Skip to content

Latest commit

 

History

History
434 lines (327 loc) · 16.5 KB

howto.md

File metadata and controls

434 lines (327 loc) · 16.5 KB
Raw

How-to Guides

The howtos are goal-oriented guides that demonstrate how to solve a specific problem using smart_open.

How to Add a New Guide

The guides are code snippets compatible with Python's doctest module. Lines that start with >>> and ... are Python commands to run via the interpreter. Lines without the above prefixes are expected standard output from the commands. The doctest module runs the commands and ensures that their output matches the expected values.

>>> foo = 'bar'
>>> print(foo)
bar

Some tips:

  • Enclose the snippets with markdowns triple backticks to get free syntax highlighting
  • End your example with a blank line to let doctest know the triple backticks aren't part of the example

Finally, ensure all the guides still work by running:

python -m doctest howto.md

The above command shouldn't print anything to standard output/error and return zero, provided your local environment is set up correctly:

  • you have a working Internet connection
  • localstack is running, and the mybucket S3 bucket has been created
  • the GITHUB_TOKEN environment variable is set to a valid access token

How to Read/Write Zip Files

smart_open does not support reading/writing zip files out of the box. However, you can easily integrate smart_open with the standard library's zipfile module:

  • smart_open handles the I/O
  • zipfile handles the compression, decompression, and file member lookup

Reading example:

>>> from smart_open import open
>>> import zipfile
>>> with open('sampledata/hello.zip', 'rb') as fin:
...     with zipfile.ZipFile(fin) as zip:
...         for info in zip.infolist():
...             file_bytes = zip.read(info.filename)
...             print('%r: %r' % (info.filename, file_bytes.decode('utf-8')))
'hello/': ''
'hello/en.txt': 'hello world!\n'
'hello/ru.txt': 'здравствуй, мир!\n'

Writing example:

>>> from smart_open import open
>>> import os
>>> import tempfile
>>> import zipfile
>>> tmp = tempfile.NamedTemporaryFile(prefix='smart_open-howto-', suffix='.zip', delete=False)
>>> with open(tmp.name, 'wb') as fout:
... 	with zipfile.ZipFile(fout, 'w') as zip:
...			zip.writestr('hello/en.txt', 'hello world!\n')
...			zip.writestr('hello/ru.txt', 'здравствуй, мир!\n')
>>> os.unlink(tmp.name)  # comment this line to keep the file for later

How to access S3 anonymously

The boto3 library that smart_open uses for accessing S3 signs each request using your boto3 credentials. If you'd like to access S3 without using an S3 account, then you need disable this signing mechanism.

>>> import boto3
>>> import botocore
>>> import botocore.client
>>> from smart_open import open
>>> config = botocore.client.Config(signature_version=botocore.UNSIGNED)
>>> params = {'client': boto3.client('s3', config=config)}
>>> with open('s3://commoncrawl/robots.txt', transport_params=params) as fin:
...    fin.readline()
'User-Agent: *\n'

How to Access S3 Object Properties

When working with AWS S3, you may want to look beyond the abstraction provided by smart_open and communicate with boto3 directly in order to satisfy your use case.

For example:

  • Access the object's properties, such as the content type, timestamp of the last change, etc.
  • Access version information for the object (versioned buckets only)
  • Copy the object to another location
  • Apply an ACL to the object
  • and anything else specified in the boto3 S3 Object API.

To enable such use cases, the file-like objects returned by smart_open have a special to_boto3 method. This returns a boto3.s3.Object that you can work with directly. For example, let's get the content type of a publicly available file:

>>> import boto3
>>> from smart_open import open
>>> resource = boto3.resource('s3')  # Pass additional resource parameters here
>>> with open('s3://commoncrawl/robots.txt') as fin:
...    print(fin.readline().rstrip())
...    boto3_s3_object = fin.to_boto3(resource)
...    print(repr(boto3_s3_object))
...    print(boto3_s3_object.content_type)  # Using the boto3 API here
User-Agent: *
s3.Object(bucket_name='commoncrawl', key='robots.txt')
text/plain

This works only when reading and writing via S3.

How to Access a Specific Version of an S3 Object

The version_id transport parameter enables you to get the desired version of the object from an S3 bucket.

.. Important:: S3 disables version control by default. Before using the version_id parameter, you must explicitly enable version control for your S3 bucket. Read https://docs.aws.amazon.com/AmazonS3/latest/dev/Versioning.html for details.

>>> import boto3
>>> from smart_open import open
>>> versions = ['KiQpZPsKI5Dm2oJZy_RzskTOtl2snjBg', 'N0GJcE3TQCKtkaS.gF.MUBZS85Gs3hzn']
>>> for v in versions:
...     with open('s3://smart-open-versioned/demo.txt', transport_params={'version_id': v}) as fin:
...         print(v, repr(fin.read()))
KiQpZPsKI5Dm2oJZy_RzskTOtl2snjBg 'second version\n'
N0GJcE3TQCKtkaS.gF.MUBZS85Gs3hzn 'first version\n'

>>> # If you don't specify a version, smart_open will read the most recent one
>>> with open('s3://smart-open-versioned/demo.txt') as fin:
...     print(repr(fin.read()))
'second version\n'

This works only when reading via S3.

How to Access the Underlying boto3 Object

At some stage in your workflow, you may opt to work with boto3 directly. You can do this by calling to the to_boto3() method. You can then interact with the object using the boto3 API:

>>> import boto3
>>> from smart_open import open
>>> resource = boto3.resource('s3')  # Pass additional resource parameters here
>>> with open('s3://commoncrawl/robots.txt') as fin:
...     boto3_object = fin.to_boto3(resource)
...     print(boto3_object)
...     print(boto3_object.get()['LastModified'])
s3.Object(bucket_name='commoncrawl', key='robots.txt')
2016-05-21 18:17:43+00:00

This works only when reading and writing via S3.

For versioned objects, the returned object will be slightly different:

>>> from smart_open import open
>>> resource = boto3.resource('s3')
>>> params = {'version_id': 'KiQpZPsKI5Dm2oJZy_RzskTOtl2snjBg'}
>>> with open('s3://smart-open-versioned/demo.txt', transport_params=params) as fin:
...     print(fin.to_boto3(resource))
s3.ObjectVersion(bucket_name='smart-open-versioned', object_key='demo.txt', id='KiQpZPsKI5Dm2oJZy_RzskTOtl2snjBg')

How to Read from S3 Efficiently

Under the covers, smart_open uses the boto3 client API to read from S3. By default, calling smart_open.open with an S3 URL will create its own boto3 client. These are expensive operations: they require both CPU time to construct the objects from a low-level API definition, and memory to store the objects once they have been created. It is possible to save both CPU time and memory by sharing the same resource across multiple smart_open.open calls, for example:

>>> import boto3
>>> from smart_open import open
>>> tp = {'client': boto3.client('s3')}
>>> for month in (1, 2, 3):
...     url = 's3://nyc-tlc/trip data/yellow_tripdata_2020-%02d.csv' % month
...     with open(url, transport_params=tp) as fin:
...         _ = fin.readline()  # skip CSV header
...         print(fin.readline().strip())
1,2020-01-01 00:28:15,2020-01-01 00:33:03,1,1.20,1,N,238,239,1,6,3,0.5,1.47,0,0.3,11.27,2.5
1,2020-02-01 00:17:35,2020-02-01 00:30:32,1,2.60,1,N,145,7,1,11,0.5,0.5,2.45,0,0.3,14.75,0
1,2020-03-01 00:31:13,2020-03-01 01:01:42,1,4.70,1,N,88,255,1,22,3,0.5,2,0,0.3,27.8,2.5

Clients are thread-safe and multiprocess-safe, so you may share them between other threads and subprocesses.

How to Write to S3 Efficiently

By default, smart_open buffers the most recent part of a multipart upload in memory. The default part size is 50MB. If you're concerned about memory usage, then you have two options. The first option is to use smaller part sizes (e.g. 5MB, the lowest value permitted by AWS):

import boto3
from smart_open import open
tp = {'min_part_size': 5 * 1024**2}
with open('s3://bucket/key', 'w', transport_params=tp) as fout:
    fout.write(lots_of_data)

This will split your upload into smaller parts. Be warned that AWS enforces a limit of a maximum of 10,000 parts per upload.

The second option is to use a temporary file as a buffer instead.

import boto3
from smart_open import open
with tempfile.NamedTemporaryFile() as tmp:
    tp = {'writebuffer': tmp}
    with open('s3://bucket/key', 'w', transport_params=tp) as fout:
        fout.write(lots_of_data)

This option reduces memory usage at the expense of additional disk I/O (writing to and reading from a hard disk is slower).

How to Specify the Request Payer (S3 only)

Some public buckets require you to pay for S3 requests for the data in the bucket. This relieves the bucket owner of the data transfer costs, and spreads them among the consumers of the data.

To access such buckets, you need to pass some special transport parameters:

>>> from smart_open import open
>>> params = {'client_kwargs': {'S3.Client.get_object': {'RequestPayer': 'requester'}}}
>>> with open('s3://arxiv/pdf/arXiv_pdf_manifest.xml', transport_params=params) as fin:
...    print(fin.readline())
<?xml version='1.0' standalone='yes'?>
<BLANKLINE>

This works only when reading and writing via S3.

How to Make S3 I/O Robust to Network Errors

Boto3 has a built-in mechanism for retrying after a recoverable error. You can fine-tune it using several ways:

Pre-configuring a boto3 client and then passing the client to smart_open

>>> import boto3
>>> import botocore.config
>>> import smart_open
>>> config = botocore.config.Config(retries={'mode': 'standard'})
>>> client = boto3.client('s3', config=config)
>>> tp = {'client': client}
>>> with open('s3://commoncrawl/robots.txt', transport_params=tp) as fin:
...     print(fin.readline())
User-Agent: *
<BLANKLINE>

To verify your settings have effect:

import logging
logging.getLogger('smart_open.s3').setLevel(logging.DEBUG)

and check the log output of your code.

How to Pass Additional Parameters to boto3

boto3 is a highly configurable library, and each function call accepts many optional parameters. smart_open does not attempt to replicate this behavior, since most of these parameters often do not influence the behavior of smart_open itself. Instead, smart_open offers the caller of the function to pass additional parameters as necessary:

>>> import boto3
>>> from smart_open import open
>>> transport_params = {'client_kwargs': {'S3.Client.get_object': {'RequestPayer': 'requester'}}}
>>> with open('s3://arxiv/pdf/arXiv_pdf_manifest.xml', transport_params=params) as fin:
...    print(fin.readline())
<?xml version='1.0' standalone='yes'?>
<BLANKLINE>

The above example influences how the S3.Client.get_object function gets called by smart_open when reading the specified URL. More specifically, the RequestPayer parameter will be set to requester for each call. Influential functions include:

  • S3.Client (the initializer function)
  • S3.Client.abort_multipart_upload
  • S3.Client.complete_multipart_upload
  • S3.Client.create_multipart_upload
  • S3.Client.get_object
  • S3.Client.head_bucket
  • S3.Client.put_object
  • S3.Client.upload_part

If you choose to pass additional parameters, keep the following in mind:

  1. Study the boto3 client API and ensure the function and parameters are valid.
  2. Study the code for the smart_open.s3 submodule and ensure smart_open is actually calling the function you're passing additional parameters for.

Finally, in some cases, it's possible to work directly with boto3 without going through smart_open. For example, setting the ACL for an object is possible after the object is created (with boto3), as opposed to at creation time (with smart_open). More specifically, here's the direct method:

import boto3
import smart_open
with open('s3://bucket/key', 'wb') as fout:
    fout.write(b'hello world!')
client = boto3.client('s3')
client.put_object_acl(ACL=acl_as_string)

Here's the same code that passes the above parameter via smart_open:

import smart_open
tp = {'client_kwargs': {'S3.Client.create_multipart_upload': {'ACL': acl_as_string}}}
with open('s3://bucket/key', 'wb', transport_params=tp) as fout:
    fout.write(b'hello world!')

If passing everything via smart_open feels awkward, try passing part of the parameters directly to boto3.

How to Read from Github API

The Github API allows users access to, among many other things, read files from repositories that you have access to. Below is an example for how users can read a file with smart_open. For more info, see the Github API documentation.

>>> import base64
>>> import gzip
>>> import json
>>> import os
>>> from smart_open import open
>>> owner, repo, path = "RaRe-Technologies", "smart_open", "howto.md"
>>> github_token = os.environ['GITHUB_TOKEN']
>>> url = f"https://api.github.com/repos/{owner}/{repo}/contents/{path}"
>>> params = {"headers" : {"Authorization" : "Bearer " + github_token}}
>>> with open(url, 'rb', transport_params=params) as fin:
...     response = json.loads(gzip.decompress(fin.read()))
>>> response["path"]
'howto.md'

Note: If you are accessing a file in a Github Enterprise org, you will likely have a different base dns than the https://api.github.com/ in the example.

How to Read/Write from localstack

localstack is a convenient test framework for developing cloud apps. You run it locally on your machine and behaves almost identically to the real AWS. This makes it useful for testing your code offline, without requiring you to set up mocks or test harnesses.

First, install localstack and start it:

$ pip install localstack
$ localstack start

The start command is blocking, so you'll need to run it in a separate terminal session or run it in the background. Before we can read/write, we'll need to create a bucket:

$ aws --endpoint-url http://localhost:4566 s3api create-bucket --bucket mybucket

where http://localhost:4566 is the default host/port that localstack uses to listen for requests.

You can now read/write to the bucket the same way you would to a real S3 bucket:

>>> import boto3
>>> from smart_open import open
>>> client = boto3.client('s3', endpoint_url='http://localhost:4566')
>>> tparams = {'client': client}
>>> with open('s3://mybucket/hello.txt', 'wt', transport_params=tparams) as fout:
...     _ = fout.write('hello world!')
>>> with open('s3://mybucket/hello.txt', 'rt', transport_params=tparams) as fin:
...     fin.read()
'hello world!'

You can also access it using the CLI:

$ aws --endpoint-url http://localhost:4566 s3 ls s3://mybucket/
2020-12-09 15:56:22         12 hello.txt

How to Download a Whole Directory From Google Cloud

Object storage providers generally don't provide real directories, and instead emulate them using object name patterns (see here for an explanation). To download all files in a directory you can do this:

>>> from google.cloud import storage
>>> from smart_open import open
>>> client = storage.Client()
>>> bucket_name = "gcp-public-data-landsat"
>>> prefix = "LC08/01/044/034/LC08_L1GT_044034_20130330_20170310_01_T2/"
>>> for blob in client.list_blobs(client.get_bucket(bucket_name), prefix=prefix):
...      with open(f"gs://{bucket_name}/{blob.name}") as f:
...          print(f.name)
...          break # just show the first iteration for the test
LC08/01/044/034/LC08_L1GT_044034_20130330_20170310_01_T2/LC08_L1GT_044034_20130330_20170310_01_T2_ANG.txt