enovapower 0.11.0

Python client for downloading electricity usage data from Enova Power

enovapower

⚠️ Unofficial project — not affiliated with, endorsed by, or supported by Enova Power Corp. "Enova Power" is a trademark of its respective owner. Use at your own risk under the Apache-2.0 license.

A Python library for downloading electricity usage data from the Enova Power customer portal.

Enova Power serves residential and commercial customers in the Kitchener-Waterloo region of Ontario, Canada. Their My Account portal provides smart meter data exports, but only through a web UI. This library automates that process so you can pull your usage data into scripts, notebooks, dashboards, or data pipelines.

Quick start

Install

Using uv (recommended):

uv add enovapower

Using pip:

pip install enovapower

Environment variables

Credentials can be set via environment variables instead of passing them directly:

export ENOVA_USERNAME="[e-mail]"
export ENOVA_PASSWORD="your_password"

Async client (primary)

The AsyncEnovaClient is the primary interface.

from datetime import date
from enovapower import AsyncEnovaClient

async with AsyncEnovaClient() as client:
    await client.login("[e-mail]", "your_password")  # Do not pass credentials if they are set via environment variables
    readings = await client.download_usage(date(2026, 2, 25), date(2026, 3, 26))
    for r in readings:
        print(f"{r.date}: {r.total:.2f} kWh")

Sync client (convenience)

The EnovaClient is a thin synchronous wrapper for scripts and non-async contexts. It runs a dedicated background event loop.

from datetime import date
from enovapower import EnovaClient

client = EnovaClient()
client.login("[e-mail]", "your_password")

readings = client.download_usage(date(2026, 2, 25), date(2026, 3, 26))
for r in readings:
    print(f"{r.date}: {r.total:.2f} kWh")

Features

• Authenticate with the Enova Power My Account portal
• Download hourly smart meter usage data as UsageReading dataclasses
• Convert hourly readings to timezone-aware UTC (timestamp, kWh) intervals via UsageReading.intervals()
• Distinguish missing hours (None) from real zero-consumption hours
• Download Green Button XML exports and parse them with parse_green_button_xml() (safe defusedxml parsing)
• Multi-meter accounts via meter_ids / select_meter()
• Optional re-authentication callback to avoid retaining credentials
• Download tariff rates for all pricing plans (Time-of-Use, Ultra-Low Overnight, Tiered)
• Automatically chunk requests for date ranges exceeding 90 days
• Store and incrementally update usage history in a local SQLite database
• Automatic retry with exponential backoff on transient errors
• Session expiry detection with automatic re-login
• Configurable base_url for custom portal endpoints
• Built-in logging with customizable logger support

Local storage

from enovapower import EnovaClient, UsageStore

client = EnovaClient()
client.login("[e-mail]", "your_password")

with UsageStore("usage.db") as store:
    store.seed(client, months=12)   # initial backfill
    store.update(client)            # incremental update
    readings = store.load("111111", from_date, to_date)

Polling interval

The Enova portal is a utility web UI, not a high-throughput API. Avoid polling more frequently than every 15 minutes. A 30-minute interval is recommended for regular updates.

Logging

The library uses Python's standard logging module. The logger name is "enovapower".

Basic usage

import logging

logging.basicConfig(level=logging.DEBUG)
# Now all enovapower logs will appear

Custom logger

You can inject a custom logger to both clients and storage:

import logging

my_logger = logging.getLogger("my_app")
my_logger.setLevel(logging.INFO)

client = AsyncEnovaClient(logger=my_logger)
store = UsageStore("usage.db", logger=my_logger)

Built-in configuration

The library provides a convenience function to set up default handlers:

from enovapower.logger import configure_logging

# Configure with default format and DEBUG level
configure_logging(level=logging.DEBUG)

# Or with custom format
configure_logging(
    level=logging.INFO,
    format_string="%(asctime)s - %(levelname)s - %(message)s"
)

Security

Credentials passed to login() are stored in memory to enable automatic re-authentication on session expiry. They are never written to disk.

The library requires HTTPS by default. Use allow_insecure_http=True only for local testing against a development endpoint.

UsageStore databases are created with owner-only permissions (0600), since hourly usage data can reveal household occupancy patterns.

Documentation

See docs/usage.md for detailed API documentation and examples.

Development

See CONTRIBUTING.md for setup instructions and development workflow.

License

Apache-2.0. See LICENSE for details.