Importing new movies via API or http requests

[giovi321]

I read the API documentation and it seems like it is not possible to import a movie via API (maybe via tmdb ID). Is there any other way to do it? Did I understand the documentation correctly?
Many thanks 🙂

[leepeuker]

Hey, you are right, there is currently no way to add a new movie to Movary via the API (we want to add it).

Beside manually adding new movies via the UI and importing from third party services, you can add new movies via a CSV import on the Settings web UI. Go to Account -> Personal data (or visit the url /settings/account/data). Here you can import e.g. history.csv. The CSV format should be title,year,tmdbId,imdbId,watchedAt,comment,location. The tmdbId is used to identify the movie and match to an existing one or to create a new one.

[giovi321]

Many thanks for the explanation! Is there a way to import a csv via command line?

[giovi321]

Just one more idea popped up in my mind... what if I use the Jellyfin webhook to add new movies? Or the integration with Jellyfin only allows to mark movies as "watched" (i.e., to scrobble them, I believe)?

[leepeuker]

Many thanks for the explanation! Is there a way to import a csv via command line?

No currently this is only possible via UI.

[leepeuker]

Just one more idea popped up in my mind... what if I use the Jellyfin webhook to add new movies? Or the integration with Jellyfin only allows to mark movies as "watched" (i.e., to scrobble them, I believe)?

Hm, this should be possible. The Jellyfin webhook can add new movies, same for the other webhooks like Plex.

[giovi321]

As a temporary workaround while the API to add movies is implemented, I wrote this simple python script to add movies exploiting the jellyfin webhook.
You'll need to run first pip install guessit requests.

#!/usr/bin/env python3
import os
import time
import requests
from guessit import guessit
from datetime import datetime, timezone

# ─── CONFIGURATION ─────────────────────────────────────────────────────────────

# Folder to scan (recursively) for movie files, without trailing slash
DOWNLOAD_DIR = "/your/directory/without/trailing/slash"

# Your Movary Jellyfin webhook URL. Available at https://movary.example.com/settings/integrations/jellyfin
WEBHOOK_URL = (
    "https://movary.example.com"
    "/api/webhook/jellyfin/"
    "your_jellyfin_webhook_uuid"
)

# The username Movary should record
MOVARY_USER = "your_username"

# Your TMDB API key
TMDB_API_KEY = "your_tmdb_api_key"

# File extensions to treat as movies
EXTENSIONS = {".mp4", ".mkv", ".avi", ".mov", ".wmv", ".flv"}

# ─── END CONFIGURATION ─────────────────────────────────────────────────────────


def scan_movies(directory):
    """
    Recursively yield paths to all files under `directory`
    whose extension is in EXTENSIONS.
    """
    for root, _, files in os.walk(directory):
        for fn in files:
            if os.path.splitext(fn)[1].lower() in EXTENSIONS:
                yield os.path.join(root, fn)


def parse_filename(path):
    """
    Use guessit to pull title and year from the filename.
    Returns (title, year) where year may be None.
    """
    info = guessit(path, {"type": "movie"})
    return info.get("title"), info.get("year")


def search_tmdb(title, year=None):
    """
    Search TMDB for `title` (and optional `year`).
    Returns the first result dict or None.
    """
    url = "https://api.themoviedb.org/3/search/movie"
    params = {"api_key": TMDB_API_KEY, "query": title}
    if year:
        params["year"] = year
    resp = requests.get(url, params=params)
    resp.raise_for_status()
    results = resp.json().get("results", [])
    return results[0] if results else None


def build_payload(movie, tmdb_id):
    """
    Construct the JSON payload Movary expects from a Jellyfin webhook.
    Uses timestamps in "YYYY-MM-DDTHH:MM:SS" format (no suffix).
    """
    now_utc = datetime.now(timezone.utc).replace(microsecond=0)
    ts = now_utc.strftime("%Y-%m-%dT%H:%M:%S")

    release_year = movie.get("release_date", "")[:4]
    return {
        "Name": movie.get("title", ""),
        "ItemType": "Movie",
        "Provider_tmdb": str(tmdb_id),
        "Year": release_year,
        "NotificationType": "PlaybackStop",
        "PlayedToCompletion": True,
        "LastPlayedDate": ts,
        "NotificationUsername": MOVARY_USER,
        "Username": MOVARY_USER,
        "ServerName": "Script",
        "DeviceName": "Script",
        "ClientName": "Script",
        "UtcTimestamp": ts,
    }


def send_to_movary(payload):
    """
    POST the JSON payload to Movary’s Jellyfin webhook endpoint.
    """
    resp = requests.post(
        WEBHOOK_URL,
        json=payload,
        headers={"Content-Type": "application/json"},
    )
    resp.raise_for_status()
    return resp


def main():
    for path in scan_movies(DOWNLOAD_DIR):
        print(f"\nProcessing: {path}")
        title, year = parse_filename(path)
        if not title:
            print("  → guessit couldn’t parse a title, skipping.")
            continue

        print(f"  → Parsed title: '{title}'", end="")
        if year:
            print(f", year: {year}")
        else:
            print()

        movie = search_tmdb(title, year)
        if not movie:
            print("  → No match on TMDB, skipping.")
            continue

        tmdb_id = movie["id"]
        print(f"  → Found TMDB ID {tmdb_id} ({movie.get('title')})")

        payload = build_payload(movie, tmdb_id)
        try:
            resp = send_to_movary(payload)
            print(f"  → Successfully notified Movary (HTTP {resp.status_code})")
        except Exception as e:
            print("  → Error sending to Movary:", e)

        # avoid hammering APIs
        time.sleep(0.3)


if __name__ == "__main__":
    main()

[leepeuker]

I have created an API endpoint to add new movies at /api/movies/add, for more infos see the API docs. This is part the the release 0.65.0. Please report back if this works for your