Posting on behalf of a client

RogelynMayAgustin

1.      The only instance where we used the deprecated refinitiv.dataplatform library was the rdp.search function. I assume that with retiring the library I should switch respectively to either rd.discovery.search or ld.discovery.search functions, which are full equivalents of the rdp.search function. I also understand that with the rebranding to lseg, there is no value in keeping both the rd.discovery.search and ld.discovery.search functions and ld.discovery.search would be a preferred choice.

2.      Currently we use both eikon and refinitive.data libraries. The area where the use of the two libraries does not overlap is the use of eikon API to perform SCREENER tool searches and the use of the rd API to perform Advanced Search tool searches. Based on the links you provided, I understand that lseg library integrates API for both searches. For the Advanced Search tool, the syntaxis of the ld search function is exactly the same as for the rd search functions. Could you please provide links to the examples of using Screener() function with the lseg library to replicate the ek.get_data(“TR(“SCREEN(U(…)))”) function? (For eikon, we export the functions from the screener tool ‘as functions’ to identify the correct syntaxis. I wonder if there is also a simple way to build the function syntaxis for the lseg Screener() function).

3.      The links you provided indicate the api key is required for both desktop and platform sessions (Example.DataLibrary.Python/Examples/4-Session/EX-4.01.01-Sessions.ipynb at lseg-data-examples · LSEG-API-Samples/Example.DataLibrary.Python · GitHub). However, we currently use the api key only with the eikon library and the rd functions use only our credentials to open a session. From the ‘easy to use’ perspective, the preference is to connect automatically to the database without using the api key (the login information is confirmed once in a few month, so the connection is created in most cases automatically without requiring any action form the user). I wonder is there a ‘recommended’ approach to create a session (e.g., setting a configuration file with the api key, credentials, and log output for each user)? Or the trade off in this case is between the connection simplicity and the connection robustness (e.g., possibility to identify easily any connection problems).

jason.ramchandani01

@RogelynMayAgustin Thanks for your questions - so my comments are below:

1. Yes - you should use the LSEG Data Library functions over Refinitiv Data Library functions as they are the latest version. The is virtually no difference in rd.discovery.search and ld.discovery.search. You can confirm this for yourself using both libraries which are available in the Codebook app in our Workspace Product or by running the respective samples from github.
2. The screener is essentially unchanged if you compare eikon.get_data to ld.get_data. The ld library does introduce a screener object which can be used as follows:

import lseg.data as ld

from lseg.data.discovery import Screener

ld.open_session()

ue_large_growth = Screener('U(IN(indices(19262093/Thomson Reuters Europe/))/UNV:Public/), TR.CompanyMarketCap>=3604900000, TR.ValMoRegionRank>=70, TR.EQCountryListRank(Period=FY0)>=20, TR.IVPriceToIntrinsicValue<=0.85, TR.PriceMoMidTermComponent>=30,CURN=EUR')

print(list(ue_large_growth))

ld.get_data(ue_large_growth, ['BID', 'ASK', 'TR.Revenue'], parameters={'curn':'eur'})

3. So the recommended approach is to create an app key and store in the lseg.data-config.json file as indicated in the article here. There are other resources available there on how to do this - but session object also allows for creating a session by passing parameters directly. Please see this example notebook which details extensively all ways to create sessions and their nuances.

I hope this can help you and your client.

RogelynMayAgustin

Hi @jason.ramchandani01 , thank you for this. Really appreciate it!