BincioActivity is a self-hosted, federated activity stats platform. You point it at a folder of GPX/FIT/TCX files, it produces a static website. The website runs anywhere — a Raspberry Pi, GitHub Pages, a USB stick. No database. No server process. No account required.

The philosophy in one sentence: your activity data is yours, it lives as plain files on your disk, and this tool turns those files into a beautiful site you control entirely.

How it works

GPX / FIT / TCX files
        │
        ▼
  bincio extract          ← Python CLI. Reads files, writes plain JSON.
        │
        ▼
  ~/bincio_data/          ← BAS data store. Human-readable JSON + GeoJSON.
   edits/*.md             ← Optional sidecar edits (titles, descriptions, photos).
        │
        ▼
  bincio render           ← Merges sidecars → _merged/. Runs Astro build.
        │
        ▼
  site/dist/              ← Drop anywhere. Open index.html. Done.

Everything in ~/bincio_data/ is plain text you can read, edit, back up, or publish to a CDN. The site build is fully reproducible from those files.

Quick start

# 1. Clone and install (requires Python >= 3.12 and uv)
git clone https://github.com/brutsalvadi/bincio-activity.git
cd bincio-activity
uv sync                     # installs the bincio package + all dependencies

# 2. Configure
cp extract_config.example.yaml extract_config.yaml
$EDITOR extract_config.yaml     # set input dirs, output dir, your name

# 3. Extract activities → BAS JSON
uv run bincio extract

# 4. Build the site (requires Node >= 20)
cd site && npm install && cd ..
cp site/.env.example site/.env  # configure BINCIO_DATA_DIR
uv run bincio render            # merges edits + runs astro build
# → open site/dist/index.html

For live development with hot reload:

uv run bincio render --serve    # merges edits, links data, starts astro dev
# → http://localhost:4321

# Optional: enable the activity edit UI
uv sync --extra edit            # install FastAPI + uvicorn (one-time)
uv run bincio edit              # starts edit server on http://localhost:4041
# Set PUBLIC_EDIT_URL=http://localhost:4041 in site/.env

Cheatsheet

Extract

bincio extract                              # uses extract_config.yaml
bincio extract --input ~/rides --output ~/bincio_data
bincio extract --file ride.gpx             # single file, prints JSON to stdout
bincio extract --since 2025-01-01          # only files newer than date

Supported formats: GPX, FIT, TCX — all with optional .gz compression. Strava bulk export: point metadata_csv at activities.csv to pull in titles, descriptions, and gear.

Extraction is incremental by default (incremental: true in config). Re-running only processes new or changed files. To force a full re-extract, delete ~/bincio_data/ or set incremental: false.

Site dev

cd site
npm run dev       # http://localhost:4321 — live reload on data or code changes
npm run build     # production build → site/dist/
npm run preview   # serve site/dist/ locally to check the production build

The site reads data from site/public/data/. Symlink your BAS store there:

ln -sf ~/bincio_data site/public/data

Python / tests

uv run pytest                          # full test suite
uv run pytest tests/test_fit.py -x     # single file, stop on first failure
uv run bincio --help                   # CLI help
uv sync                                # install / update dependencies

Configuration

`extract_config.yaml`

owner:
  handle: yourname
  display_name: Your Name

input:
  dirs:
    - ~/Activities/gpx
    - ~/Activities/fit
  metadata_csv: ~/strava_export/activities.csv   # optional — Strava titles/descriptions

output:
  dir: ~/bincio_data

default_privacy: public   # public | blur_start | no_gps | private

track:
  rdp_epsilon: 0.0001     # GPS track simplification (~11 m at equator)
  timeseries_hz: 1        # data samples per second stored in JSON

incremental: true         # skip files whose hash hasn't changed

Privacy levels

Level	GPS track	Stats	Appears in index
`public`	Full	Yes	Yes
`blur_start`	First/last 200 m removed	Yes	Yes
`no_gps`	Not published	Yes	Yes
`private`	Not published	No	No

Privacy is enforced at extract time. A private activity never enters index.json and is never served.

The BAS data store

bincio extract produces a directory of plain files — the BincioActivity Schema (BAS) store:

~/bincio_data/
  index.json                    ← summary of all activities + owner info
  activities/
    2024-05-15T08:30:00Z.json   ← full activity: stats, laps, timeseries
    2024-05-15T08:30:00Z.geojson  ← simplified GPS track (RDP)

index.json is everything the feed page needs — no extra fetches until you open an activity. {id}.json contains the full timeseries (elevation, speed, HR, cadence, power at 1 Hz) for charts and the detail map. Both are human-readable and editable with any text editor.

See SCHEMA.md for the full specification.

Federation (work in progress)

Add a friend's published index.json URL to your site_config.yaml:

data_sources:
  - type: local
    path: ~/bincio_data
  - type: remote
    handle: alice
    url: https://alice.example.com/bincio/index.json

At build time the renderer fetches their public data and renders it under /friends/alice/. Your site, their data — with full attribution. They control what they publish; you control what you display.

Tech stack

Layer	Technology
Extract	Python 3.12, click, fitdecode, gpxpy, lxml
Site framework	Astro 4 (static output)
UI components	Svelte 5
Styling	Tailwind CSS v3
Charts	Observable Plot
Maps	MapLibre GL v5 + OpenFreeMap tiles
Python packages	uv
Node packages	npm

Project layout

bincio/                 Python package
  extract/
    cli.py              `bincio extract` entry point
    parsers/            GPX, FIT, TCX parsers
    sport.py            sport name normalisation
    metrics.py          haversine stats (single-pass)
    timeseries.py       1 Hz downsampling
    simplify.py         RDP track simplification
    dedup.py            hash-based + near-duplicate detection
    strava_csv.py       Strava activities.csv reader
    writer.py           BAS JSON + GeoJSON writer
  render/
    cli.py              `bincio render` — merge + astro build/serve
    merge.py            sidecar edit overlay (produces _merged/)
  edit/
    cli.py              `bincio edit` — local edit server
    server.py           FastAPI write API for the edit drawer
schema/
  bas-v1.schema.json    JSON Schema for BAS format
SCHEMA.md               Human-readable BAS specification
site/                   Astro project
  src/
    pages/
      index.astro               Activity feed
      activity/[id].astro       Single activity detail
      stats/index.astro         Yearly heatmaps + totals
    components/
      ActivityFeed.svelte        Card grid, sport filter, pagination
      ActivityDetail.svelte      Map + stats + charts + photo gallery
      ActivityMap.svelte         MapLibre GL map
      ActivityCharts.svelte      Observable Plot charts
      StatsView.svelte           Heatmap, percentile scaling, sport filter
      EditDrawer.svelte          Slide-in activity editor
    lib/
      types.ts                  BAS TypeScript types
      format.ts                 Formatting helpers

Why no database?

Databases add operational complexity — backups, migrations, running processes, credentials. Activity data is append-only and read-heavy. Plain JSON files handle this perfectly, are trivially backed up with cp or rsync, can be diffed in git, and work offline. The site is a folder you can zip and email.

Why federation?

Strava, Garmin Connect, and similar platforms are silos. If the company shuts down or changes its terms, your data and your social graph go with it. BincioActivity's federation model is inspired by the open web: you host your own data at a URL, friends subscribe to that URL, and no central authority is involved.

Languages

Python 53.4%

Svelte 24.1%

Astro 17.9%

TypeScript 2.1%

HTML 1.5%

Other 1%