Obtain duplicate previous_data on days when there is no data

jwlee08

It's been a long time.

I have a few questions.

First of all, it is an issue that occurred in the process of bringing gold prices from LBMA ( AM and PM).

If you look at this image below , there's no VALUE on the 23rd and 30th.

If you use the function GET_HISTORY you provided,

On the 23rd and 30th, each day has same price with previous day.

Is it able to controll the day which does not has value by putting some parameter or other ways?

On the other hands , CMZN0 which is the ZN value of LME. In this case, the date itself doesn't come up and the price doesn't come up on holidays.

Does each market have a different policy?

We would like to solve this problem using the GET_HISTORY function.

In the case of GETDATA, it is not easy to identify a definite date or bring past data.

In addition I just found out that someone says that by using get_timeseries can show up like image below which mean it shows blank value as<NA>

However in my case it does not have difference with using get_history.

Is it a difference with license?

I will wait for your kind reply.

And happy new year!

raksina.samasiri

Hi @jwlee08 ,

We apologize for this inconvenience. Ticket number 12068757 has been raised on your behalf and the technical support team is going to contact you to assist with this.

jwlee08

this is the one that shows <NA> on the day which does not have data

this is my case that shows duplicate data

nick.zincone

Hi @jwlee08

Can you confirm the version of the Refinitiv Data Library for Python you are using?

I was able to retrieve the expected data using the following API call:

rd.get_history(universe=['XAUFIXPM=', 'XAUFIXAM='], fields=['TRDPRC_1'])

I'm using CodeBook and the version installed is: 1.0.0b27.

Here is the result:

jwlee08

could you tell me how could I check the version of Refinitiv Data Library for Python

and if i needed how could i update the version?

jwlee08

I found out it is

rd : 1.0.0b24

jwlee08

I checked that I use the same version of library in codebook

but result is still same

jwlee08

I checked that I use the same version of library in codebook

but result is still same

nick.zincone

Hi @jwlee08

Thanks for confirming your version and your results.

I believe the issue you are facing has been resolved in a newer version of the library. If you are required to do this within CodeBook, I would suggest you contact Help & Support within the desktop via the Help menu option or by the function key (F1) and determine if your desktop can be updated within a newer version of the library. If you have your own Jupyter Notebook environment, you should be able to update your library version yourself via PyPi.

Trying the same command within my instance resulted in a correct response.

jwlee08

this is from codebook

this is from my visual studio code

I've updated the library to recent version but I see the same thing...

raksina.samasiri

Hi @jwlee08 ,

I cannot reproduce the issue. However, as an alternative way, can you please try to put the parameter 'raw_output=True' to get the output in Python dict format instead. With this parameter, I assume the date with NA will not be returned in the output as shown in the screenshot below.

Then this Python dict can be converted to the dataframe as you would like. Please also post the raw_output result here for further investigation.

timeseries_json = ek.get_timeseries(['XAUFIXPM=', 'XAUFIXAM='], raw_output=True)
timeseries_json

The help function in Python is going to return the available parameters and theirs description, including the description of raw_output parameter

help(ek.get_timeseries)

raksina.samasiri

Plus, this code is running in the Eikon Codebook app which uses the same version of the Eikon Data API Python library as yours and I still cannot reproduce an issue.

jwlee08

You said you can't reproduce this error.

Both the library version and the eikon desktop version are up to date, which means that there are other minor or unexpected problems, so can I ask for remote assistance?

nick.zincone

Hi @jwlee08

Please use the rd.get_history() instead - that is what I'm using.

jwlee08

I've received remote support from the developer in charge.

they decided to contact the team that can solve the issue now and contact you again, but can you tell me how long you'll contact me?

raksina.samasiri

Hi @jwlee08 ,

Now the issue is under investigation and the support team will keep you updated.

wasin.w

Hello @jwlee08

I checked case 12068757, the issue is resolved now.