Models

Latitude = Annotated[float, Field(ge=-90, le=90)]

Latitude in degrees, from -90 to 90.

Longitude = Annotated[float, Field(ge=-180, le=180)]

Longitude in degrees, from -180 to 180.

DataList = Annotated[list[float], Field(min_length=1)]

A non-empty list of floats.

Hour = int

Hour of the day, 0 to 23. TimePeriod checks the range.

Month = int

Month of the year, 1 to 12. TimePeriod checks the range.

Day = int

Day of the month, 1 to 31 and within its month. TimePeriod checks it.

Bases: BaseModel

Base class of the SDK's request payloads.

A payload is immutable and validated on creation. Unknown fields are rejected, and surrounding whitespace in strings is stripped. Fields are serialized under their kebab-case aliases by default; the alias and the Python field name are both accepted on input.

to_dict

to_dict(*, by_alias: bool = True, exclude_none: bool = True) -> dict

Return the payload as a plain dict.

Parameters:

Returns:

to_json

to_json(*, by_alias: bool = True, exclude_none: bool = True) -> str

Return the payload as a JSON string.

Parameters:

Returns:

to_gzip

to_gzip()

Return the payload's JSON, UTF-8 encoded and gzip-compressed.

Returns:

Bases: Payload

One analysis window, as the six integers the filter takes.

Warnings

A window is a date span with a daily hour range, not one continuous stretch of time: the hours the model counts. "1 March 08:00 to 30 September 18:00" selects hours 08 to 18 of every day from 1 March to 30 September. The hours are EPW local standard time: no daylight-saving hour is added or removed.

A window may cross the year end (for example 1 December to 28 February); it selects December, January and February of the typical year, in calendar order of the year.

Malformed values are refused: a month outside 1-12, an hour outside 0-23, a day outside its own month's length, and a window inside one month whose first day is after its last day. An analysis request also refuses an hour range across midnight (start_hour after end_hour) for solar radiation, direct sun hours, daylight availability and the two thermal comfort analyses. The check happens before the tiles are built and acquired, and therefore before anything is billed.

A single-hour window (start == end) is valid: it selects one hour of each day in the date span.

Attributes:

Bases: Payload

A point on Earth.

Attributes:

Bases: Payload

One hourly weather record.

Each field is the EPW column of the same name, with the EPW column's units. Every field is optional and None means no reading. Numeric strings are converted to floats, and empty strings and the tokens null, none, na, n/a and nan become None. Unknown keys are ignored.

Warnings

The pre-0.10 key znithLuminance is still accepted for zenithLuminance with a DeprecationWarning.

Attributes:

coerce_numeric_strings classmethod

coerce_numeric_strings(v)

• None stays None
• numbers stay numbers
• numeric strings -> float
• empty strings -> None
• 'null', 'na', 'nan' -> None
• non-numeric strings -> ValueError

to_camel_case(s: str) -> str

Convert a snake_case string to camelCase.

to_kebab_case(s: str) -> str

Convert a snake_case string to kebab-case.

check_time_window(
window: "TimePeriod", analysis_type: Optional[str] = None
) -> None

Refuse a window the SDK rules out before any request.

Without analysis_type the window is a weather selection. With one, the analyses the server checks also refuse an hour range across midnight. Raises ValueError.

extract_weather_fields(
weather_data: list[WeatherDataPoint], fields: list[str]
) -> dict[str, Any]

Extract the given fields from weather data points into flat lists.

A gap (None) in a requested field raises instead of being skipped, because skipping it would shorten that column and misalign it against every other array. For a complete weather window the output is the same as in 0.5.3.

Parameters:

Returns:

Raises: