How do I pass values from a CSV to a JSON in Python?

Question

I tried two ways to read a CSV file and pass the values in the JSON value parameter, but it does not return any result, if I pass the direct values, it works.

Code 1:

import boto3
from datetime import datetime
import csv

client = boto3.client('cloudwatch', region_name='us-east-1')

with open('metrics_tcc.csv', mode='r') as csv_file:
csv_reader = csv.DictReader(csv_file)
for row in csv_reader:
    response = client.get_metric_data(
        MetricDataQueries=[
            {
                'Id': 'tcc',
                'MetricStat': {
                    'Metric': {
                        'Namespace': 'metrics.tcc',
                        'MetricName': 'responseTime',
                        "Dimensions": [
                    {
                        "Name": "isColdStart",
                        "Value": row['isColdStart']
                    },
                    {
                        "Name": "requestId",
                        "Value": row['requestId']
                    },
                    {
                        "Name": "interval",
                        "Value": row['interval']
                    },
                    {
                        "Name": "target",
                        "Value": row['target']
                    }
                        ]
                    },
                    'Period': 300,
                    'Stat': 'Maximum',
                    'Unit': 'Milliseconds'
                },
                    'ReturnData': True
            },
        ],
        StartTime=datetime(2018, 10, 8),
        EndTime=datetime(2018, 10, 9), 
    )   
    print(response['MetricDataResults'])

The exit is this JSON:

[{u'Timestamps': [], u'StatusCode': 'Complete', u'Values': [], u'Id': 'tcc', u'Label': 'responseTime'}]

In Values it was supposed to have come a value, however, it’s like you don’t understand the value passed in JSON, hence returns empty.

Second attempt:

import boto3
import numpy as np
from datetime import datetime

target, interval, isColdStart, requestId = 
np.loadtxt('metrics_tcc.csv', delimiter = ',', unpack = True, dtype = 
'str')
client = boto3.client('cloudwatch', region_name='us-east-1')

t = target[1]
i = interval[1]
bcs = isColdStart[1]
rq = requestId[1]

response = client.get_metric_data(
MetricDataQueries=[
    {
        'Id': 'tcc',
        'MetricStat': {
            'Metric': {
                'Namespace': 'metrics.tcc',
                'MetricName': 'responseTime',
                "Dimensions": [
            {
                "Name": "isColdStart",
                "Value": t
            },
            {
                "Name": "requestId",
                "Value": i
            },
            {
                "Name": "interval",
                "Value": bcs
            },
            {
                "Name": "target",
                "Value": rq
            }
                ]
            },
            'Period': 300,
            'Stat': 'Maximum',
            'Unit': 'Milliseconds'
        },
            'ReturnData': True
    },
],
        StartTime=datetime(2018, 10, 8),
        EndTime=datetime(2018, 10, 9), 
)
print(response['MetricDataResults'])

The exit is the same:

[{u'Timestamps': [], u'StatusCode': 'Complete', u'Values': [], u'Id': 'tcc', u'Label': 'responseTime'}]

But if I pass the direct value, it works:

import boto3
from datetime import datetime

client = boto3.client('cloudwatch', region_name='us-east-1')

response = client.get_metric_data(
MetricDataQueries=[
    {
        'Id': 'tcc',
        'MetricStat': {
            'Metric': {
                'Namespace': 'metrics.tcc',
                'MetricName': 'responseTime',
                "Dimensions": [
            {
                "Name": "isColdStart",
                "Value": "false"
            },
            {
                "Name": "requestId",
                "Value": "0"
            },
            {
                "Name": "interval",
                "Value": "660"
            },
            {
                "Name": "target",
                "Value": "when-will-i-coldstart-dev-system-under-test-256"
            }

                ]
            },
            'Period': 300,
            'Stat': 'Maximum',
            'Unit': 'Milliseconds'
        },
            'ReturnData': True
    },
],
StartTime=datetime(2018, 10, 8),
EndTime=datetime(2018, 10, 9), 
)


print(response['MetricDataResults'])

The exit is this JSON:

[{u'Timestamps': [datetime.datetime(2018, 10, 8, 20, 0, tzinfo=tzutc())], u'StatusCode': 'Complete', u'Values': [141.89045000006445], u'Id': 'tcc', u'Label': 'responseTime'}]

Note that Values is now right. Could you help me?

Answer 1

I’m not sure, but it seems the problem is in your csv; It is not being read correctly, perhaps because it is empty or because it is not using the correct delimiter. If you still have problems put part of csv file in question.