AutoBrr List that contains "New on Netflix" from three genres (true crime, documentary films, documentary series)

netflix.py

#!/bin/python

import json
from typing import List, Set

import requests
from bs4 import BeautifulSoup

TRUE_CRIME_URL = "https://www.netflix.com/browse/genre/108820"
DOCUMENTARY_URL = "https://www.netflix.com/browse/genre/2243108"
DOCUMENTARY_SERIES_URL = "https://www.netflix.com/browse/genre/10105"

# Stable section title in the UI (not content-dependent).
SECTION_TITLE = "New on Netflix"

# Button/label text we must not treat as titles when scraping the row.
_NON_TITLE_TEXT: Set[str] = frozenset(
    {"Join now", "Play", "More Info", "Add to My List", "Next", "Previous"}
)


def _titles_from_html(soup: BeautifulSoup) -> List[str]:
    """Extract titles from the 'New on Netflix' section using the section heading only."""
    for h2 in soup.find_all("h2"):
        if h2.get_text(strip=True) != SECTION_TITLE:
            continue
        # Row can be: next sibling of h2, or next sibling of h2's parent (current Netflix layout).
        row = h2.find_next_sibling()
        if not row:
            parent = h2.find_parent()
            if parent is not None:
                row = parent.find_next_sibling()
        if not row:
            continue
        titles = [
            p.get_text(strip=True)
            for p in row.find_all("p")
            if (p.get_text(strip=True) and p.get_text(strip=True) not in _NON_TITLE_TEXT)
        ]
        if titles:
            return titles
    return []


def _titles_from_json_ld(soup: BeautifulSoup) -> List[str]:
    """Use JSON-LD only when an ItemList is explicitly named 'New on Netflix'."""
    for script in soup.find_all("script", type="application/ld+json"):
        try:
            data = json.loads(script.string)
        except (TypeError, json.JSONDecodeError):
            continue
        if data.get("@type") != "ItemList" or data.get("name") != SECTION_TITLE:
            continue
        elements = data.get("itemListElement") or []
        return [
            e["item"]["name"]
            for e in elements
            if isinstance(e.get("item"), dict) and e.get("item", {}).get("name")
        ]
    return []


def get_new_on_netflix_titles(html: str) -> List[str]:
    soup = BeautifulSoup(html, "html.parser")

    # 1) Prefer HTML: section identified by the stable heading "New on Netflix".
    titles = _titles_from_html(soup)
    if titles:
        return titles

    # 2) Fallback: JSON-LD ItemList named "New on Netflix" (if Netflix adds it).
    return _titles_from_json_ld(soup)


def main() -> None:
    headers = {"User-Agent": "Mozilla/5.0 (X11; Linux x86_64) Chrome/141.0.7150.0"}
    all_titles: List[str] = []

    for url in (TRUE_CRIME_URL, DOCUMENTARY_URL, DOCUMENTARY_SERIES_URL):
        response = requests.get(url, headers=headers)
        response.raise_for_status()
        all_titles.extend(get_new_on_netflix_titles(response.text))

    for title in sorted(set(all_titles), key=str.lower):
        print(title.lower())

if __name__ == "__main__":
    main()

netflix.txt

a child of my own
a toxic love story
america's sweethearts: dallas cowboys cheerleaders
big chicken: a fast food conspiracy
bãhubali the torch bearer
chompoo: lost & forgotten
chris & martina: the final set
conversations with a killer: the charles manson tapes
death of the pastor's wife
elize: shadows of a woman
famous last words: gloria steinem
fito páez: the world within a song
freefall: a reckoning for boeing
future council
inside the trustor scandal
instadocs: the decoy plane
instadocs: the prediction games
miguel ángel blanco: the 48 hours that changed spain
mourinho
potential “tadayoshi ohkura”
quarterback
ride
shipwrecked: nightmare at sea
teenage wasteland
the american experiment
the idaho murders: college nightmare
the knowledge keepers
the last salute
the root of the game
tony armstrong's extra-ordinary things
true south
turning point: generation 9/11
untold mr. t: i pity the fool
untold raygun: breaking badly
untold: the testimony of vince young
world series film
worst neighbor ever
wwe: unreal

⚠️ This list gets automatically updated daily ⚠️

How to use this in AutoBrr?

1. Go to your https://<AUTOBRR_IP>:<AUTOBRR_PORT>/settings/lists
2. Add new list, enter the name
3. Type: Plaintext
4. List URL: https://gist.githubusercontent.com/luckylittle/58363570af7c432ad3e429bcd4b2155f/raw/netflix.txt
5. Match Release: OFF
6. Bypass the clanup and sanitization and use the list as-is: OFF
7. Filters: [Select your filter here]