A Python module to interact with Lariat API and perform various operations like querying indicators, fetching datasets, etc.
To install the package, run the following command:
pip install lariat_python_sdkFirst, import the necessary components:
from lariat_client import configure, get_raw_datasets, get_datasets, get_indicators, get_indicator, query, Filter, FilterClause
import datetimeSet up your API key and application key:
API_KEY = "your-api-key"
APP_KEY = "your-app-key"
configure(api_key=API_KEY, application_key=APP_KEY)Get raw datasets:
raw_datasets = get_raw_datasets(dataset_ids=[1, 2, 3])Get computed datasets:
datasets = get_datasets()Get indicators:
indicators = get_indicators(datasets=datasets)Get a specific indicator:
indicator = get_indicator(id=1)Query an indicator:
from_ts = datetime.datetime(2022, 1, 1)
to_ts = datetime.datetime(2022, 2, 1)
group_by = ["country"]
filter_clause = FilterClause(field="country", operator="in", values="US,UK")
query_filter = Filter(clauses=[filter_clause], operator="and")
results = query(indicator.id, from_ts, to_ts, group_by, query_filter=query_filter)
# Convert results to a DataFrame
results_df = results.to_df()
# Save results to a CSV file
results.to_csv("results.csv")Use the RawQuery interface to add additional query arguments
Note: Query arguments attached via the RawQuery interface are subject to changes in their backend interpretation. Use with caution
import lariat_client
import datetime
lariat_client.configure(api_key="some_key", application_key="some_other_key")
indicator = lariat_client.get_indicator(id=1234)
from_ts = datetime.datetime(2023, 5, 1)
to_ts = datetime.datetime(2023, 5, 10)
filter_clause = lariat_client.FilterClause(field="country", operator="in", values="USA")
query_filter = lariat_client.Filter(clauses=[filter_clause], operator="and")
raw_query = lariat_client.RawQuery(
indicator_id=indicator.id,
from_ts=from_ts,
to_ts=to_ts,
aggregate="distinct",
query_filter=query_filter
)
raw_query.add_query_argument("x_axis", "custom_x_axis")
records = raw_query.send()
