Just my notes

AWS in Python with Boto3

Massimiliano Figini's photo
Massimiliano Figini
·

3 min read

To interact with AWS in Python, there is the Boto3 library.

1. AWS S3

S3 is the AWS Storage solution.

import boto3

# Generate the boto3 client for interacting with S3
s3 = boto3.client('s3', 
     region_name='eu-south-1',    #region where your resources are located
     aws_access_key_id=AWS_KEY,    #your aws key
     aws_secret_access_key=AWS_SECRET)     #your aws secret

# Create a bucket
s3.create_bucket(Bucket='new_bucket_230383')

# Delete a buckets
s3.delete_bucket(Bucket='new_bucket_230383')

# Get the list_buckets response
response = s3.list_buckets()
# Iterate over Buckets from .list_buckets() response
for bucket in response['Buckets']: 
     # Print the Name for each bucket
     print(bucket['Name'])

# Upload file to the bucket
s3.upload_file(Filename='m455y.csv',    #this is the local file path
     Bucket='new_bucket_230383',
     Key='m455y.csv')     #this is the name the object will have in s3

# Get object metadata
s3.head_object(Bucket='new_bucket_230383',
     Key='m455y.csv')

# Download file
s3.download_file(Filename='m455y.csv',    #this is the local path we want to download the file
     Bucket='new_bucket_230383',
     Key='m455y.csv')     #this is the name the object will have in s3

# 1. Access file directly
obj = s3.get_object(Bucket='new_bucket_230383',Key='m455y.csv')

# 2. Then read the file with pandas
pd.read_csv(obj['Body'])   # Body contains the data, the rest are the metadata

By default, files in S3 are private.
In the web console or with some boto methods, you can generate public file or you can temporary shared private files:

#Temporary shared file with get_object method
share_url = s3.generate_presigned_url(
     ClientMethod='get_object',
     ExpiresIn=3600,
     Params={'Bucket': 'new_bucket_230383', Key: 'm455y.csv'})

#Open Temporary shared files in pandas
pd.read_csv(share_url)

2. AWS SNS

SNS = Simple Notification Service.
With SNS you can send email, text message and push notifications.

# Initialize the SNS client
sns = boto3.client('sns',
     region_name='eu-central-1',
     aws_access_key_id=AWS_KEY,    #your aws key
     aws_secret_access_key=AWS_SECRET)     #your aws secret

# Create a topic
sns.create_topic(Name='my_alerts')['TopicArn']

# Listing Topics
sns.list_topics()['Topics']

# We can delete topic using its arn
sns.delete_topic(TopicArn='arn:aws:sns:eu-central-1:320333378981:my_alerts'

# Create a new SMS subscription
sns.subscribe(
     TopicArn = 'arn:aws:sns:eu-central-1:320333378981:my_alerts',
     Protocol = 'SMS',
     Endpoint = '+393331234567'

# Create a new email subscription
sns.subscribe(
     TopicArn = 'arn:aws:sns:eu-central-1:320333378981:my_alerts',
     Protocol = 'email',
     Endpoint = 'primodrudi@fakemail.com'

# List topic subscriptions
sns.list_subscriptions_by_topic(TopicArn = 'arn:aws:sns:eu-central-1:320333378981:my_alerts')

# List all topics subscriptions
sns.list_subscriptions()['Subscriptions']

# Delete a subscription
sns.unsubscribe(SubscriptionArn='arn:aws:sns:eu-central-1:320333378981:my_alerts:6b0e71bd-7e97-4d97-80ce-4a0994e55286')

# Publish a topic to send email and sms
sns.publish(
     TopicArn = 'arn:aws:sns:eu-central-1:320333378981:my_alerts',
     Message = 'Hello world!',
     Subject = 'Hi')    #subject is visible only for email

# Send a single SMS (don't need topic and subscriptions)
sns.publish(
     PhoneNumber = '+393331234567'
     Message = 'Hello world!')

3. AWS REKOGNITION

Computer vision API by AWS.

# Initialize the S3 client
s3 = boto3.client('s3',
     region_name='eu-central-1',
     aws_access_key_id=AWS_KEY,    #your aws key
     aws_secret_access_key=AWS_SECRET)     #your aws secret

# Upload a file
s3.upload_file(Filename='file1.jpg', Key='file1.jpg', Bucket='m455y-img')

# Initialize the Rekognition client
rekog = boto3.client('rekognition',
     region_name='eu-central-1',
     aws_access_key_id=AWS_KEY,
     aws_secret_access_key=AWS_SECRET)

# Object recognition
rekog.detect_labels(
     Image={'S3Object': {
          'Bucket': 'm455y-img',
          'Name': 'file1.jpg'},
     MaxLabels=10,     #optional max amount of labels to return
     MinConfidence=95})     #optional min accettable confidence in the match

# Text detenction
rekog.detect_labels(
     Image={'S3Object': {
          'Bucket': 'm455y-img',
          'Name': 'file1.jpg'}})

4. AWS TRANSLATE

Real-time translation service by AWS.

# Initialize the Translate client
translate = boto3.client('translate',
     region_name='eu-central-1',
     aws_access_key_id=AWS_KEY,    #your aws key
     aws_secret_access_key=AWS_SECRET)     #your aws secret

# Translate
translate.translate_text(
     Text='Ciao, come stai?',
     SourceLanguageCode='auto',
     TargetLanguageCode='en')['Translated_Text']

5. AWS COMPREHEND

Text comprehention service by AWS.

# Initialize the Comprehend client
comprehend = boto3.client('comprehend',
     region_name='eu-central-1',
     aws_access_key_id=AWS_KEY,    #your aws key
     aws_secret_access_key=AWS_SECRET)     #your aws secret

# Detect language
comprehend.detect_dominant_language(
     Text=''Ciao, come stai? Io bene, grazie.")

# Detect text sentiment
comprehend.detect_sentiment(
     Text=''This article is really good.",
     LanguageCode='en')
PythonAWSdata