python setup.py install
- Retrieve an API key from NASS's QuickStats API page. Save this key to a file or provide it directly as an argument.
- Import the nass_api class from nasspy into a Python environment and create an api object.
- Retrieve dataframes of available parameters and definitions.
- Retrieve a list of input options for any one parameter.
- Use the above parameters and options to build a query.
- And retrieve the results of that query as a Pandas data frame.
from nasspy import nass_api1) Create an api object. To use your NASS API key follow the prompts to set and save the key to a file using the default save path. The default save path is ~/.keys/nass_api_key.txt. You may also save the key to a specific file and specify that file path when creating the object. You may also enter the key directly as an argument and have it saved to the file path specified.
api = nass_api()
# or
api = nass_api(keypath='/any/path/anyfile.txt')
# or
api = nass_api(key="KEYCHARACTERSHERE", keypath='/any/path/anyfile.txt')whats = api.what_parameters
wheres = api.where_parameters
whens = api.when_parameters
print(whats.head(4))
#> Parameter Max Length Definition
#> 0 source_desc 60 Source of data (CENSUS or SURVEY). Census prog...
#> 1 sector_desc 60 Five high level, broad categories useful to na...
#> 2 group_desc 80 Subsets within sector (e.g., under sector = CR...
#> 3 commodity_desc 80 The primary subject of interest (e.g., CORN, C... api.get_parameter_options("commodity_desc") # A 'What' option
#> ['AG LAND',
#> 'AG SERVICES',
#> 'AG SERVICES & RENT',
#> 'ALCOHOL COPRODUCTS',
#> 'ALMONDS',
#> 'ALPACAS',
#> ...]print(api.operator_options)
#> CALL FAMILIAR ANALOG DESCRIPTION
#> 0 __LE <= Less than or equal to.
#> 1 __LT < Less than.
#> 2 __GT > Greater than.
#> 3 __GE >= Greater than or equal to.
#> 4 __NE != Not equal to.
#> 5 __LIKE like Pattern is in.
#> 6 __NOT_LIKE not like Pattern is not in.5) For numeric parameters, it appears necessary to use an operator (e.g.: "year__GE=1980" for years since 1980).
query = ['state_name=IOWA', 'commodity_desc=CORN', 'year__GE=2018', 'freq_desc=WEEKLY']data = api.get_query(query)
print(data.head())
#> CV (%) Value agg_level_desc asd_code ... watershed_desc week_ending year zip_5
#> 0 6 STATE ... 2019-06-02 2019
#> 1 7 STATE ... 2019-06-09 2019
#> 2 8 STATE ... 2019-06-16 2019
#> 3 10 STATE ... 2019-06-23 2019
#> 4 11 STATE ... 2019-06-30 2019
#>
#> [5 rows x 39 columns]query = ['state_name=IOWA', 'commodity_desc=ALPACAS', 'year__GE=2018', 'freq_desc=WEEKLY']
api.get_query(query)
#> {'error': ['bad request - invalid query']}9) And there is 50,000 record limit on returns. Here, corn records since 1950 for all states with an "I" in their name.
query = ['state_name_like=I', 'commodity_desc=CORN', 'year__GE=1950', 'freq_desc=WEEKLY']
api.get_query(query)
#> {'error': ['exceeds limit=50000']}