TV Time Archive Viewer

A client-side web app that reads your TV Time GDPR data export, a .zip of CSV tables, and turns it into a browsable, searchable archive of the shows and movies you've watched, rated, and reacted to.

Everything runs in your browser. The export is parsed locally and never uploaded; it's kept in the browser's own storage (IndexedDB) so it reloads automatically on your next visit.

Not affiliated with TV Time / Whip Media.

Disclaimer

This is a reader for the data backup TV Time gives you, not a replacement for it. Some of it is best-effort guesswork (how reactions are encoded, how the data fits together across many tables). And some things aren't in the export at all, they live on TV Time's servers behind images and IDs (comment/badge/avatar images, the characters you voted for, your friends' names), so they only survive a shutdown if you capture them first: see Extended backup.

I have no intention of running a server to hold this data centrally or to fill in the gaps by crowdsourcing, managing other people's data is a can of worms I want no part in. The app is provided as-is, and whether anything more gets added to it is entirely at my discretion. You're of course free to fork it.

If you're looking for a replacement rather than an archive: I moved to Trakt (note that it has no social features); other options include Sofa Time and Refract, whose exports this app also reads (see Refract exports).

Get your export

TV Time's GDPR self-service has shut down; new exports can no longer be requested. If you downloaded an export while it was up, that .zip is all this app needs.

The export contains personal information (name, email, IP history, and more). Treat it as private, don't commit it to a public repository or host it anywhere.

Use it

Either open the hosted app at https://tvt.remls.io and drop your .zip on the page, or run it yourself.

The app loads a few libraries from a CDN, so serve it over HTTP rather than opening the file directly:

cd viewer
python3 -m http.server 8777
# open http://localhost:8777/

Drop the export .zip on the page (or tap to choose it). It's remembered between visits; "Change source .zip file" in the ⚙ menu forgets it and clears local storage.

Refract exports

A second viewer at https://tvt.remls.io/refract/ (or /refract/ on your own copy) reads Refract CSV exports. In the Refract app, go to Settings > Import & Export > Export as CSV and drop the resulting .zip on that page.

Refract exports carry less data than TV Time's, so the view set is smaller: Home, TV Shows, Anime, Movies, Watch history, Lists, Ratings, Reactions (mood tags), Reviews, and All data. Reviews render Refract's markup (bold, italics, inline spoilers, quotes, lists, links).

Notes on how the data is interpreted:

• The export has no database ids, so everything joins by title. Show posters and episode titles come from TVmaze by name search, guided by release year when a title exists twice in your library or when a fuzzy match looks wrong.
• Refract's database numbers some shows as one continuous season. When TVmaze splits the same show into seasons, watched episodes are remapped onto that split (the detail page says so when it happens).
• Both viewers share the TVmaze and Wikidata metadata caches. Each keeps its own copy of your export in browser storage; loading one never touches the other.

Views

Three nav tabs are umbrellas that group related views, Watch (Shows, Movies, Watch history, Lists), Ratings (Ratings, Reactions, Character votes), and Community (Comments, Notifications, Friends, Badges). On desktop their items nest in the sidebar; on mobile, tapping the tab opens a popup menu.

View	(Group)	What it shows
Home		Headline totals, most-watched shows, biggest marathons, and episodes/hours/movies per month
Shows	Watch	Followed / watched / archived shows, episode progress, ratings
Movies	Watch	Watched / watchlisted / rated / reacted movies, with watch dates
Watch history	Watch	A chronological watch timeline of episodes and movies
Lists	Watch	Your custom lists and collections, resolved to titles with cover art
Ratings	Ratings	The Bad→Wow star ratings you gave, shows, movies and episodes
Reactions	Ratings	"How did you feel?" reactions, grouped per episode/movie
Character votes	Ratings	Characters you voted for, actor, poster, show/episodes (names & posters from the extended backup)
Comments	Community	Every comment you posted, with attached images, likes, and reply threads
Notifications	Community	Read-only activity feed, likes, replies, mentions, follow requests, badges, airing reminders
Friends	Community	Your friends, real names and avatars (from the extended backup)
Badges	Community	Badges you earned, grouped by type with counts and the shows that earned them
Profile		Account details, avatar, and cover image
All data		Browse, sort, filter, and export any CSV table in the archive

Curated views support search, sort, filter, and CSV/JSON export.

Extended backup

Some of your data lives behind TV Time's servers, not in the export: images on its CDN (comment images, notification avatars, badge art, friends' avatars) and names behind numeric IDs (the characters you voted for, and your friends' real names). While TV Time is online the app fills these in live, but when the servers go offline they're gone, unless you capture them first.

extended-backup.py builds one portable zip with everything. It's plain Python 3 (no other dependencies), runs the same on macOS, Linux and Windows, and needs no login, the names come from TV Time's public API, the images from its CDN.

1. Unzip your export somewhere (so meme.csv and friends sit in a folder).

2. Run it while TV Time is still up:

   python3 extended-backup.py path/to/your/export
   # writes tvt-extended-backup.zip

   It packs everything into one tvt-extended-backup.zip, in folders the app understands: comments/, avatars/, badges/, characters/, friends/, plus characters.json / friends.json for the resolved names. Comment memes are saved in both a clean and a watermarked "marked" variant. Re-running is safe and resumes over already-downloaded images (and keeps previously-resolved names if the API can no longer be reached).

3. In the app, open ⚙ → Import extended backup (or the Import backup button at the top of the Comments view) and choose that zip.

Imported data is stored in your browser and shown from the local copy, so it keeps working after TV Time is gone. Missing images fall back to a placeholder.

Character posters come from TheTVDB (which outlives TV Time), and show posters / episode stills come from the keyless TVmaze API (see below), neither needs backing up.

Optional metadata

Off by default; toggle in the ⚙ Settings menu. Both cache results locally and send only a show or movie name to the API when enabled:

• Show metadata, episode titles, posters, and thumbnails from the keyless TVmaze API (matched by TheTVDB id).
• Movie titles, English titles for localized names via Wikidata.

Notes on the data

• Ratings vs reactions. TV Time reused numeric IDs across many versioned "sets" over the years, and the export dropped the set name, so IDs are decoded by source. The ratings-* files (+ tv_show_rate) are the 5-level star scale (Bad/Meh/Okay/ Good/Wow) shown under Ratings across shows, movies and episodes; the emotions-* files (+ the feelings hidden in episode_emotion) drive Reactions, the 12 "how did you feel?" feelings plus older emoji-grid reactions.
• Comments. Your comments are gathered from several tables (episode_comment, show_comment, profile_comment, and the newer comments-prod-comments), with images joined from meme.csv. Replies keep their parent's text only when the parent is also one of your comments, other people's comments aren't in the export.
• Notifications & badges. Your activity feed (notifications-prod-notifications) and earned badges (user_badge) each get a view; badge art and per-badge shows are reconstructed by joining IDs across tables, and follow-request notifications even recover a few usernames.
• Friends & characters. friend.csv and show_character_episode_vote.csv hold only numeric IDs, no names in the export. Their real names/avatars/posters are resolved from TV Time's public API by the extended-backup step above (before shutdown). Likes you gave have no dedicated view, but every table stays browsable under All data.

Run your own copy

Static site, no build step. netlify.toml publishes the viewer/ folder as-is; any static host works. The Refract viewer ships from the same folder, at viewer/refract/. Keep your own exports out of the published folder and out of git.

License

MIT.