Automapa Map Services Python SDK

European Geocoding, Routing & Address Autocomplete

Python SDK for the Automapa Map Services REST API. Simplifies integration with geocoding, routing, and address autocomplete services for Europe.

---

Getting started with your .env file

You will receive a .env file from us containing your API credentials. Place it in the root directory of this repository:

AUTOMAPA_API_KEY=your-key
AUTOMAPA_API_PASSWORD=your-password

Once the .env file is in place, you can test all endpoints against the live API using Docker — no local Python installation required:

# Run the full test suite (unit + integration) inside Docker
make docker-test

# Run only integration tests (live API calls)
make docker-test-integration

# Run only unit tests (no API calls)
make docker-test-unit

# Run all checks (unit, integration, lint, type checks)
make docker-check

Zero dependencies. The SDK has no runtime dependencies — it uses only Python's built-in urllib. It is compatible with Python 3.11 and above.

---

1. Overview

The SDK covers:

• Geocoding — convert an address to coordinates and back
• Routing — calculate routes, distance matrices, and point-order optimization
• Address autocomplete — suggestions as the user types
• Google Geocoding API compatibility — optional google response format

For the full list of endpoints and parameters, see the Automapa REST API documentation.

---

2. Installation

pip install automapa-map-services

Or with uv:

uv add automapa-map-services

Requirements: Python 3.11+, no external runtime dependencies.

---

3. Client configuration

from automapa_map_services import ApiClient, Config

client = ApiClient(Config(
    key="YOUR_API_KEY",
    password="YOUR_PASSWORD",
))

---

4. Session handling

A session is opened automatically (lazy) on the first call that requires authorization. To open it manually (fetches a salt, hashes the configured password and stores the session id for subsequent calls):

session_id = client.open_session()

Sessions opened through Session.login() / generate_session() are not stored in the client; use client.open_session() when you want the client itself to reuse the session.

---

5. Available services

Class	Methods
Geocoder	geocode(), geocodemulti(), revgeocode(), revgeocodemulti()
Autocomplete	search(), search_address()
Road	get_segment_info(), speed_check(), speed_check_multi()
RoadPermit	add(), overwrite(), remove(), list()
Routes	matrix(), optimize(), optimize_queue(), optimize_queue_result(), route()
Speed	speed_report(), speed_report_result()

All methods also accept camelCase aliases (searchAddress(), getSegmentInfo(), etc.) for consistency with the PHP SDK.

Geocoder

# Address → coordinates
response = client.geocoder().geocode({
    "address":    {"city": "Warsaw", "street": "Domaniewska", "house": "37"},
    "maxResults": 1,
})
results = response.data()  # list of results, each with x, y, city, street, ...

# Reverse geocoding — coordinates → address
response = client.geocoder().revgeocode({
    "point":  [21.0073642, 52.18288],
    "params": {},
})
address = response.data()  # city, street, house, pcode, ...

Autocomplete

Address and POI autocomplete — returns suggestions as the user types.

search() — search by phrase

response = client.autocomplete().search({
    "query": "Warsaw Domaniewska 37",
    # "types": ["place"],  # optional: "place", "poi", or both
})

items = response.data()["items"]

for item in items:
    print(item["name"])          # "Domaniewska 37, 02-672 Warsaw"
    print(item["type"])          # "place"
    print(item["subtype"])       # "building"
    print(item["coords"]["x"])   # longitude (lng)
    print(item["coords"]["y"])   # latitude (lat)

Each items entry contains: type, subtype, name, province, district, community, place, quarter, street, number, postal_code, coords.{x,y}.

search_address() — step-by-step address search

Useful for multi-field address forms (city → street → number → postcode).

response = client.autocomplete().search_address({
    "query": {
        "city":    "Warsaw",
        "street":  "Dom",     # partial street name
        "number":  "",
        "zipcode": "02-672",
    },
    "source":   "street",              # field being searched: "place", "street", "number", "zipcode"
    "selected": ["city", "zipcode"],   # fields already confirmed by the user
    # "limit": 10,                     # optional: max results (default 10)
})

items = response.data()["items"]
# [{"city": "Warsaw", "street": "Domaniewska", "zipcode": "02-672", "count": 1}]

Each entry may contain: city, street, number, zipcode, count, and optionally coords.{x,y}.

Note on Autocomplete coordinates: Unlike Geocoder, coordinates are nested: coords.x = longitude, coords.y = latitude.

Routes

Route calculation, distance matrix, and point-order optimisation.

route() — calculate a route

response = client.routes().route({
    "points": [
        {"x": 21.0073642, "y": 52.2297},  # Warsaw (x=lng, y=lat)
        {"x": 18.6282,    "y": 54.3520},  # Gdańsk
    ],
    "route": {
        "type":    "short",  # "quick" | "short" | "optimal"
        "traffic": True,     # True = take live traffic into account
    },
    "object": {"type": "car"},
})

data = response.data()
print(round(data["length"] / 1000, 1), "km")
import time; print(time.strftime("%H:%M:%S", time.gmtime(data["eta"])))  # estimated travel time
print(data["mapurl"])   # link to map
# data["desc"]  — turn-by-turn manoeuvre list
# data["tolls"] — toll charges

matrix() — distance / time matrix

response = client.routes().matrix({
    "points": [
        {"x": 21.0073642, "y": 52.2297},
        {"x": 18.6282,    "y": 54.3520},
        {"x": 16.9252,    "y": 52.4064},
    ],
    "type": 1,  # 1 = time (duration in minutes), 2 = distance (length in metres)
})

for row in response.data():
    print(f"Segment {row['points']}: {round(row['length'] / 1000)} km, {row['duration']} min")

optimize() — synchronous optimisation (max 10 points)

response = client.routes().optimize({
    "points":   [{"x": 21.0, "y": 52.2}, {"x": 18.6, "y": 54.4}],
    "type":     1,      # 1 = time, 2 = distance
    "fixedEnd": False,  # True = last point is the destination; False = return to start
    "object":   {"type": "car"},
})

data = response.data()
data["order"]    # optimal index sequence, e.g. "0,3,2,1"
data["sections"] # segment details: x, y, eta, dist

optimize_queue() + optimize_queue_result() — asynchronous optimisation (max 50 points)

import time

# Step 1: submit — each point requires an "id" field
response   = client.routes().optimize_queue({
    "points":     [{"id": "Warsaw", "x": 21.0, "y": 52.2}],
    "optimizeBy": "time",  # "time" or "distance"
    "object":     {"type": "car"},
})
request_id = response.data()["requestId"]

# Step 2: poll for the result
while True:
    time.sleep(2)
    result = client.routes().optimize_queue_result({"requestId": request_id})
    data   = result.data()
    if data["resultReady"] is True:
        break

data["optimizedSequence"]  # list of point IDs in the optimal order

---

6. Examples

# Copy configuration
cp .env.example .env
# Fill in AUTOMAPA_API_KEY and AUTOMAPA_API_PASSWORD in .env

# Routes
python examples/routes_route.py                # point-to-point route
python examples/routes_matrix.py               # distance/time matrix
python examples/routes_optimize.py             # sync optimisation (max 10 points)
python examples/routes_optimize_async.py       # async optimisation (max 50 points)

# Geocoder
python examples/geocoder_geocode.py            # single address → coordinates
python examples/geocoder_geocodemulti.py       # batch address geocoding
python examples/geocoder_revgeocode.py         # coordinates → single address
python examples/geocoder_revgeocodemulti.py    # coordinates → multiple addresses

# Autocomplete
python examples/autocomplete_search.py         # search places and POI by phrase
python examples/autocomplete_search_address.py # search address objects

# Road
python examples/road_segment_info.py           # road segment metadata
python examples/road_speed_check.py            # speed limit for a single point
python examples/road_speed_check_multi.py      # speed limits for multiple points

# Road permits
python examples/road_permit_list.py            # list permits
python examples/road_permit_add.py             # create a new permit
python examples/road_permit_overwrite.py <id>  # replace an existing permit
python examples/road_permit_remove.py <id>     # delete a permit

# Speed reports
python examples/speed_report.py                # submit async speed report
python examples/speed_report_result.py <id>    # poll for report results

Or simply run make examples to run all examples.

---

7. Passing additional parameters

The SDK does not restrict unknown parameters — all fields are forwarded to the API as-is:

client.geocoder().geocode({
    "address":       {"city": "Warsaw"},
    "myCustomParam": "value",  # passed through without modification
})

---

8. Low-level call() and request()

# Via service + method name (preferred)
client.call("Geocoder", "newMethod", {"param": "value"})

# Or via full URL path (fallback)
client.request("POST", "/v3/Geocoder/newMethod", {"param": "value"})

---

9. Response formats: native vs google

# Native format (default)
response = client.geocoder().geocode({"address": {"city": "Kraków"}})
data = response.data()   # API data
raw  = response.raw()    # full raw response (always available)

# Google format
response = client.geocoder().geocode({
    "address": {"city": "Kraków"},
    "format":  "google",
})

---

10. Google format field limitations

Google field	Value
geometry.viewport	always null (not provided by the API)
place_id	always null (not provided by the API)

Note: The Automapa API uses x = longitude and y = latitude — the opposite of Google's convention.

---

11. Error handling

Situation	Exception class
Invalid API key / password	AuthenticationException
Expired session	SessionException
Parameter validation failure	ValidationException
Resource not found	NotFoundException
Rate limit exceeded	RateLimitException
Network / urllib error	NetworkException
Server error (5xx)	ServerException
Unexpected response	UnknownResponseException

from automapa_map_services.exceptions import AuthenticationException, AutomapaException

try:
    response = client.geocoder().geocode({"address": {"city": "Warsaw"}})
except AuthenticationException as e:
    print("Authentication error:", e)
except AutomapaException as e:
    print("API error:", e)

---

License

MIT © 2026 Automapa sp. z o.o.