duprly

A CLI for downloading and processing DUPR data into a local SQLite database.

What Changed

The CLI has been rebuilt around a single entrypoint with task-based commands:

• duprly sync ...
• duprly browse ...
• duprly explore ...
• duprly export ...
• duprly db ...
• duprly doctor ...

Running duprly with no subcommand opens an interactive menu in TTY sessions.

Setup

Create a .env file (or pass one via --config):

DUPR_USERNAME=<username>
DUPR_PASSWORD=<password>
DUPR_CLUB_ID=8436164521

Install dependencies:

pip install -r requirements.txt

CLI Usage

Global options

duprly [--verbose] [--quiet] [--no-color] [--json] [--config PATH] [--interactive|--no-interactive] ...

Core workflows

# interactive mode (TTY)
duprly

# sync everything
duprly sync all

# sync only players from a specific club
duprly sync players --club-id 8436164521

# sync matches for everyone currently in local DB
duprly sync matches --all-players

# refresh ratings only where missing
duprly sync ratings

# browse a player and persist to DB
duprly browse player 6886613721

# fetch one player's full match history (ALL by default) and persist
duprly browse matches 6886613721

# fetch one player's match history in a bounded date range
duprly browse matches 6886613721 --start-date 2025-01-01 --end-date 2025-12-31

# fetch one player's rating history (defaults: both types, last 2 years)
duprly browse rating-history 6886613721

# fetch only doubles rating history for a specific date range
duprly browse rating-history 6886613721 --type doubles --start-date 2025-12-14 --end-date 2026-02-21

# live lookup by player name with suggestions, then optional match download
duprly browse lookup --entity player --query "aidan bai"

# open interactive saved-data explorer (no API calls)
duprly explore

# list saved players and export to CSV
duprly explore players --query "sternlicht" --export csv --output explore_players.csv

# inspect one saved player and drill into ratings/matches
duprly explore player 6886613721
duprly explore player 6886613721 ratings --type both
duprly explore player 6886613721 matches

# inspect one saved match with full raw payload
duprly explore match 4295493637 --export json --output explore_match_4295493637.json

# browse raw metadata snapshots from local tables
duprly explore raw --kind player --player-dupr-id 6886613721

# launch local web explorer (Datasette)
duprly explore web --open

# import DUPR auth token from browser cookies (top-level shortcut)
duprly import-browser-token --browser comet --domain dupr.com

# import DUPR auth token from browser cookies (Safari default)
duprly auth import-browser-token --browser safari

# search clubs
duprly browse clubs "NYC pickle"

# export club players (search/select)
duprly export club-players "NYC pickle"

# export rankings
duprly export rankings 8436164521

# export workbook from local DB
duprly export workbook --output dupr.xlsx

# DB stats / rebuild reporting table
duprly db stats
duprly db rebuild-match-detail

# diagnostics
duprly doctor check
duprly doctor check --api

Question-Driven Datasette

duprly explore web --open now launches a guided Datasette dashboard with canned analytics queries for pickleball ratings.

Each dashboard card now supports direct data retrieval:

• Get JSON: opens the query as JSON (_shape=objects).
• Get CSV: opens the query as CSV.
• Copy API URL: copies a ready-to-call JSON endpoint URL.

Dashboard homepage also includes Add player to local DB with live DUPR autocomplete (name or ID):

• searches DUPR as you type (CLI-style completion labels: Full Name [dupr_id])
• marks suggestions already saved locally
• imports the selected player into local SQLite
• optional one-click rating-history and full match-history sync during import

For player_rating_over_time, the query page itself also includes:

• Request latest player data (fetches/persists player profile + rating history from DUPR)
• optional include full match history
• auto-refresh of the page after fetch completes

Common question URLs

# this player's rating over time (default doubles)
http://127.0.0.1:8001/dupr/player_rating_over_time?dupr_id=6886613721&rating_type=DOUBLES

# recent form (last 90 days by default)
http://127.0.0.1:8001/dupr/player_recent_form?dupr_id=6886613721&days=90

# partner breakdown
http://127.0.0.1:8001/dupr/player_partner_breakdown?dupr_id=6886613721&days=90

# opponent breakdown
http://127.0.0.1:8001/dupr/player_opponent_breakdown?dupr_id=6886613721&days=90

# club top risers
http://127.0.0.1:8001/dupr/club_top_risers?club_id=7735643894&days=90&rating_type=DOUBLES

# direct JSON data retrieval (object rows)
http://127.0.0.1:8001/dupr/player_recent_form.json?dupr_id=6886613721&days=90&_shape=objects

# direct CSV data retrieval
http://127.0.0.1:8001/dupr/player_recent_form.csv?dupr_id=6886613721&days=90

Parameter glossary

• dupr_id: numeric DUPR player ID
• club_id: numeric DUPR club ID (optional in club queries)
• rating_type: DOUBLES (default) or SINGLES
• days: rolling window for "recent" analysis (default 90)
• start_date / end_date: optional YYYY-MM-DD date bounds for trend queries

Legacy Compatibility

Old command names are still available with deprecation warnings:

• get-data -> sync all
• get-all-players -> sync players
• get-all-player-matches -> sync matches --all-players
• get-player -> browse player
• get-matches -> browse matches
• update-ratings -> sync ratings
• build-match-detail -> db rebuild-match-detail
• stats -> db stats
• write-excel -> export workbook

The old script entrypoint still works:

python /Users/leosternlicht/repos/duprly/duprly.py --help

Standalone helper scripts are now thin wrappers:

• get_players.py
• get_club_rankings.py
• get_matches.py

Notes

• DUPR endpoints used here are unofficial and may change.
• Match history uses the browser-proven request shape (POST /player/v1.0/{id}/history).
• duprly explore is read-only and works entirely from local SQLite data.
• Explorer export supports CSV/JSON to file and best-effort clipboard copy.
• duprly explore web depends on Datasette 0.59, which currently needs setuptools<81 (pkg_resources).
• duprly explore web now generates guided metadata and templates at launch for canned analytics queries.
• datasette-vega enables richer chart rendering; if missing, query pages still work in table mode.
• For easier web browsing, rating now includes cached player_dupr_id and player_full_name columns.
• Data is written to dupr.sqlite.
• duprly doctor check is useful before first sync.