live with a mere number limit ... the first 3 or 5 would be more than enough for me ...
#5
(2016-07-02, 14:05)Edworld Wrote: What about not scraping if no picture is available?

I do not think is is a good idea.
a) There are movies with main actors which do not have any picture - so they won't get scraped.
b) Not all scrapers have that much actor images as imdb/tmdb (maybe there are scrapers out there, which do not provide any actor images..)

we will think about an option to limit the amount of actors scraped, but a decision by picture is too dangerous
tinyMediaManager - THE media manager of your choice - available for Windows, macOS and Linux
Help us translate tinyMediaManager at Weblate | Translations at 66%
Found a bug or want to submit a feature request? Contact us at GitLab
#6
I'm a long-time Ember user who is trying out TMM, and so far I'm impressed.

EMM has two different ways to filter cast: (1) scrape only the first xx actors, and/or (2) scrape only actors with images.

I realize that neither filter is perfect: sometimes IMDB lists the cast in order of appearance so a main actor may be 20th or lower on the list, and as you point out, not all major actors actually have images. But it's too unwieldy to have a cast database filled with thousands who have only had 1-2 bit parts in movies I own, so this is a risk worth taking, if you could emulate the EMM filters.

My ideal actor DB would include only actors who appear in, say, 3+ movies which I own. As I was writing this, I just had a thought on a possible filter, but it may be too cumbersome.

1. When scraping an actor, check how many other movies in our TMM list actor appears in.
2. If under xx movies, the actor's data is put in a comment in the NFO, rather than the <actor> tag.
3. Have an option for the user to have TMM manually scan all NFOs, and if an actor's name now appears in xxx or more movies, all NFOs where he/she was in the comment are changed to the <actor> tag.
4. Generate a list of movies thus modified so we can then refresh them in Kodi.

Probably way too much work, but it would create the ideal cast list.

Carl.
CoreELEC 9.03
Aeon Nox 5: SiLVO
Minix Neo U9-H, Logitech Harmony Elite
#7
Another option would be that you can give kodi the option to display only actors who appears 2, 3, 4 etc times in your library. Like a filter.
#8
Something definitely should be done to limit the number of actors scraped. For one movie I get like 60-70 actors which is ridiculous.
#9
At least an option to not scrape "uncredited" cast members would help a lot.
#10
(2020-09-12, 18:40)puenktchen Wrote: At least an option to not scrape "uncredited" cast members would help a lot.

That's great idea!
Another option would be using IMDb dataset instead that contains only principal cast. But it only lists 4 to 5 actors for each movie, which some users may find too few.

e.g.

Around the World in 80 Days (1956) 

total actors: 1301
credited actors: 50
IMDb dataset: 4

The Ten Commandments (1956)

total actors: 371
credited actors: 82
IMDb dataset: 4
#11
Well, that should be doable, because scraping _all_ actors was just an extra step. Scraping just the credited ones should be easier..

as you see, it is rather impossible to make it right for all users without a ton extra settings Sad
tinyMediaManager - THE media manager of your choice - available for Windows, macOS and Linux
Help us translate tinyMediaManager at Weblate | Translations at 66%
Found a bug or want to submit a feature request? Contact us at GitLab
#12
Thanks for implementing it!
With v4, I re-scraped actors for all movies in my collection, expecting it will reduce database size and initial launching time significantly.
It did. It almost halved both the db size and launching time.
But it didn't happen immediately. It actually slightly increased the db size (about 10%). So I tried removing and re-importing all items and now the size got halved.
Why do you think it requires re-creating the db to reduce the size?
#13
the database engine has an internal mechanic to re-arrange/compress/defragment itself.. sometimes it just takes some time to do that work.
With removing/re-adding you gave the database a huge change so it immediately did its work - maybe this is why you saw the change there faster
tinyMediaManager - THE media manager of your choice - available for Windows, macOS and Linux
Help us translate tinyMediaManager at Weblate | Translations at 66%
Found a bug or want to submit a feature request? Contact us at GitLab
#14
(2020-09-23, 17:42)mlaggner Wrote: the database engine has an internal mechanic to re-arrange/compress/defragment itself.. sometimes it just takes some time to do that work.
With removing/re-adding you gave the database a huge change so it immediately did its work - maybe this is why you saw the change there faster

You mean like running Update Sources command instead of re-creating the db?
Or can it get smaller without anything being done?
#15
just had a look at the code - the compacting is done when the database shuts down (tmm shuts down). So it should be "self cleaning/shrinking" when tmm finally shuts down
tinyMediaManager - THE media manager of your choice - available for Windows, macOS and Linux
Help us translate tinyMediaManager at Weblate | Translations at 66%
Found a bug or want to submit a feature request? Contact us at GitLab

Logout Mark Read Team Forum Stats Members Help
Scraping: Number of Actors ...0