Python pass output of statement for evaluation in another statement

Question

Trying to take output of a code and pass it to another statement.

t = ", ".join(DTYPE)
output = ('data', list), ('freq', str)

Would like to using "output" and pass into the following:

DF_TYPE = numpy.dtype([t)])
DF = numpy.empty(0, dtype=DF_TYPE)

That is, evaluate

DF_TYPE = numpy.dtype([('data', list), ('freq', str)])
DF = numpy.empty(0, dtype=DF_TYPE)

Error received:

Exception has occurred: TypeError
data type not understood

Sample data:

JSON_DATA = {
  "data": [
    {
      "Members": 29,
      "period": "2020-06-30",
      "TCOC": 5880070,
    },
    {
      "Members": 21,
      "period": "2020-03-31",
      "TCOC": 5689074,
    }
  ],
  "freq": "quarterly"
}

Have tried using exec() and eval() to no avail.

P. Mobley · Accepted Answer · 2020-05-08 01:46:08Z

1

Apparently I need 50 rep to comment. But could you provide additional information like toy data or any error messages? Thanks

Also, I noticed you refer to numpy as numpy and np. The system will only recognize whatever your import statement says so:

import numpy is referred to as numpy.whatever and import numpy as np is referred to as np.whatever

edited May 8, 2020 at 1:46

answered May 8, 2020 at 1:40

P. Mobley

216 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

hpaulj · Accepted Answer · 2020-05-08 03:13:44Z

With a few corrections, these structured array creations work:

In [244]: dt = np.dtype([('data',object), ('freq', 'U10')])                                            
In [245]: x = np.empty(3, dtype=dt)                                                                    
In [246]: x                                                                                            
Out[246]: 
array([(None, ''), (None, ''), (None, '')],
      dtype=[('data', 'O'), ('freq', '<U10')])

or data as a list of tuples:

In [247]: x = np.array([([1,2,3], 'quarterly'), ([4,5], 'yearly')], dt)                                
In [248]: x                                                                                            
Out[248]: 
array([(list([1, 2, 3]), 'quarterly'), (list([4, 5]), 'yearly')],
      dtype=[('data', 'O'), ('freq', '<U10')])
In [249]: x['data']                                                                                    
Out[249]: array([list([1, 2, 3]), list([4, 5])], dtype=object)

and with the JSON_DATA dictionary:

In [251]: JSON_DATA['data']                                                                            
Out[251]: 
[{'Members': 29, 'period': '2020-06-30', 'TCOC': 5880070},
 {'Members': 21, 'period': '2020-03-31', 'TCOC': 5689074}]

In [254]: x = np.empty(2, dt)                                                                          
In [255]: x['data'] = JSON_DATA['data']                                                                
In [256]: x['freq'] = JSON_DATA['freq']                                                                

In [257]: x                                                                                            
Out[257]: 
array([({'Members': 29, 'period': '2020-06-30', 'TCOC': 5880070}, 'quarterly'),
       ({'Members': 21, 'period': '2020-03-31', 'TCOC': 5689074}, 'quarterly')],
      dtype=[('data', 'O'), ('freq', '<U10')])
In [258]: x['data']                                                                                    
Out[258]: 
array([{'Members': 29, 'period': '2020-06-30', 'TCOC': 5880070},
       {'Members': 21, 'period': '2020-03-31', 'TCOC': 5689074}],
      dtype=object)

An element of the 'data' field is actually a dictionary:

In [259]: x['data'][0]                                                                                 
Out[259]: {'Members': 29, 'period': '2020-06-30', 'TCOC': 5880070}

Thanks for answering. I am trying to make such that I don't have to enter the values of 'data' or 'freq' manually. These values are outputs form the join statement. I want to pass them into another line of code and run it

Collectives™ on Stack Overflow

Python pass output of statement for evaluation in another statement

2 Answers 2

Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related