http

Table of Content

• Name
• Status
• Synopsis
• Description
• Exceptions
  • http.HttpError
  • http.LineTooLongError
  • http.ChunkedSizeError
  • http.NotConnectedError
  • http.ResponseNotReadyError
  • http.HeadersError
  • http.BadStatusLineError
• Constants
  • http.Client.status
  • http.Client.has_read
  • http.Client.headers
  • http.Client.content_length
  • http.Client.chunked
• Classes
  • http.Client
• Methods
  • http.Client.send_request
  • http.Client.read_status
  • http.Client.read_headers
  • http.Client.read_response
  • http.Client.request
  • http.Client.send_body
  • http.Client.read_body
  • http.Client.get_trace_str
  • http.headers_add_host
  • http.request_add_host
• Author
• Copyright and License

Name

http

Status

The library is considered production ready.

Synopsis

from pykit import http

headers = {
    'Host': '127.0.0.1',
    'Accept-Language': 'en, mi',
}

try:
    h = http.Client('127.0.0.1', 80)

    # send http reqeust without body
    # read response status line
    # read response headers
    h.request('/test.txt', method='GET', headers=headers)

    status = h.status
    # response code return from http server, type is int
    # 200
    # 302
    # 404
    # ...

    res_headers = h.headers
    # response headers except status line
    # res_headers = {
    #   'Content-Type': 'text/html;charset=utf-8',
    #   'Content-Length': 1024,
    #   ...
    # }

    # get response body
    print(h.read_body(None))
except (socket.error, http.HttpError) as e:
    print(repr(e))

from pykit import http
import urllib

content = urllib.urlencode({'f': 'foo', 'b': 'bar'})
headers = {
    'Host': 'www.example.com',
    'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
    'Content-Length': len(content),
}

try:
    h = http.Client('127.0.0.1', 80)

    # send http reqeust
    h.send_request('http://www.example.com', method='POST', headers=headers)

    # send http request body
    h.send_body(content)

    # read response status line and headers
    status, headers = h.read_response()

    # read response body
    print(h.read_body(None))
except (socket.error, http.HttpError) as e:
    print(repr(e))

Description

HTTP/1.1 client

We find that httplib must work in blocking mode and it can not have a timeout when recving response.

Use this module, we can set timeout, if timeout raise a socket.timeout.

Exceptions

http.HttpError

syntax: http.HttpError

The base class of the other exceptions in this module. It is a subclass of Exception.

http.LineTooLongError

syntax: http.LineTooLongError

A subclass of HttpError. Raise if length of line is greater or equal than 65536 when read response status line, headers and get length of chunked block.

http.ChunkedSizeError

syntax: http.ChunkedSizeError

A subclass of HttpError. Raise if failed to get length of chunked block.

http.NotConnectedError

syntax: http.NotConnectedError

A subclass of HttpError. Raise if send data without connecting to server.

http.ResponseNotReadyError

syntax: http.ResponseNotReadyError

A subclass of HttpError. Raise if response is unavailable.

http.HeadersError

syntax: http.HeadersError

A subclass of HttpError. Raise if failed to read response headers.

http.BadStatusLineError

syntax: http.BadStatusLineError

A subclass of HttpError. Raise if failed to read response status line.

Constants

http.Client.status

syntax: http.Client.status

Status code returned by server.

http.Client.has_read

syntax: http.Client.has_read

The length of response body that has been read.

http.Client.headers

syntax: http.Client.headers

A dict(header name, header value) of response headers.

http.Client.content_length

syntax: http.Client.content_length

Length of http response body, if body is chunked encoding, it is None.

http.Client.chunked

syntax: http.Client.chunked

Whether response body is chunked encoding or not.

Classes

http.Client

syntax: http.Client(host, port, timeout=60, stopwatch_kwargs=None, https_context=None)

HTTP client class

arguments:

• host: server address, ip or domain, type is str

• port: server port, type is int

• timeout: set a timeout on blocking socket operations, unit is second, default is 60. The value argument can be a nonnegative float expressing seconds, or None. if None, it is equivalent to socket.setblocking(1). if 0.0, it is equivalent to socket.setblocking(0).

• stopwatch_kwargs: is an dictionary used as keyword arguments when initializing stopwatch instance.

  class StopWatch(object):
      def __init__(self,
                   strict_assert=True,
                   export_tracing_func=None,
                   export_aggregated_timers_func=None,
                   max_tracing_spans_for_path=1000,
                   min_tracing_milliseconds=3,
                   time_func=None):
  

• https_context: a ssl.SSLContext instance describing the various SSL options. Defaults to None.

Methods

http.Client.send_request

syntax: http.Client.send_request(uri, method='GET', headers={})

Connect to server and send http request.

arguments:

• uri: specifies the object being requested

• method: specifies an HTTP request method

• headers: a dict(header name, header value) of http request headers

return: nothing

http.Client.read_status

syntax: http.Client.read_status()

Read response status line and return the status code. Cache the response status code with http.Client.status

arguments: nothing

return: response status code.

http.Client.read_headers

syntax: http.Client.read_headers()

Read and return the response headers. Cache the response headers with http.Client.headers

arguments: nothing

return: response headers, it is a dict(header name , header value).

http.Client.read_response

syntax: http.Client.read_response()

Read response status line, headers and return them. Cache the status code with http.Client.status Cache the response headers with http.Client.headers

arguments: nothing

return: two values,

• status: response status code, type is int

• headers: response headers, it is a dict(header name , header value).

http.Client.request

syntax: http.Client.request(uri, method='GET', headers={})

Send http request without body and read response status line, headers. After it, get response status code with http.Client.status, get response headers with http.Client.headers.

arguments:

• uri: specifies the object being requested

• method: specifies an HTTP request method

• headers: a dict(header name, header value) of http request headers

return: nothing

http.Client.send_body

syntax: http.Client.send_body(body)

Send http request body. It should be a str of data to send after the headers are finished

arguments:

• body: a str of http request body.

return: nothing

http.Client.read_body

syntax: http.Client.read_body(size)

Read and return the response body.

arguments:

• size: need read size, if greater than left size use the min one, if None read left unread body.

return: the response body.

http.Client.readlines

syntax: http.Client.readlines(delimiter)

Read and yield one line in response body each time.

arguments:

• delimiter: specify the delimiter between each line, defalut supporting \n.

return: a generator yileding one line including delimiter in response body each time.

http.Client.get_trace_str

syntax: http.Client.get_trace_str()

return: a string shows time spent on each phase of a request. The following fields would presents in it:

conn: 0.000001 -; send_header: 0.000000 -; recv_status: 0.000000 -; recv_header: 0.000000 -; recv_body: 0.000000 -; recv_body: 0.000000 -; exception: 0.000000 -; pykit.http.Client: 0.000002 Exception:ValueError

The numbers are time in second.

There might be less fields in the result, if request failed.

It is also possible some of the above fields appear more than once.

For example if a server responds a 100-Continue status line and a 200-OK status line, there would be two recv_status fields.

If the caller calls http.Client.read_body more than one time, there would be more than one recv_body fields.

http.headers_add_host

syntax: http.headers_add_host(headers, address)

If there is no Host field in the headers, insert the address as a Host into the headers.

arguments:

• headers: a dict(header name, header value) of http request headers

• address: a string represents a domain name

return: headers after adding

http.request_add_host

syntax: http.request_add_host(request, address)

If the request has no headers field, or request['headers'] does not have a Host field, then add address to Host.

arguments:

• request: a dict(request key, request name) of http request

• address: a string represents a domain name

return: request after adding

Author

Baohai Liu(刘保海) [email protected]

Copyright and License

The MIT License (MIT)

Copyright (c) 2017 Baohai Liu(刘保海) [email protected]