Added support for Jellyfin

personal_data/fetchers/jellyfin_watch_history.py

@@ -0,0 +1,73 @@
+import dataclasses
+import datetime
+import logging
+import re
+import bs4
+from typing import Any
+from collections.abc import Iterator
+from jellyfin_apiclient_python import JellyfinClient
+
+from ..data import DeduplicateMode, Scraper
+from .. import secrets, parse_util, html_util, _version
+
+logger = logging.getLogger(__name__)
+
+URL_SITE_ROOT = 'https://steamcommunity.com/'
+
+URL_GAME_ACHIVEMENTS = URL_SITE_ROOT+'id/{username}/stats/appid/{appid}'
+
+FORMAT_DATE_HEADER = '%d/%m/%YYYY'
+
+def iterate_series(client):
+    result = client.jellyfin.user_items(params = {
+        'includeItemTypes': 'Series',
+        'parentId': 'a656b907eb3a73532e40e44b968d0225',
+        'userId': 'dd95c1085c1b4e83ba8e8853fbc644ab',
+    })
+    yield from result['Items']
+
+def iterate_watched_episodes_of_series(client, series_id: str):
+    result = client.jellyfin.user_items(params = {
+        'filters': 'IsPlayed',
+        'recursive': True,
+        'includeItemTypes': 'Episode',
+        'parentId': series_id,
+        'userId': 'dd95c1085c1b4e83ba8e8853fbc644ab',
+        'fields': 'AirTime',
+    })
+    yield from result['Items']
+
+@dataclasses.dataclass(frozen=True)
+class JellyfinWatchHistoryScraper(Scraper):
+    dataset_name = 'show_episodes_watched'
+    deduplicate_mode = DeduplicateMode.BY_ALL_COLUMNS
+
+    def scrape(self) -> Iterator[dict[str, Any]]:
+        client = JellyfinClient()
+
+        client.config.app('personal_data', _version.__version__,
+                          'test_machine', 'unique_id_1')
+
+        client.config.data["auth.ssl"] = False
+        client.auth.connect_to_address(secrets.JELLYFIN_URL)
+        client.auth.login(secrets.JELLYFIN_URL, secrets.JELLYFIN_USERNAME, secrets.JELLYFIN_PASSWORD)
+
+        for series_data in iterate_series(client):
+            series_id = series_data['Id']
+            for episode_data in iterate_watched_episodes_of_series(client, series_id):
+                episode_index = episode_data.get('IndexNumber')
+                if episode_index is None:
+                    continue
+                yield {
+                    'series.name': episode_data['SeriesName'],
+                    'season.name': episode_data['SeasonName'],
+                    'episode.index': int(episode_index),
+                    'episode.name': episode_data['Name'],
+                    'me.last_played_time': episode_data['UserData']['LastPlayedDate'],
+                    'episode.duration_seconds': episode_data['RunTimeTicks'] / 10000000,
+                    'episode.premiere_date': episode_data.get('PremiereDate'),
+                }
+
+                del episode_data
+            del series_data, series_id
+

personal_data/main.py

@@ -124,7 +124,7 @@ def main(
             OUTPUT_PATH / f'{scraper.dataset_name}.csv',
             result_rows,
             deduplicate_mode=scraper.deduplicate_mode,
-            deduplicate_ignore_columns=scraper.deduplicate_ignore_columns,
+            deduplicate_ignore_columns=scraper.deduplicate_ignore_columns(),
         )
         logger.info('Scraper done: %s', scraper.dataset_name)
 

personal_data/secrets.py

@@ -23,7 +23,7 @@ KUCOIN_KEY = load_secret('KUCOIN_KEY')
 KUCOIN_SECRET = load_secret('KUCOIN_SECRET')
 KUCOIN_PASS = load_secret('KUCOIN_PASS')
 
-# KRAKEN
+# Kraken
 KRAKEN_KEY = load_secret('KRAKEN_KEY')
 KRAKEN_SECRET = load_secret('KRAKEN_SECRET')
 
@@ -35,3 +35,8 @@ HOME_ASSISTANT_LLAK = load_secret('HOME_ASSISTANT_LLAK')
 MAILGUN_API_KEY = load_secret('MAILGUN_API_KEY')
 MAILGUN_DOMAIN = load_secret('MAILGUN_DOMAIN')
 MAILGUN_RECIPIENT = load_secret('MAILGUN_RECIPIENT')
+
+# Jellyfin
+JELLYFIN_URL = load_secret('JELLYFIN_URL')
+JELLYFIN_USERNAME = load_secret('JELLYFIN_USERNAME')
+JELLYFIN_PASSWORD = load_secret('JELLYFIN_PASSWORD')

personal_data/util.py

@@ -89,6 +89,7 @@ def deduplicate_by_ignoring_certain_fields(
 
     Output order is stable.
     """
+
     to_remove = set()
     for idx1, first in enumerate(dicts):
         for idx2, second in enumerate(dicts[idx1 + 1 :], idx1 + 1):
@@ -163,7 +164,7 @@ def load_csv_file(csv_file: Path) -> list[frozendict]:
 
 def extend_csv_file(
     csv_file: Path,
-    new_dicts: list[dict],
+    new_dicts: list[dict[str,typing.Any]],
     deduplicate_mode: data.DeduplicateMode,
     deduplicate_ignore_columns: list[str],
 ) -> dict: