sda.api.load_data_cloud

Timeseries data loader that calls the SDA cloud /api/timeseries endpoint.

Used as a fallback when no local PostgreSQL credentials are found in ~/sda.json. Authentication uses the MSAL interactive / device-code flow (via sda.api.graph_auth.get_token_auto()) and the token is cached at ~/.sda/token_cache.bin so the user only needs to log in once.

To enable the cloud fallback, add SDA_CLOUD_URL to ~/sda.json:

{
    "SDA_CLOUD_URL": "https://sda.example.com"
}

The environment variable SDA_CLOUD_URL is also accepted and takes precedence over the config file value.

Functions

get_cloud_url()	Return the SDA cloud base URL, or None if not configured.
load_data_via_cloud(tag[, start_time, end_time])	Load timeseries data by calling the SDA cloud /api/timeseries endpoint.

Module Contents

sda.api.load_data_cloud.get_cloud_url()

Return the SDA cloud base URL, or None if not configured.

Checks the SDA_CLOUD_URL environment variable first, then ~/sda.json. Trailing slashes are stripped.

sda.api.load_data_cloud.load_data_via_cloud(tag, start_time=None, end_time=None)

Load timeseries data by calling the SDA cloud /api/timeseries endpoint.

Acquires a Microsoft access token via MSAL (silent refresh → browser popup → device-code flow). The token is cached at ~/.sda/token_cache.bin so the user only needs to log in once.

Parameters:

• tag – Test tag to query (e.g. gif1_t392). The same partial / case-insensitive matching logic as the direct DB path applies on the server side.

• start_time – Start of the time range (inclusive). Forwarded as an ISO-8601 string.

• end_time – End of the time range (inclusive). Forwarded as an ISO-8601 string.

Returns:

Wide DataFrame with a time column and one column per measurement, identical in structure to the direct-DB path.

Return type:

pandas.DataFrame

Raises:

• RuntimeError – If SDA_CLOUD_URL is not configured, authentication fails, or the HTTP request fails.

• ValueError – If the SDA cloud reports that no tables were found for tag.