secstate/vrobbler
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[videos] Merge TMDB + OMDB metadata for reliable run time and ratings

Browse Source
This commit is contained in:
Colin Powell
2026-08-25 19:31:27 -04:00
parent 79ee294e71
commit 9140e0cb95
5 changed files with 192 additions and 39 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
PROJECT.org
View File

@ -18,7 +18,7 @@ tasks, Todoist tasks, web pages I've read and trails I've hiked has turned out
to be sometimes cathartic and sometimes functional as I try to remember when I
did a thing.
* Backlog [0/30] :vrobbler:project:personal:
* Backlog [0/31] :vrobbler:project:personal:
** TODO [#C] Configure IMAP folder/start in user profile :imap:settings:
*** Description
@ -590,4 +590,8 @@ The Edit log form should have from top to bottom:
- People (which should be similar to the Bird widget on BirdLocation and allow setting per user score, win true/false, rank, new true/false, seat_ordrer)
- Expansion ids (which should a multi-select widget of expansions for this game)
- Location (which should be a drop down of BoardGameLocations for this user)
** STRT [#B] Video metadata is not getting base run time seconds reliably :videos:metadata:
:PROPERTIES:
:ID: cdb7cf49-a9df-3368-7cf8-12ee0775a3ce
:END:

91
vrobbler/apps/videos/management/commands/cleanup_video_metadata.py Normal file
View File

@ -0,0 +1,91 @@
import logging
from django.core.management.base import BaseCommand
from django.db import models, transaction
logger = logging.getLogger(__name__)
BOGUS_RUN_TIME_SECONDS = 900
class Command(BaseCommand):
help = "Enrich video metadata (run time, ratings) from TMDB/OMDB APIs"
def add_arguments(self, parser):
parser.add_argument(
"--force",
action="store_true",
help="Overwrite existing cover image",
)
parser.add_argument(
"--dry-run",
action="store_true",
help="Show what would be done without making changes",
)
parser.add_argument(
"--imdb-id",
type=str,
help="Only process videos with this imdb_id",
)
parser.add_argument(
"--all",
action="store_true",
help="Process every video with an imdb_id, not just ones missing metadata",
)
def handle(self, *args, **options):
from videos.models import Video
force = options["force"]
dry_run = options["dry_run"]
imdb_id = options["imdb_id"]
process_all = options["all"]
qs = Video.objects.filter(imdb_id__isnull=False).exclude(imdb_id="")
if imdb_id:
qs = qs.filter(imdb_id=imdb_id)
if not process_all:
qs = qs.filter(
models.Q(base_run_time_seconds__isnull=True)
| models.Q(base_run_time_seconds=BOGUS_RUN_TIME_SECONDS)
| models.Q(imdb_rating__isnull=True)
| models.Q(tmdb_rating__isnull=True)
)
videos = list(qs)
total = len(videos)
self.stdout.write(f"Total videos to process: {total}")
if dry_run:
for video in videos:
label = video.title or str(video.uuid)
runtime = video.base_run_time_seconds
if runtime == BOGUS_RUN_TIME_SECONDS:
runtime = "900 (bogus)"
status = (
f"runtime={runtime}, imdb_rating={video.imdb_rating}, "
f"tmdb_rating={video.tmdb_rating}"
)
self.stdout.write(f" [DRY RUN] Would fix {label} ({status})")
return
updated = 0
errors = 0
for video in videos:
label = video.title or str(video.uuid)
try:
with transaction.atomic():
video.fix_metadata(force_update=force)
updated += 1
self.stdout.write(f" [{updated}/{total}] {label}")
except Exception as e:
errors += 1
self.stdout.write(
self.style.ERROR(
f" Error updating {label} (imdb_id={video.imdb_id}): {e}"
)
)
self.stdout.write(
self.style.SUCCESS(f"\nDone! {updated} videos updated, {errors} errors")
)

60
vrobbler/apps/videos/metadata.py
View File

@ -34,9 +34,7 @@ class VideoMetadata:
title: str
video_type: VideoType = VideoType.UNKNOWN
base_run_time_seconds: int = (
60 # Silly default, but things break if this is 0 or null
)
base_run_time_seconds: Optional[int] = None
imdb_id: Optional[str]
tmdb_id: Optional[str]
youtube_id: Optional[str]
@ -69,7 +67,7 @@ class VideoMetadata:
imdb_id: Optional[str] = "",
youtube_id: Optional[str] = "",
twitch_id: Optional[str] = "",
base_run_time_seconds: int = 900,
base_run_time_seconds: Optional[int] = None,
):
self.title = ""
self.imdb_id = imdb_id
@ -87,3 +85,57 @@ class VideoMetadata:
if "tv_series_imdb_id" in video_dict.keys():
series_id = video_dict.pop("tv_series_imdb_id")
return video_dict, series_id, cover, genres
MERGE_FIELDS = (
"title",
"video_type",
"base_run_time_seconds",
"imdb_id",
"tmdb_id",
"youtube_id",
"twitch_id",
"episode_number",
"season_number",
"next_imdb_id",
"year",
"plot",
"overview",
"imdb_rating",
"tmdb_rating",
"cover_url",
"tv_series_imdb_id",
)
def merge_metadata(
primary: Optional[VideoMetadata],
secondary: Optional[VideoMetadata],
) -> Optional[VideoMetadata]:
"""Merge two metadata lookups, filling gaps in ``primary`` from ``secondary``.
``primary`` (typically the TMDB result) wins; ``secondary`` (typically the
OMDB result) only fills fields the primary source did not set.
"""
if not primary:
return secondary
if not secondary:
return primary
for field in MERGE_FIELDS:
current = getattr(primary, field, None)
if field == "video_type":
if current != VideoType.UNKNOWN:
continue
elif current not in (None, ""):
continue
value = getattr(secondary, field, None)
if value in (None, ""):
continue
setattr(primary, field, value)
secondary_genres = getattr(secondary, "genres", None)
if not getattr(primary, "genres", None) and secondary_genres:
primary.genres = secondary_genres
return primary

60
vrobbler/apps/videos/models.py
View File

@ -21,7 +21,7 @@ from scrobbles.mixins import (
ScrobblableMixin,
)
from taggit.managers import TaggableManager
from videos.metadata import VideoMetadata
from videos.metadata import VideoMetadata, merge_metadata
from videos.sources.omdb import lookup_video_from_omdb
from videos.sources.tmdb import lookup_video_from_tmdb
from videos.sources.youtube import lookup_video_from_youtube
@ -37,6 +37,17 @@ logger = logging.getLogger(__name__)
BNULL = {"blank": True, "null": True}
def lookup_merged_video_metadata(imdb_id: str) -> Optional[VideoMetadata]:
"""Look a title up on both TMDB and OMDB, merging the results."""
try:
tmdb_metadata = lookup_video_from_tmdb(imdb_id)
except Exception as e:
logger.warning(f"TMDB lookup failed for {imdb_id}: {e}")
tmdb_metadata = None
omdb_metadata = lookup_video_from_omdb(imdb_id)
return merge_metadata(tmdb_metadata, omdb_metadata)
@dataclass
class VideoLogData(BaseLogData, WithPeopleLogData):
rating: Optional[int] = None
@ -341,22 +352,26 @@ class Series(TimeStampedModel):
logger.warning(f"No results on TMDB for {self.name}")
return
video_metadata = lookup_video_from_tmdb(self.imdb_id)
video_metadata = lookup_merged_video_metadata(self.imdb_id)
if not video_metadata or not video_metadata.title:
logger.warning(f"No metadata for {self}")
return
if video_metadata.cover_url and (not self.cover_image or force_update):
if getattr(video_metadata, "cover_url", None) and (
not self.cover_image or force_update
):
r = requests.get(video_metadata.cover_url)
if r.status_code == 200:
fname = f"{self.name}_{self.uuid}.jpg"
self.cover_image.save(fname, ContentFile(r.content), save=True)
self.plot = video_metadata.plot or ""
self.imdb_rating = getattr(video_metadata, "imdb_rating", None)
self.plot = getattr(video_metadata, "plot", None) or ""
self.imdb_rating = (
getattr(video_metadata, "imdb_rating", None) or self.imdb_rating
)
self.save()
if video_metadata.genres:
if getattr(video_metadata, "genres", None):
self.genre.add(*video_metadata.genres)
@classmethod
@ -367,14 +382,7 @@ class Series(TimeStampedModel):
logger.info("Series not created and overwrite=False, returning")
return series
try:
metadata = lookup_video_from_tmdb(imdb_id)
except Exception as e:
logger.warning(f"TMDB lookup failed for series {imdb_id}: {e}")
metadata = None
if not metadata or not metadata.title:
metadata = lookup_video_from_omdb(imdb_id)
metadata = lookup_merged_video_metadata(imdb_id)
if not metadata or not metadata.title:
logger.warning(f"No metadata found for series {imdb_id} from TMDB or OMDB")
@ -385,6 +393,8 @@ class Series(TimeStampedModel):
vdict["name"] = vdict.pop("title")
for k, v in vdict.items():
if v is None:
continue
setattr(series, k, v)
series.save()
@ -570,14 +580,7 @@ class Video(ScrobblableMixin):
return
if self.imdb_id:
try:
metadata = lookup_video_from_tmdb(self.imdb_id)
except Exception as e:
logger.warning(f"TMDB lookup failed for {self.imdb_id}: {e}")
metadata = None
if not metadata or not metadata.title:
metadata = lookup_video_from_omdb(self.imdb_id)
metadata = lookup_merged_video_metadata(self.imdb_id)
if not metadata or not metadata.title:
logger.warning(f"No metadata found for {self} from TMDB or OMDB")
@ -586,6 +589,8 @@ class Video(ScrobblableMixin):
vdict, series_id, cover, genres = metadata.as_dict_with_cover_and_genres()
for k, v in vdict.items():
if v is None:
continue
setattr(self, k, v)
if series_id:
@ -629,14 +634,7 @@ class Video(ScrobblableMixin):
if not created and not overwrite:
return video
try:
metadata = lookup_video_from_tmdb(imdb_id)
except Exception as e:
logger.warning(f"TMDB lookup failed for {imdb_id}: {e}")
metadata = None
if not metadata or not metadata.title:
metadata = lookup_video_from_omdb(imdb_id)
metadata = lookup_merged_video_metadata(imdb_id)
if not metadata or not metadata.title:
logger.warning(f"No metadata found for {imdb_id} from TMDB or OMDB")
@ -646,6 +644,8 @@ class Video(ScrobblableMixin):
if created or overwrite:
for k, v in vdict.items():
if v is None:
continue
setattr(video, k, v)
if series_id:

14
vrobbler/apps/videos/sources/tmdb.py
View File

@ -47,7 +47,9 @@ def lookup_video_from_tmdb(name_or_id: str, kind: str = "movie") -> VideoMetadat
video_metadata.year = pendulum.parse(media.release_date).year
video_metadata.genres = [g.get("name", "") for g in media.genres]
video_metadata.tmdb_id = media.id
video_metadata.base_run_time_seconds = media.runtime * 60
video_metadata.base_run_time_seconds = (
media.runtime * 60 if media.runtime else None
)
video_metadata.plot = media.overview
video_metadata.overview = media.overview
video_metadata.tmdb_rating = media.vote_average
@ -56,10 +58,10 @@ def lookup_video_from_tmdb(name_or_id: str, kind: str = "movie") -> VideoMetadat
media = TV().details(tmdb_result.tv_results[0].id)
video_metadata.video_type = VideoType.TV_EPISODE.value
video_metadata.title = media.name
video_metadata.cover_url = (
TMDB_IMAGE_URL + media.poster_path
video_metadata.cover_url = TMDB_IMAGE_URL + media.poster_path
video_metadata.year = (
pendulum.parse(media.first_air_date).year if media.first_air_date else None
)
video_metadata.year = pendulum.parse(media.first_air_date).year if media.first_air_date else None
video_metadata.genres = [g.get("name", "") for g in media.genres]
video_metadata.tmdb_id = media.id
video_metadata.base_run_time_seconds = (
@ -91,6 +93,10 @@ def lookup_video_from_tmdb(name_or_id: str, kind: str = "movie") -> VideoMetadat
video_metadata.tmdb_id = media.id
video_metadata.plot = media.overview
video_metadata.overview = media.overview
video_metadata.tmdb_rating = getattr(media, "vote_average", None)
runtime = getattr(media, "runtime", None)
if runtime:
video_metadata.base_run_time_seconds = runtime * 60
if not media:
logger.warning("Video not found on TMDB", extra={"imdb_id": imdb_id})