0xIbra / python-downloader-light

Downloader light

Lightweight python library that let's you download and process files concurrently.
This package was developed to allows serverless deployment.

Dependencies

Installed automatically with pip

• requests
• pysftp

• Installation
• Usage examples

Installation

pip install downloader-light

Usage examples

Download and upload files to AWS S3 For this to work, AWS CLI must be configured

from blackfeed.downloader import Downloader
from blackfeed.adapter.s3 import S3Adapter

queue = [
    {
        'url': 'https://www.example.com/path/to/image.jpg', # Required
        'destination': 'some/key/image.jpg' # S3 key - Required
    },{
        'url': 'https://www.example.com/path/to/image2.jpg',
        'destination': 'some/key/image2.jpg'
    }
]

downloader = Downloader(
    S3Adapter(bucket='bucketname'),
    multi=True, # If true, uploads files to images to S3 with multithreading
    stateless=False # If set to False, it generates and stores md5 hashes of files in a file
    state_id='flux_states' # name of the file where hashes will be stored (states.txt) not required
    bulksize=200 # Number of concurrent downloads
)
downloader.process(queue)
stats = downloader.get_stats() # Returns a dict with information about the process

Download files with states

Loading states can be useful if you don't want to re-download the same file twice.

from blackfeed.downloader import Downloader
from blackfeed.adapter.s3 import S3Adapter

queue = [
...
]

downloader = Downloader(
    S3Adapter(bucket='bucketname'),
    multi=True,
    stateless=False,
    state_id='filename'
)

# You can add a callback function if needed
# This function will be called after each bulk is processed
def callback(responses):
    # response: {
    #    'destination': destination of the file can be local or can be S3 key,
    #    'url': URL from where the file was downloaded,
    #    'httpcode': HTTP code returned by the server,
    #    'status': True|False,
    #    'content-type': Mime type of the downloaded resource Example: image/jpeg
    # }
    # responses: response[]

    pass # Your logic

downloader.set_callback(callback)

downloader.load_states('filename') # This will load states from "filename.txt"
downloader.process(queue)
stats = downloader.get_stats() # Statistics

ElasticDownloader

Let's you to download/retrieve files from FTP, SFTP and HTTP/S servers easily.

Examples

Downloading file from FTP

from blackfeed.elasticdownloader import ElasticDownloader

uri = 'ftp://user:[e-mail]/path/to/file.csv'

retriever = ElasticDownloader()
res = retriever.download(uri, localpath='/tmp/myfile.csv') # localfile is optional
# .download() function returns False if there was an error or return the local path of the downloaded file if it was a success.
print(res)

/tmp/myfile.csv

Retrieving binary content of file from FTP

from blackfeed.elasticdownloader import ElasticDownloader

uri = 'ftp://user:[e-mail]/path/to/file.csv'

retriever = ElasticDownloader()
res = retriever.retrieve(uri) # Return type: io.BytesIO | False

with open('/tmp/myfile.csv', 'wb') as f:
    f.write(res.getvalue())

ElasticDownloader can handle FTP, SFTP and HTTP URIs automatically. Use the method download to download file locally and use the retrieve method to get the binary content of a file.