Parse your Swiggy Food and Instamart PDF bills, store structured data in PostgreSQL, and visualize your spending with pre-built Grafana dashboards.
- Extracts order data from Swiggy-exported summary PDFs (Food + Instamart)
- Downloads detailed invoices via embedded PDF hyperlinks
- Parses every field: items, prices, discounts, per-item taxes (HSN, CGST, SGST), handling fees
- Loads everything into PostgreSQL with full upsert support (safe to re-run)
- Provides a Grafana dashboard with 30+ panels for spending analysis
- Supports multiple users (e.g., family members) from separate PDF exports
Swiggy provides account statements as PDF summaries. Here's how to request them:
- Open the Swiggy app and tap your Profile icon
- Select Account Statements
- Select the Duration (maximum 6 months per request)
- Select the Category — Food or Instamart
- Tap Get Report — takes approximately 3 hours to arrive via email
Note: The summary PDF contains a header with totals, followed by a table of orders. Each order has a "View" link pointing to the detailed invoice PDF. SwiggyIt automatically downloads and parses all linked detail invoices.
You'll need to request separate reports for Food and Instamart. If you want more than 6 months of history, request multiple reports with different date ranges — SwiggyIt handles deduplication automatically.
- Docker and Docker Compose
git clone https://github.com/dannotes/swiggyit.git
cd swiggyit
cp .env.example .env
# Edit .env and set your passwordsinput/
├── food/
│ └── order_summary_food_<uuid>.pdf
└── instamart/
└── order_summary_instamart_<uuid>.pdf
docker compose up -dThis pulls the pre-built image from ghcr.io/dannotes/swiggyit, starts PostgreSQL (auto-creates the schema on first run), runs the parser to load your PDFs, and brings up Grafana.
Open http://localhost:3000 and log in with admin / admin (or your configured password from .env).
SwiggyIt - Swiggy Bill Parser
========================================
Input dir: /data/input
Temp dir: /data/.tmp
Database: connected
[Food] Parsing summary: order_summary_food_<uuid>.pdf
Customer: John Doe (john@example.com)
Orders: 36
Downloading detail PDFs...
food: 36 downloaded, 0 cached, 0 failed
Loaded: 36/36 orders (0 failed)
[Instamart] Parsing summary: order_summary_instamart_<uuid>.pdf
Customer: John Doe (john@example.com)
Orders: 55
Downloading detail PDFs...
instamart: 55 downloaded, 0 cached, 0 failed
Loaded: 55/55 orders (0 failed)
Done!
A pre-built Grafana dashboard with 30+ panels visualizes your spending across four sections:
Total orders, grand total spend, food vs instamart breakdown, monthly and weekly spending trends.
Top restaurants by spend and frequency, most ordered items, most expensive items, tax per restaurant, monthly tax trends.
Top sellers, most bought items, handling fee breakdown, discount savings, monthly handling fee and discount trends.
Orders by day of week, order value distribution, cumulative spend over time, repeat purchases, outlier detection.
To build and run locally instead of pulling the pre-built image:
docker compose -f docker-compose-dev.yml up --buildRun tests:
pip install -r requirements.txt
pytest -m "not integration"Safe to run multiple times. The pipeline is idempotent:
| Step | Behavior |
|---|---|
| Download detail PDFs | Skipped if already cached in .tmp/ |
| Customers | Upserted by email |
| Orders | Upserted by order_id |
| Order items | Deleted and reinserted per order |
| Handling fees | Upserted by order_id |
Drop their exported PDFs into input/food/ and input/instamart/ and re-run. A new customer record is created automatically based on their email from the PDF header. Use the customer filter dropdown in Grafana to switch between users.
customers
├── food_orders → food_order_items
└── instamart_orders → instamart_order_items
→ instamart_handling_fees
6 tables capturing every field from the invoices — item descriptions, quantities, unit prices, discounts, per-item tax breakdowns (CGST/SGST/Cess), HSN codes, restaurant/seller GSTIN, FSSAI numbers, and more.
swiggyit/
├── Dockerfile
├── LICENSE
├── README.md
├── requirements.txt
├── pytest.ini
├── .env.example
├── docker-compose.yml # Production (pulls from GHCR)
├── docker-compose-dev.yml # Development (local build)
├── .github/
│ └── workflows/
│ └── ci.yml # Tests + GHCR image publish
├── grafana/
│ └── provisioning/
│ ├── datasources/
│ │ └── datasource.yml # Auto-provisioned PostgreSQL connection
│ └── dashboards/
│ ├── dashboard.yml # Dashboard provider config
│ └── json/
│ └── swiggyit.json # Combined dashboard (30+ panels)
├── screenshots/ # Dashboard and bill screenshots
├── sql/
│ ├── schema.sql # Database DDL (auto-runs on first start)
│ └── analyze.sql # Standalone analysis queries
├── docs/
│ └── architecture.md # Detailed design document
├── input/ # Drop exported PDFs here (gitignored)
│ ├── food/
│ └── instamart/
├── tests/
│ ├── conftest.py
│ ├── test_validator.py
│ ├── test_summary_parser.py
│ ├── test_food_parser.py
│ ├── test_instamart_parser.py
│ ├── test_loader.py
│ ├── test_downloader.py
│ └── test_integration.py # Requires actual PDFs
└── src/
├── main.py # CLI entrypoint
├── validator.py # Input validation
├── downloader.py # Fetch detail PDFs from S3 URLs
├── loader.py # PostgreSQL upsert logic
└── parser/
├── summary_parser.py # Parse summary PDFs (PyMuPDF)
├── food_parser.py # Parse food invoices (pdfplumber)
└── instamart_parser.py # Parse instamart invoices (pdfplumber)
| Component | Technology |
|---|---|
| Language | Python 3.12+ |
| PDF link extraction | PyMuPDF (fitz) |
| PDF table parsing | pdfplumber |
| Database | PostgreSQL 17 |
| DB driver | psycopg 3 |
| HTTP client | httpx |
| Infrastructure | Docker Compose |
| Dashboards | Grafana |
| Testing | pytest + pytest-cov |
python3 src/main.py --help
options:
--input PATH Input directory with food/ and instamart/ folders (default: input)
--tmp PATH Temp directory for downloaded detail PDFs (default: .tmp)
--db-url URL PostgreSQL connection URL (default: built from env vars)
| Variable | Default | Description |
|---|---|---|
POSTGRES_HOST |
localhost |
Database host |
POSTGRES_PORT |
5432 |
Database port |
POSTGRES_DB |
swiggyit |
Database name |
POSTGRES_USER |
swiggyit |
Database user |
POSTGRES_PASSWORD |
(required) | Database password |
GF_ADMIN_USER |
admin |
Grafana admin username |
GF_ADMIN_PASSWORD |
admin |
Grafana admin password |
GRAFANA_PORT |
3000 |
Grafana port |
-- Monthly food spending
SELECT date_trunc('month', date_of_invoice) as month,
count(*) as orders,
sum(invoice_total) as total
FROM food_orders
GROUP BY 1 ORDER BY 1;
-- Top restaurants by spend
SELECT restaurant_name, count(*) as orders, sum(invoice_total) as total
FROM food_orders
GROUP BY 1 ORDER BY 3 DESC;
-- Most bought instamart items
SELECT description, sum(quantity) as qty, sum(total_amount) as total
FROM instamart_order_items
GROUP BY 1 ORDER BY 2 DESC LIMIT 10;
-- Total tax paid
SELECT 'Food' as type, sum(total_taxes) as tax FROM food_orders
UNION ALL
SELECT 'Instamart', sum(cgst_amount + sgst_amount + cess_amount)
FROM instamart_order_items;