gmaps-scraper

A request-based Google Maps scraper that extracts place information and reviews. Light weight and fast.

Install

pip install -r requirements.txt

Usage

• sqlite3: Default db path is output/gmaps.db but you can specify any path with --db.

List places (search only, no scraping)

python main.py list "web developers in Stockholm" --max-places=20
python main.py list "hospitals in Nairobi"

Scrape a single place by ID

python main.py place "0x465f9b8fa03f30b3:0xb52faeba880b7674" --max-reviews=100 --db output/places.db

Search and scrape all results

python main.py search "coffee shops in London" --max-places=10 --max-reviews=50 --db output/london.db
python main.py search "restaurants in Nairobi" --db output/nairobi.db

Place details only, no reviews

python main.py search "hotels in London" --max-places=20 --max-reviews=0
python main.py place "0x465f9b8fa03f30b3:0xb52faeba880b7674" --max-reviews=0

Commands

Command	What it does	Returns
list	Search only - no details or reviews fetched	name, address, rating, coords, categories
place	Full scrape of a single known place	all fields + reviews → saved to DB
search	Search + full scrape of every result	all fields + reviews for every matched place

list -> Discover places and their IDs. One request per page. Does not fetch phone, opening hours, or reviews.

python main.py list "web developers in Stockholm" --max-places=20
# → quick list with place_id values

place -> Use when you already have a place_id (from list or elsewhere). Fetches full details and reviews, saves to DB.

python main.py place "0x465f9b8fa03f30b3:0xb52faeba880b7674" --max-reviews=100 --db output/places.db

search -> Runs a search then scrapes full details and reviews for every result. Most data, most requests.

python main.py search "hospitals in Nairobi" --max-places=5 --max-reviews=20 --db output/nairobi.db

Typical workflow:

# Step 1: find places and note the place_id
python main.py list "coffee shops in Paris" --max-places=20

# Step 2: scrape a specific one in full
python main.py place "0x47e671fd3387e0dd:0x174ea8ece7dffa45" --max-reviews=200 --db output/paris.db

Options

Flag	Description	Default
--db	SQLite database path	output/gmaps.db
--max-places	Max number of places to return	20
--max-reviews	Max reviews per place (0 = skip reviews)	50 (search), 100 (place)
--lat / --lng	Center coordinates (optional)	0
--proxy	Proxy URL (socks5://... or http://...)	None
--lang	Language code	en
--gl	Country/region code	us
--delay	Min seconds between requests	1.5

Database schema

Results are saved to two tables:

places -> one row per place, upserted on place_id

• place_id, name, address, lat, lng, rating, review_count
• website, phone, price_level, description
• categories, hours, photos, about, menu, booking_links (JSON columns)
• reviews_fetched - 0 if only place details were saved, 1 once reviews have been fetched

Use reviews_fetched to find places that still need reviews scraped:

SELECT place_id, name FROM places WHERE reviews_fetched = 0;

reviews -> one row per review, deduplicated on review_id

• review_id, place_id, reviewer_name, reviewer_profile_url, reviewer_is_local_guide
• rating, text, date, language, photos, owner_reply, owner_reply_date

What the scraper extracts

Place info

• Name, address, coordinates
• Rating, review count, price level
• Categories, website, phone number
• Opening hours
• Description (editorial summary)
• About section: services, accessibility, dining options, amenities (grouped yes/no attributes)
• Menu items with name, description, price (restaurants only)
• Booking / reservation links (OpenTable, TheFork, etc.)

Reviews

• Author name, profile URL, avatar, review count, local guide badge
• Rating, text, language, date
• Review photos (direct image URLs)
• Owner reply text and date

Contributions are welcome!

Disclaimer

This Scraper is provided for educational and research purposes only. By using it, you agree to comply with local and international laws regarding data scraping and privacy. The authors and contributors are not responsible for any misuse of this software. This tool should not be used to violate the rights of others, or for unethical purposes.