How to pass a list of RICs through get_timeseries?

calice.l

Hi dev community!

I am absolutely new to this but would like to use a list from eiko PORT app with the get_timeseries function.

This works well with the get_data function

However, when I try to pass the list through get_timeseries function I get an error.

import eikon as ek

import pandas as pd

[2]:

ek.set_app_key('##########################')

[3]:

rics=(['Portfolio(500867312/*Total Asset List*/)'])

[4]:

print(rics)

['Portfolio(500867312/*Total Asset List*/)']

[5]:

dataset = ek.get_timeseries(rics,

                         fields='CLOSE',

                   start_date='2020.01.01',

                         end_date='2021.01.02',

                   normalize='True'

)

WARNING:pyeikon:Error with Portfolio(500867312/*Total Asset List*/): Invalid RIC ERROR:pyeikon:Portfolio(500867312/*Total Asset List*/): Invalid RIC | 

---------------------------------------------------------------------------
EikonError                                Traceback (most recent call last)
<ipython-input-5-af7c7776ca2a> in <module>
      7                    start_date='2020.01.01',
      8                          end_date='2021.01.02',
----> 9                    normalize='True'
     10 ) 
/opt/conda/lib/python3.7/site-packages/eikon/time_series.py in get_timeseries(rics, fields, start_date, end_date, interval, count, calendar, corax, normalize, raw_output, debug)
    200     if len(ts_status_errors) == len(ts_timeserie_data):
    201         logger.error(ts_error_messages)
--> 202         raise EikonError(-1, message=ts_error_messages)
    203     204     if raw_output: return ts_result 
EikonError: Error code -1 | Portfolio(500867312/*Total Asset List*/): Invalid RIC | 

I've also tried a workaround by first getting the RIC list with the get_data function and then converting the retrieved RIC list and pasting them through the get_timseries function without success.

import eikon as ek

[2]:

ek.set_app_key('############################')

[3]:

rics=(['Portfolio(500867312/*Total Asset List*/)'])

[4]:

dataset, err=ek.get_data(rics,['tr.PriceClose'])

[5]:

print(dataset)

         Instrument  Price Close 0           OREP.PA     323.4000 1           SAPG.DE     103.0200 2           0941.HK      53.6000 3          0593xq.L    1840.0000 4            SOON.S     248.2000 ..              ...          ... 96          CYBA.AS       5.4062 97     NL0000388619      50.2600 98   912828YM6=RRPS          NaN 99     NL0000009355          NaN 100          BRKb.N     251.5000 
[101 rows x 2 columns]

[6]:

import pandas as pd

[7]:

rics2 = pd.DataFrame(dataset.loc[:0,'Instrument'])

[8]:

print(rics2)

  Instrument 0    OREP.PA

[9]:

rics3 = rics2.astype(str)

[10]:

print (rics3)

  Instrument 0    OREP.PA

[11]:

s_array = rics3['Instrument'].to_numpy()

[12]:

print(s_array)

['OREP.PA']

[13]:

dataset = ek.get_timeseries(s_array,

                         fields='CLOSE',

                   start_date='2020.01.01',

                         end_date='2021.01.02',

                   normalize='True'

)

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-13-8e9e9cf810cb> in <module>
      7                    start_date='2020.01.01',
      8                          end_date='2021.01.02',
----> 9                    normalize='True'
     10 ) 
/opt/conda/lib/python3.7/site-packages/eikon/time_series.py in get_timeseries(rics, fields, start_date, end_date, interval, count, calendar, corax, normalize, raw_output, debug)
    111     112     # set the ric(s) in the payload
--> 113     check_for_string_or_list_of_strings(rics, 'rics')
    114     if is_string_type(rics):
    115         rics = [rics.strip()]

/opt/conda/lib/python3.7/site-packages/eikon/tools.py in check_for_string_or_list_of_strings(parameter, name)
     53 def check_for_string_or_list_of_strings(parameter, name):
     54     if type(parameter) != list and (not parameter or not is_string_type(parameter)):
---> 55         raise ValueError('The parameter {} should be a string or a list of string, found {}'.format(name,type(parameter)))
     56     if type(parameter) == list and not is_list_of_string(parameter):
     57         raise ValueError('All items in the parameter {} should be of data type string, found {}'.format(name,[type(v) for v in parameter]))

ValueError: The parameter rics should be a string or a list of string, found <class 'numpy.ndarray'>

Any help is much appreciated!

Ludwig

pierre.faurel

calice.l ,

You provided wrong type (numpy.ndarray) to get_timeseries().

Update your last example as below:

dataset, err=ek.get_data(rics,['tr.PriceClose']) rics2 = dataset["Instruments"] rics3 = rics2.tolist()

Then this request should work:

dataset = ek.get_timeseries(rics3, ...)

calice.l

Pierre, thanks a lot! For what it's worth, gave you my one and only point.