I have written a data function to retrieve data from an API. I have used the information in the response to create a pandas data frame and I have set the dtype for each column. Spotfire does not use this typing and then fails when trying to auto assign.

[Bill Hall]

I have written a data function to retrieve data from an API. I have used the information in the response to create a pandas data frame and I have set the dtype for each column. Spotfire does not use this typing and then fails when trying to auto assign.

[Gaia Paolini]

There is a (little documented) Python package called 'spotfire', which is pre-installed.

You import it as usual with: import spotfire

Then you use it for instance like this:

spotfire.set_spotfire_types(df, {'col1':'String','col2':'Integer'})

use the Spotfire data types.

[Bill Hall]

Excellent, Thank you! I'll test it this afternoon and report back.

[Bill Hall]

That worked for me, thank you for your quick response. I included a snippet of my code in case someone else finds this in the future.

data_set = myapi.Dataset.get_by_full_path(data_path)

field_info = data_set.fields()

returned_data_set = data_set.query()

pd_field_list = {}

sf_field_list = {}

pd_data_type_map = {

  "string":"object",

  "text":"object",

  "auto":"object",

  "integer":"int64",

  "long":"int64",

  "double":"float64",

  "float":"float64",

  "boolean":"bool",

  "date":"datetime64"

}

sf_data_type_map = {

  "string":"String",

  "text":"String",

  "auto":"String",

  "integer":"Integer",

  "long":"LongInteger",

  "double":"Real",

  "float":"Real",

  "boolean":"Boolean",

  "date":"DateTime"

}

for field in field_info:

  pd_field_list[field.name]=pd.Series(dtype=pd_data_type_map.get(field.data_type,"object"))

  sf_field_list[field.name]=sf_data_type_map.get(field.data_type,"String")

qp_data_set = pd.DataFrame(returned_data_set, columns = pd_field_list)

spotfire.set_spotfire_types(qp_data_set, sf_field_list)

[Gaia Paolini]

Thank you for sharing this.