Parsing deeply (multiple) nested JSON blocks with Python 3.6.8

user_15069291 0
Viewed 133

I've seen several related "nested json in Python" questions but the syntax for this Corona virus JSON data is giving me problems. Here's a sample:

{"recovered":524855,"list":[
    {"countrycode":"US","country":"United States of America","state":"South Carolina","latitude":"34.22333378","longitude":"-82.46170658","confirmed":15228,"deaths":568},
    {"countrycode":"US","country":"United States of America","state":"Louisiana","latitude":"30.2950649","longitude":"-92.41419698","confirmed":43612,"deaths":2957}
]}

If I just want to get to Louisiana, here's what I was trying:

import json
import requests

url = "https://covid19-data.p.api.com/us"

headers = {
    'x-api-key': "<api-key>",
    'x-api-host': "covid19-data.p.api.com"
    }

response = requests.request("GET", url, headers=headers)

coronastats = json.loads(response.text)

la_deaths = coronastats["list"][0]["countrycode"]["US"]["country"]["United States of America"]["state"]["Louisiana"]["deaths"]

print("Value: %s" % la_deaths)

I get: "TypeError: string indices must be integers"

This is obviously a list (I'm a detective and deduced that a variable named "list" might be a list) but the long key-value list is throwing me off (I think).

nested-lists python json
2 Answers

The problem is that once you get the first element of the list, you're left with only a depth-one dictionary. The data isn't as nested as you think it is. You're getting to a string quickly, and then trying to indice it using the US string, which raises the exception.

In [2]: data
Out[2]: 
{'recovered': 524855,
 'list': [{'countrycode': 'US',
   'country': 'United States of America',
   'state': 'South Carolina',
   'latitude': '34.22333378',
   'longitude': '-82.46170658',
   'confirmed': 15228,
   'deaths': 568},
  {'countrycode': 'US',
   'country': 'United States of America',
   'state': 'Louisiana',
   'latitude': '30.2950649',
   'longitude': '-92.41419698',
   'confirmed': 43612,
   'deaths': 2957}]}

In [3]: data["list"][0]
Out[3]: 
{'countrycode': 'US',
 'country': 'United States of America',
 'state': 'South Carolina',
 'latitude': '34.22333378',
 'longitude': '-82.46170658',
 'confirmed': 15228,
 'deaths': 568}

In [7]: data["list"][0]["countrycode"]
Out[7]: 'US'

In [8]: type(data["list"][0]["countrycode"])
Out[8]: str

In [9]: data["list"][0]["countrycode"]["asdf"]
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-9-cb9dbc39ef82> in <module>
----> 1 data["list"][0]["countrycode"]["asdf"]

TypeError: string indices must be integers

To get to a specific country what you want to do is to FIND the state in the list, for example with code:

In [14]: [f"{row['state']}: {row['deaths']} deaths. Wear a mask!" for row in data["list"] if row["state"] == "Louisiana"]
Out[14]: ['Louisiana: 2957 deaths. Wear a mask!']

You can also use filter, pandas, and a million other solutions to sort through a table.

user_41199060

Try this:

coronastats = json.loads(response.text)
[coronastats["list"][i]["deaths"] for i in range(len(coronastats["list"])) if coronastats["list"][i]["state"] == "Louisiana"]
user_122136120
Related