FX Option - The query for a Fx barrier does not work due to a JSON Serialization error

yaokoffi.kouassi

Hello,

I'm trying to execute a IPA query and it does not work.

I got this error: ""

Here is the code

import refinitiv.data as rd
import refinitiv.data.content.ipa.financial_contracts as rdf
from refinitiv.data.content.ipa.financial_contracts import option

rd.open_session()

response = option.Definition(
    instrument_tag = "EURGBP_1M",
    underlying_type = option.UnderlyingType.FX,
    underlying_definition = option.FxUnderlyingDefinition("EURGBP"),
    barrier_definition =  option.FxBarrierDefinition(barrier_mode = "American",
                                                      in_or_out = "Out",
                                                      up_or_down = "Down",
                                                      level = 0.85449),
    tenor = "1M",
    call_put = "Call",
    exercise_style = "EURO",
    strike = 0.8597,
    notional_amount = 1000000,
    notional_ccy = "EUR",
    pricing_parameters = option.PricingParameters(valuation_date = "2023-09-15T00:00:00Z")
).get_data()

response.data.df

here is the error got :

Can someone has a look ?

Thanks:

Note: The same error occurs when trying to use a FxBinaryDefinition

wasin.w

Hello @yaokoffi.kouassi

I can replicate the issue on Codebook (Refinitiv Data Library for Python version 1.2.0) and a standalone Jupyter Notebook using Refinitiv Data Library for Python version 1.3.1.

However, the latest version of the library is 1.4.0 (https://pypi.org/project/refinitiv-data/1.4.0/). I have tried it on a standalone Jupyter notebook with your code. It works fine.

It seems the issue has been fixed in version 1.4.0. If you (or the client) use a standalone Jupyter notebook, I highly recommend upgrading the library.

pip install refinitiv-data==1.4.0

yaokoffi.kouassi

Hello Wasin and thank you for your reply.

Unfortunately, the customer palns to use the script within Codebook.

Do we know whne Codebook will use the version 1.4.0?

Thanks

Jirapongse

@yaokoffi.kouassi

The workaround is using the following code.

import refinitiv.data as rd
import refinitiv.data.content.ipa.financial_contracts as rdf
from refinitiv.data.content.ipa.financial_contracts import option
 
rd.open_session()
 
response = option.Definition(
    instrument_tag = "EURGBP_1M",
    underlying_type = option.UnderlyingType.FX,
    underlying_definition = option.FxUnderlyingDefinition("EURGBP"),
    barrier_definition = {'barrierMode': 'American', 'inOrOut': 'Out', 'upOrDown': 'Down', 'level': 0.85449},
    tenor = "1M",
    call_put = "Call",
    exercise_style = "EURO",
    strike = 0.8597,
    notional_amount = 1000000,
    notional_ccy = "EUR",
    pricing_parameters = option.PricingParameters(valuation_date = "2023-09-15T00:00:00Z")
).get_data()
 
response.data.df

wasin.w

Hello @yaokoffi.kouassi

I suggest you contact the Codebook team about when will the app upgrade its libraries.

m.bunkowski

Hi @yaokoffi.kouassi

Customer can also use a direct request to an endpoint:

import refinitiv.data as rd
import pandas as pd
rd.open_session()

request = rd.delivery.endpoint_request.Definition(
                        method=rd.delivery.endpoint_request.RequestMethod.POST,
                        url='https://api.refinitiv.com/data/quantitative-analytics/v1/financial-contracts',
                        body_parameters={"universe": [
    {
      "instrumentType": "Option",
      "instrumentDefinition": {
        "instrumentTag": "EURGBP_1M",
        "underlyingType": "FX",
        "underlyingDefinition": {
          "fxCrossCode": "EURGBP"
        },
        "barrierDefinition": {
          "level": 0.85449,
          "barrierMode": "American",
          "inOrOut": "Out",
          "upOrDown": "Down"
        },
        "tenor": "1M",
        "callPut": "Call",
        "exerciseStyle": "EURO",
        "strike": 0.8597,
        "notionalAmount": 1000000,
        "notionalCcy": "EUR"
      },
      "pricingParameters": {
        "valuationDate": "2023-09-15"
      }
    }
  ]
})
response = request.get_data()
names = [i["name"] for i in response.data.raw["headers"]]
pd.DataFrame(response.data.raw["data"], columns=names)

yaokoffi.kouassi

Thanks guys for your reply.

The customer will use the script in Codebook.

I've contacted Codebook team to have a date for the release and I will inform the client.