multipart/form-data Encoder¶

The main attraction is a streaming multipart form-data object, MultipartEncoder. Its API looks like this:

from requests_toolbelt import MultipartEncoder
import requests

m = MultipartEncoder(
    fields={'field0': 'value', 'field1': 'value',
            'field2': ('filename', open('file.py', 'rb'), 'text/plain')}
    )

r = requests.post('http://httpbin.org/post', data=m,
                  headers={'Content-Type': m.content_type})

You can also use multipart/form-data encoding for requests that don’t require files:

from requests_toolbelt import MultipartEncoder
import requests

m = MultipartEncoder(fields={'field0': 'value', 'field1': 'value'})

r = requests.post('http://httpbin.org/post', data=m,
                  headers={'Content-Type': m.content_type})

Or, you can just create the string and examine the data:

# Assuming `m` is one of the above
m.to_string()  # Always returns unicode

User-Agent constructor¶

You can easily construct a requests-style User-Agent string:

from requests_toolbelt import user_agent

headers = {
    'User-Agent': user_agent('my_package', '0.0.1')
    }

r = requests.get('https://api.github.com/users', headers=headers)

SSLAdapter¶

The SSLAdapter was originally published on Cory Benfield’s blog. This adapter allows the user to choose one of the SSL protocols made available in Python’s ssl module for outgoing HTTPS connections:

from requests_toolbelt import SSLAdapter
import requests
import ssl

s = requests.Session()
s.mount('https://', SSLAdapter(ssl.PROTOCOL_TLSv1))

Known Issues¶

On Python 3.3.0 and 3.3.1, the standard library’s http module will fail when passing an instance of the MultipartEncoder. This is fixed in later minor releases of Python 3.3. Please consider upgrading to a later minor version or Python 3.4. There is absolutely nothing this library can do to work around that bug.

FingerprintAdapter¶

By default, requests will validate a server’s certificate to ensure a connection is secure. In addition to this, the user can provide a fingerprint of the certificate they’re expecting to receive. Unfortunately, the requests API does not support this fairly rare use-case. When a user needs this extra validation, they should use the FingerprintAdapter class to perform the validation.

class requests_toolbelt.adapters.fingerprint.FingerprintAdapter(fingerprint, **kwargs)¶

A HTTPS Adapter for Python Requests that verifies certificate fingerprints, instead of certificate hostnames.

Example usage:

import requests
import ssl
from requests_toolbelt.adapters.fingerprint import FingerprintAdapter

twitter_fingerprint = '...'
s = requests.Session()
s.mount(
    'https://twitter.com',
    FingerprintAdapter(twitter_fingerprint)
)

The fingerprint should be provided as a hexadecimal string, optionally containing colons.

SSLAdapter¶

The SSLAdapter is the canonical implementation of the adapter proposed on Cory Benfield’s blog, here. This adapter allows the user to choose one of the SSL/TLS protocols made available in Python’s ssl module for outgoing HTTPS connections.

In principle, this shouldn’t be necessary: compliant SSL servers should be able to negotiate the required SSL version. In practice there have been bugs in some versions of OpenSSL that mean that this negotiation doesn’t go as planned. It can be useful to be able to simply plug in a Transport Adapter that can paste over the problem.

For example, suppose you’re having difficulty with the server that provides TLS for GitHub. You can work around it by using the following code:

from requests_toolbelt.adapters.ssl import SSLAdapter

import requests
import ssl

s = requests.Session()
s.mount('https://github.com/', SSLAdapter(ssl.PROTOCOL_TLSv1))

Any future requests to GitHub made through that adapter will automatically attempt to negotiate TLSv1, and hopefully will succeed.

class requests_toolbelt.adapters.ssl.SSLAdapter(ssl_version=None, **kwargs)¶

A HTTPS Adapter for Python Requests that allows the choice of the SSL/TLS version negotiated by Requests. This can be used either to enforce the choice of high-security TLS versions (where supported), or to work around misbehaving servers that fail to correctly negotiate the default TLS version being offered.

Example usage:

>>> import requests
>>> import ssl
>>> from requests_toolbelt import SSLAdapter
>>> s = requests.Session()
>>> s.mount('https://', SSLAdapter(ssl.PROTOCOL_TLSv1))

You can replace the chosen protocol with any that are available in the default Python SSL module. All subsequent requests that match the adapter prefix will use the chosen SSL version instead of the default.

SourceAddressAdapter¶

The SourceAddressAdapter allows a user to specify a source address for their connnection.

class requests_toolbelt.adapters.source.SourceAddressAdapter(source_address, **kwargs)¶

A Source Address Adapter for Python Requests that enables you to choose the local address to bind to. This allows you to send your HTTP requests from a specific interface and IP address.

Example usage:

import requests
from requests_toolbelt.adapters.source import SourceAddressAdapter

s = requests.Session()
s.mount('http://', SourceAddressAdapter('10.10.10.10'))

SocketOptionsAdapter¶

The SocketOptionsAdapter allows a user to pass specific options to be set on created sockets when constructing the Adapter without subclassing. The adapter takes advantage of urllib3‘s support for setting arbitrary socket options for each urllib3.connection.HTTPConnection (and HTTPSConnection).

To pass socket options, you need to send a list of three-item tuples. For example, requests and urllib3 disable Nagle’s Algorithm by default. If you need to re-enable it, you would do the following:

import socket
import requests
from requests_toolbelt.adapters.socket_options import SocketOptionsAdapter

nagles = [(socket.IPPROTO_TCP, socket.TCP_NODELAY, 0)]
session = requests.Session()
for scheme in session.adapters.keys():
    session.mount(scheme, SocketOptionsAdapter(socket_options=nagles))

This would re-enable Nagle’s Algorithm for all http:// and https:// connections made with that session.

class requests_toolbelt.adapters.socket_options.SocketOptionsAdapter(**kwargs)¶

An adapter for requests that allows users to specify socket options.

Since version 2.4.0 of requests, it is possible to specify a custom list of socket options that need to be set before establishing the connection.

Example usage:

>>> import socket
>>> import requests
>>> from requests_toolbelt.adapters import socket_options
>>> s = requests.Session()
>>> opts = [(socket.IPPROTO_TCP, socket.TCP_NODELAY, 0)]
>>> adapter = socket_options.SocketOptionsAdapter(socket_options=opts)
>>> s.mount('http://', adapter)

You can also take advantage of the list of default options on this class to keep using the original options in addition to your custom options. In that case, opts might look like:

>>> opts = socket_options.SocketOptionsAdapter.default_options + opts

TCPKeepAliveAdapter¶

The TCPKeepAliveAdapter allows a user to pass specific keep-alive related options as keyword parameters as well as arbitrary socket options.

An example usage of the adapter:

import requests
from requests_toolbelt.adapter.socket_options import TCPKeepAliveAdapter

session = requests.Session()
keep_alive = TCPKeepAliveAdapter(idle=120, count=20, interval=30)
session.mount('https://region-a.geo-1.compute.hpcloudsvc.com', keep_alive)
session.post('https://region-a.geo-1.compute.hpcloudsvc.com/v2/1234abcdef/servers',
             # ...
             )

In this case we know that creating a server on HP Public Cloud can cause requests to hang without using TCP Keep-Alive. So we mount the adapter specifically for that domain, instead of adding it to every https:// and http:// request.

class requests_toolbelt.adapters.socket_options.TCPKeepAliveAdapter(**kwargs)¶

An adapter for requests that turns on TCP Keep-Alive by default.

The adapter sets 4 socket options:

• SOL_SOCKET SO_KEEPALIVE - This turns on TCP Keep-Alive
• IPPROTO_TCP TCP_KEEPINTVL 20 - Sets the keep alive interval
• IPPROTO_TCP TCP_KEEPCNT 5 - Sets the number of keep alive probes
• IPPROTO_TCP TCP_KEEPIDLE 60 - Sets the keep alive time if the socket library has the TCP_KEEPIDLE constant

The latter three can be overridden by keyword arguments (respectively):

• idle
• interval
• count

You can use this adapter like so:

>>> from requests_toolbelt.adapters import socket_options
>>> tcp = socket_options.TCPKeepAliveAdapter(idle=120, interval=10)
>>> s = requests.Session()
>>> s.mount('http://', tcp)

AuthHandler¶

The AuthHandler is a way of using a single session with multiple websites that require authentication. If you know what websites require a certain kind of authentication and what your credentials are.

Take for example a session that needs to authenticate to GitHub’s API and GitLab’s API, you would set up and use your AuthHandler like so:

import requests
from requests_toolbelt.auth.handler import AuthHandler

def gitlab_auth(request):
    request.headers['PRIVATE-TOKEN'] = 'asecrettoken'

handler = AuthHandler({
    'https://api.github.com': ('sigmavirus24', 'apassword'),
    'https://gitlab.com': gitlab_auth,
})

session = requests.Session()
session.auth = handler
r = session.get('https://api.github.com/user')
# assert r.ok
r2 = session.get('https://gitlab.com/api/v3/projects')
# assert r2.ok

class requests_toolbelt.auth.handler.AuthHandler(strategies)¶

The AuthHandler object takes a dictionary of domains paired with authentication strategies and will use this to determine which credentials to use when making a request. For example, you could do the following:

from requests import HTTPDigestAuth
from requests_toolbelt.auth.handler import AuthHandler

import requests

auth = AuthHandler({
    'https://api.github.com': ('sigmavirus24', 'fakepassword'),
    'https://example.com': HTTPDigestAuth('username', 'password')
})

r = requests.get('https://api.github.com/user', auth=auth)
# => <Response [200]>
r = requests.get('https://example.com/some/path', auth=auth)
# => <Response [200]>

s = requests.Session()
s.auth = auth
r = s.get('https://api.github.com/user')
# => <Response [200]>

Warning

requests.auth.HTTPDigestAuth is not yet thread-safe. If you use AuthHandler across multiple threads you should instantiate a new AuthHandler for each thread with a new HTTPDigestAuth instance for each thread.

add_strategy(domain, strategy)¶

Add a new domain and authentication strategy.

Parameters:	domain (str) – The domain you wish to match against. For example: 'https://api.github.com' strategy (str) – The authentication strategy you wish to use for that domain. For example: ('username', 'password') or requests.HTTPDigestAuth('username', 'password')

a = AuthHandler({})
a.add_strategy('https://api.github.com', ('username', 'password'))

get_strategy_for(url)¶

Retrieve the authentication strategy for a specified URL.

Parameters:	url (str) – The full URL you will be making a request against. For example, 'https://api.github.com/user'
Returns:	Callable that adds authentication to a request.

import requests
a = AuthHandler({'example.com', ('foo', 'bar')})
strategy = a.get_strategy_for('http://example.com/example')
assert isinstance(strategy, requests.auth.HTTPBasicAuth)

remove_strategy(domain)¶

Remove the domain and strategy from the collection of strategies.

Parameters:	domain (str) – The domain you wish remove. For example, 'https://api.github.com'.

a = AuthHandler({'example.com', ('foo', 'bar')})
a.remove_strategy('example.com')
assert a.strategies == {}

GuessAuth¶

The GuessAuth authentication class automatically detects whether to use basic auth or digest auth:

import requests
from requests_toolbelt.auth import GuessAuth

requests.get('http://httpbin.org/basic-auth/user/passwd',
             auth=GuessAuth('user', 'passwd'))
requests.get('http://httpbin.org/digest-auth/auth/user/passwd',
             auth=GuessAuth('user', 'passwd'))

Detection of the auth type is done via the WWW-Authenticate header sent by the server. This requires an additional request in case of basic auth, as usually basic auth is sent preemptively. If the server didn’t explicitly require authentication, no credentials are sent.

class requests_toolbelt.auth.guess.GuessAuth(username, password)¶

Guesses the auth type by the WWW-Authentication header.

HTTPProxyDigestAuth¶

The HTTPProxyDigestAuth use digest authentication between the client and the proxy.

import requests
from requests_toolbelt.auth.http_proxy_digest import HTTPProxyDigestAuth


proxies = {
    "http": "http://PROXYSERVER:PROXYPORT",
    "https": "https://PROXYSERVER:PROXYPORT",
}
url = "https://toolbelt.readthedocs.org/"
auth = HTTPProxyDigestAuth("USERNAME", "PASSWORD")
requests.get(url, proxies=proxies, auth=auth)

Program would raise error if the username or password is rejected by the proxy.

class requests_toolbelt.auth.http_proxy_digest.HTTPProxyDigestAuth(*args, **kwargs)¶

HTTP digest authentication between proxy

Parameters:	stale_rejects (int) – The number of rejects indicate that: the client may wish to simply retry the request with a new encrypted response, without reprompting the user for a new username and password. i.e., retry build_digest_header

Streaming Multipart Data Encoder¶

Requests has support for multipart uploads, but the API means that using that functionality to build exactly the Multipart upload you want can be difficult or impossible. Additionally, when using Requests’ Multipart upload functionality all the data must be read into memory before being sent to the server. In extreme cases, this can make it impossible to send a file as part of a multipart/form-data upload.

The toolbelt contains a class that allows you to build multipart request bodies in exactly the format you need, and to avoid reading files into memory. An example of how to use it is like this:

import requests
from requests_toolbelt.multipart.encoder import MultipartEncoder

m = MultipartEncoder(
    fields={'field0': 'value', 'field1': 'value',
            'field2': ('filename', open('file.py', 'rb'), 'text/plain')}
    )

r = requests.post('http://httpbin.org/post', data=m,
                  headers={'Content-Type': m.content_type})

The MultipartEncoder has the .to_string() convenience method, as well. This method renders the multipart body into a string. This is useful when developing your code, allowing you to confirm that the multipart body has the form you expect before you send it on.

The toolbelt also provides a way to monitor your streaming uploads with the MultipartEncoderMonitor.

class requests_toolbelt.multipart.encoder.MultipartEncoder(fields, boundary=None, encoding='utf-8')¶

The MultipartEncoder oject is a generic interface to the engine that will create a multipart/form-data body for you.

The basic usage is:

import requests
from requests_toolbelt import MultipartEncoder

encoder = MultipartEncoder({'field': 'value',
                            'other_field', 'other_value'})
r = requests.post('https://httpbin.org/post', data=encoder,
                  headers={'Content-Type': encoder.content_type})

If you do not need to take advantage of streaming the post body, you can also do:

r = requests.post('https://httpbin.org/post',
                  data=encoder.to_string(),
                  headers={'Content-Type': encoder.content_type})

If you want the encoder to use a specific order, you can use an OrderedDict or more simply, a list of tuples:

encoder = MultipartEncoder([('field', 'value'),
                            ('other_field', 'other_value')])

Changed in version 0.4.0.

You can also provide tuples as part values as you would provide them to requests’ files parameter.

encoder = MultipartEncoder({
    'field': ('file_name', b'{"a": "b"}', 'application/json',
              {'X-My-Header': 'my-value'})
])

Warning

This object will end up directly in httplib. Currently, httplib has a hard-coded read size of 8192 bytes. This means that it will loop until the file has been read and your upload could take a while. This is not a bug in requests. A feature is being considered for this object to allow you, the user, to specify what size should be returned on a read. If you have opinions on this, please weigh in on this issue.

Monitoring Your Streaming Multipart Upload¶

If you need to stream your multipart/form-data upload then you’re probably in the situation where it might take a while to upload the content. In these cases, it might make sense to be able to monitor the progress of the upload. For this reason, the toolbelt provides the MultipartEncoderMonitor. The monitor wraps an instance of a MultipartEncoder and is used exactly like the encoder. It provides a similar API with some additions:

• The monitor accepts a function as a callback. The function is called every time requests calls read on the monitor and passes in the monitor as an argument.
• The monitor tracks how many bytes have been read in the course of the upload.

You might use the monitor to create a progress bar for the upload. Here is an example using clint which displays the progress bar.

To use the monitor you would follow a pattern like this:

import requests
from requests_toolbelt.multipart import encoder

def my_callback(monitor):
    # Your callback function
    pass

e = encoder.MultipartEncoder(
    fields={'field0': 'value', 'field1': 'value',
            'field2': ('filename', open('file.py', 'rb'), 'text/plain')}
    )
m = encoder.MultipartEncoderMonitor(e, my_callback)

r = requests.post('http://httpbin.org/post', data=m,
                  headers={'Content-Type': m.content_type})

If you have a very simple use case you can also do:

import requests
from requests_toolbelt.multipart.encoder import MultipartEncoderMonitor

def my_callback(monitor):
    # Your callback function
    pass

m = MultipartEncoderMonitor.from_fields(
    fields={'field0': 'value', 'field1': 'value',
            'field2': ('filename', open('file.py', 'rb'), 'text/plain')},
    callback=my_callback
    )

r = requests.post('http://httpbin.org/post', data=m,
                  headers={'Content-Type': m.content_type})

class requests_toolbelt.multipart.encoder.MultipartEncoderMonitor(encoder, callback=None)¶

An object used to monitor the progress of a MultipartEncoder.

The MultipartEncoder should only be responsible for preparing and streaming the data. For anyone who wishes to monitor it, they shouldn’t be using that instance to manage that as well. Using this class, they can monitor an encoder and register a callback. The callback receives the instance of the monitor.

To use this monitor, you construct your MultipartEncoder as you normally would.

from requests_toolbelt import (MultipartEncoder,
                               MultipartEncoderMonitor)
import requests

def callback(encoder, bytes_read):
    # Do something with this information
    pass

m = MultipartEncoder(fields={'field0': 'value0'})
monitor = MultipartEncoderMonitor(m, callback)
headers = {'Content-Type': montior.content_type}
r = requests.post('https://httpbin.org/post', data=monitor,
                  headers=headers)

Alternatively, if your use case is very simple, you can use the following pattern.

from requests_toolbelt import MultipartEncoderMonitor
import requests

def callback(encoder, bytes_read):
    # Do something with this information
    pass

monitor = MultipartEncoderMonitor.from_fields(
    fields={'field0': 'value0'}, callback
    )
headers = {'Content-Type': montior.content_type}
r = requests.post('https://httpbin.org/post', data=monitor,
                  headers=headers)

Streaming Data from a Generator¶

There are cases where you, the user, have a generator of some large quantity of data and you already know the size of that data. If you pass the generator to requests via the data parameter, requests will assume that you want to upload the data in chunks and set a Transfer-Encoding header value of chunked. Often times, this causes the server to behave poorly. If you want to avoid this, you can use the StreamingIterator. You pass it the size of the data and the generator.

import requests
from requests_toolbelt.streaming_iterator import StreamingIterator

generator = some_function()  # Create your generator
size = some_function_size()  # Get your generator's size
content_type = content_type()  # Get the content-type of the data

streamer = StreamingIterator(size, generator)
r = requests.post('https://httpbin.org/post', data=streamer,
                  headers={'Content-Type': content_type})

The streamer will handle your generator for you and buffer the data before passing it to requests.

If, for example, you need to upload data being piped into standard in, you might otherwise do:

import requests
import sys

r = requests.post(url, data=sys.stdin)

This would stream the data but would use a chunked transfer-encoding. If instead, you know the length of the data that is being sent to stdin and you want to prevent the data from being uploaded in chunks, you can use the StreamingIterator to stream the contents of the file without relying on chunking.

import requests
from requests_toolbelt.streaming_iterator import StreamingIterator
import sys

stream = StreamingIterator(size, sys.stdin)
r = requests.post(url, data=stream,
                  headers={'Content-Type': content_type})

class requests_toolbelt.streaming_iterator.StreamingIterator(size, iterator, encoding='utf-8')¶

This class provides a way of allowing iterators with a known size to be streamed instead of chunked.

In requests, if you pass in an iterator it assumes you want to use chunked transfer-encoding to upload the data, which not all servers support well. Additionally, you may want to set the content-length yourself to avoid this but that will not work. The only way to preempt requests using a chunked transfer-encoding and forcing it to stream the uploads is to mimic a very specific interace. Instead of having to know these details you can instead just use this class. You simply provide the size and iterator and pass the instance of StreamingIterator to requests via the data parameter like so:

from requests_toolbelt import StreamingIterator

import requests

# Let iterator be some generator that you already have and size be
# the size of the data produced by the iterator

r = requests.post(url, data=StreamingIterator(size, iterator))

You can also pass file-like objects to StreamingIterator in case requests can’t determize the filesize itself. This is the case with streaming file objects like stdin or any sockets. Wrapping e.g. files that are on disk with StreamingIterator is unnecessary, because requests can determine the filesize itself.

Naturally, you should also set the Content-Type of your upload appropriately because the toolbelt will not attempt to guess that for you.