Search class

[duck7000]

Post your findings/issues here

[duck7000]

@GeorgeFive
You probably using externalIdSearch() because you use the person class?

I presume that this is working as it should?
There are more external id searches possible like facebook, twitter or tiktok but i didn't include them as probably no one will use that?

here are the other choices:
https://developer.themoviedb.org/reference/find-by-id

Do you use textSearch() as well?
if so what are your findings about that?

[duck7000]

I added includeAdult parameter to textSearch()
Use true or false to in or exclude adult results

[duck7000]

@GeorgeFive

I changed externalIdSearch()

• renamed method name to imdbToTmdb()
• output is now int or false, false is in this case better as an empty array

Wiki page is updated

[mosource21]

@duck7000 is it possible to send $results = $tmdb->imdbToTmdb($imdbId); requests through the cache system and create something like tt0000000_find.json?

The imdbToTmdb() is "static" data much like movie, tv and person so it would benefit from the caching and save many repeated API calls.

Thanks

[duck7000]

Well it is possible but i assumed that caching would not be necessary, just as textSearch() as it would create a lot of cache files.
As for imdbToTmdb() what if the imdbId is not found and thus returns false, store a cache file with value false?
And what if that imdb is at this time not found but inside the cacheExpire time is? It will remain false until cacheExpire time runs out and this method is called again.

I'll think about it

[mosource21]

send $results = $tmdb->imdbToTmdb($imdbId); requests through the cache system and create something like tt0000000_find.json

A few more notes to hopefully put the case forward that it should be cached.

Well it is possible but i assumed that caching would not be necessary, just as textSearch() as it would create a lot of cache files.

That is fine just like if movie, tv or person is cached that also creates lots of files. Also remember the cache doesn't need to be a file it could be being stored in a database etc.

As for imdbToTmdb() what if the imdbId is not found and thus returns false, store a cache file with value false?

Yes if it is not found the cache should store whatever the API returned. It should work just like the other data that is cached.

And what if that imdb is at this time not found but inside the cacheExpire time is? It will remain false until cacheExpire time runs out

Yes that is fine just like if a movie, tv or person is updated on the site it will not update the cache until it is expired. This is the expected behaviour.

I'll think about it

In my mind the imdbToTmdb is just like the movie, tv and person data albeit it only returns a very small amount of data but should be treated in a similar way. The fact that all easily fit the same structure of an identifier_category.json type further supports this in my opinion.

Thanks

[duck7000]

Well thank you for your detailed explanation!
I am convinced now that this would be an idea although i not agree with your statement that cache doesn't need to be a file?
The cache system is based on cache files, i don't see how this works with a database? please explain.

I will add this feature

[duck7000]

This is added in latest git commit.
Would you test this please?

[mosource21]

i not agree with your statement that cache doesn't need to be a file? The cache system is based on cache files, i don't see how this works with a database? please explain.

Sorry wasn't very clear here - yes the default is cache files but the beauty of the Psr/SimpleCache/CacheInterface.php class is you can easily write your own so it can store in a database if you want. You then send that cache interface into all the classes instead of the default one and you have cache stored in a database instead of files.

This is added in latest git commit.
Would you test this please?

Yes will do but getting very late today so it will more than likely be tomorrow now.

Thanks

[mosource21]

end $results = $tmdb->imdbToTmdb($imdbId); requests through the cache system and create something like tt0000000_find.json

This is added in latest git commit. Would you test this please?

Working great 👍

Thanks

[duck7000]

Nice! thanks for testing, that is the only way to make it better for everyone.