This proof of concept (POC) is based on RediSearch and written in Python. Not much energy has been put into linting or PEP-8, nor is the code optimized in any way. It's actually quite messy.
After downloading the repository, simply run the following commands:
> docker compose up --build --detach
> docker exec -it search-app python ingest.py
The docker exec... command is only required the first time building the service (or after running docker compose down) hence can be left out.
When installed, the simple search API is available at POST http://localhost:8800/search.
In order to perform a search include the query parameter query, example:
curl --location --request POST 'http://localhost:8800/search?query=the%20matrix'
- Why POST when using query parameters? No particular reason. I guess I was thinking of using json payloads originally... It's a POC, you'll live ;-)
- Why port
8800? Because port 8000 was already used in my environment.
The resulting response could be something like this
Query: "the matrix"
{
"movies": [
{
"_popularity": 81.263,
"_score": 1.0,
"adult": "0",
"id": "movie:603",
"original_title": "The Matrix",
"payload": null,
"popularity": "81.263",
"type": "movie",
"video": "0"
},
{
"_popularity": 1.1915777777777778,
"_score": 0.9555555555555556,
"adult": "0",
"id": "movie:509944",
"original_title": "The Matwix",
"payload": null,
"popularity": "1.247",
"type": "movie",
"video": "0"
},
{
"_popularity": 36.787882352941175,
"_score": 0.9058823529411765,
"adult": "0",
"id": "movie:604",
"original_title": "The Matrix Reloaded",
"payload": null,
"popularity": "40.61",
"type": "movie",
"video": "0"
},
{
"_popularity": 6.4998000000000005,
"_score": 0.9,
"adult": "0",
"id": "movie:14543",
"original_title": "The Matrix Revisited",
"payload": null,
"popularity": "7.222",
"type": "movie",
"video": "0"
},
{
"_popularity": 27.42802,
"_score": 0.89,
"adult": "0",
"id": "movie:605",
"original_title": "The Matrix Revolutions",
"payload": null,
"popularity": "30.818",
"type": "movie",
"video": "0"
},
{
"_popularity": 3.0459714285714288,
"_score": 0.8857142857142858,
"adult": "0",
"id": "movie:221495",
"original_title": "The Matrix Recalibrated",
"payload": null,
"popularity": "3.439",
"type": "movie",
"video": "1"
},
{
"_popularity": 62.559709090909095,
"_score": 0.8818181818181818,
"adult": "0",
"id": "movie:624860",
"original_title": "The Matrix Resurrections",
"payload": null,
"popularity": "70.944",
"type": "movie",
"video": "0"
},
{
"_popularity": 2.32925,
"_score": 0.875,
"adult": "0",
"id": "movie:684731",
"original_title": "The Matrix Reloaded: Pre-Load",
"payload": null,
"popularity": "2.662",
"type": "movie",
"video": "0"
},
{
"_popularity": 1.170224,
"_score": 0.872,
"adult": "0",
"id": "movie:684428",
"original_title": "The Matrix: What Is Bullet-Time?",
"payload": null,
"popularity": "1.342",
"type": "movie",
"video": "0"
},
{
"_popularity": 1.680344,
"_score": 0.872,
"adult": "0",
"id": "movie:684729",
"original_title": "The Matrix Reloaded: Car Chase",
"payload": null,
"popularity": "1.927",
"type": "movie",
"video": "0"
}
],
"persons": [
{
"_popularity": 0.526984126984127,
"_score": 0.8783068783068783,
"adult": "0",
"id": "person:2540910",
"name": "Mattrix",
"payload": null,
"popularity": "0.6",
"type": "person"
},
{
"_popularity": 0.4955555555555556,
"_score": 0.825925925925926,
"adult": "0",
"id": "person:2368741",
"name": "John Matrix",
"payload": null,
"popularity": "0.6",
"type": "person"
},
{
"_popularity": 0.4828282828282829,
"_score": 0.8047138047138048,
"adult": "0",
"id": "person:1351398",
"name": "Jenny Matrix",
"payload": null,
"popularity": "0.6",
"type": "person"
},
{
"_popularity": 0.5502629629629631,
"_score": 0.7783068783068784,
"adult": "0",
"id": "person:1283932",
"name": "Jeff Hatrix",
"payload": null,
"popularity": "0.707",
"type": "person"
},
{
"_popularity": 0.45425685425685425,
"_score": 0.757094757094757,
"adult": "0",
"id": "person:3875331",
"name": "Mitch Mattix",
"payload": null,
"popularity": "0.6",
"type": "person"
},
{
"_popularity": 0.43333333333333335,
"_score": 0.7222222222222222,
"adult": "0",
"id": "person:983872",
"name": "Mike Marix",
"payload": null,
"popularity": "0.6",
"type": "person"
},
{
"_popularity": 0.4166666666666666,
"_score": 0.6944444444444443,
"adult": "0",
"id": "person:2314977",
"name": "Mihai Mitrix Mitrică",
"payload": null,
"popularity": "0.6",
"type": "person"
},
{
"_popularity": 0.4150793650793651,
"_score": 0.6917989417989419,
"adult": "0",
"id": "person:3455896",
"name": "Mirose Patrix",
"payload": null,
"popularity": "0.6",
"type": "person"
},
{
"_popularity": 0.4061050061050061,
"_score": 0.6768416768416768,
"adult": "0",
"id": "person:2850346",
"name": "Lorenzo Mataix",
"payload": null,
"popularity": "0.6",
"type": "person"
},
{
"_popularity": 1.2787264957264957,
"_score": 0.6680911680911681,
"adult": "0",
"id": "person:2799682",
"name": "Maetrix Fitten",
"payload": null,
"popularity": "1.914",
"type": "person"
}
],
"series": [
{
"_popularity": 15.447911111111111,
"_score": 0.9037037037037037,
"id": "series:711",
"original_name": "Threat Matrix",
"payload": null,
"popularity": "17.094",
"type": "series"
},
{
"_popularity": 10.351666666666667,
"_score": 0.8333333333333334,
"id": "series:11393",
"original_name": "Matrix",
"payload": null,
"popularity": "12.422",
"type": "series"
},
{
"_popularity": 0.545,
"_score": 0.8333333333333334,
"id": "series:31636",
"original_name": "Matrix",
"payload": null,
"popularity": "0.654",
"type": "series"
},
{
"_popularity": 1.1801703703703703,
"_score": 0.7925925925925926,
"id": "series:116501",
"original_name": "Escape The Matrix",
"payload": null,
"popularity": "1.489",
"type": "series"
},
{
"_popularity": 0.6311218855218856,
"_score": 0.6033670033670034,
"id": "series:220827",
"original_name": "Rua da Matriz",
"payload": null,
"popularity": "1.046",
"type": "series"
},
{
"_popularity": 0.38526984126984126,
"_score": 0.5873015873015873,
"id": "series:38368",
"original_name": "Approval Matrix",
"payload": null,
"popularity": "0.656",
"type": "series"
},
{
"_popularity": 0.4930718954248366,
"_score": 0.586990351696234,
"id": "series:68447",
"original_name": "Matrix Chiambretti",
"payload": null,
"popularity": "0.84",
"type": "series"
},
{
"_popularity": 0.37241666666666656,
"_score": 0.5694444444444443,
"id": "series:18401",
"original_name": "Paranormal Matrix",
"payload": null,
"popularity": "0.654",
"type": "series"
}
]
}
Usage: ingest.py [OPTIONS]
Options:
--records / --no-records ingest TMDb records
--index / --no-index create Redis searchable indexes
--help Show this message and exit.
By default ingest.py will ingest the TMDb records and create Redis indexes. This should not harm anything, just
overwrite any keys that already exist. Error handling ensures that attempts to create indexes that already exist will be
handled somewhat gracefully.
This application uses TMDB data but is not endorsed, certified, or otherwise approved by TMDB.
The exports are in English only, so this POC generally only supports English searches for international titles. Local titles should be available in the local language though, eg. searching for "adams æbler" or "Das Leben der Anderen" will still return a valid results.