JakobGreen / lua-requests翻译 / 编辑






License Coverage Status Build Status

The same friendly Python Requests interface for Lua!

Basic Usage

> requests = require('requests')
> response = requests.get('http://httpbin.org/get')
> print(response.status_code)
> response = requests.post{'http://httpbin.org/post', data='random data'}
> json_data = response.json()
> print(json_data.data)
random data







The HTTP backend can be swapped out for anything that has the same API as LuaSocket's socket.http. This is done by setting the value of requests.http_socket. Swapping the HTTPS backend can be done by swapping out requests.https_socket.


Tests are located in the tests directory and are written using busted.

Install busted:

$ luarocks install busted

Run Tests:

$ busted -p _tests tests


lua-requests is licensed under the MIT license. See LICENSE.md for details on the MIT license.


Simple requests

Importing the lua-requests is quite simple.

> requests = require('requests')

Making a GET request is not much more difficult.

> response = requests.get('http://httpbin.org/get')

Other methods like POST, HEAD, OPTIONS, TRACE, PATCH, PUT, and DELETE are just as simple.

> response = requests.post('http://httpbin.org/post')
> response = requests.put('http://httpbin.org/put')

The second argument of a request is a table that can be used to make more advanced requests. Any request can be made with either a second argument or as a table.

> response = requests.post{url = 'http://httpbin.org/post', data = 'random data'}


> response = requests.post{'http://httpbin.org/post', data = 'random data'}


> response = requests.post('http://httpbin.org/post', {data = 'random data'})

NOTE: This documentation mostly uses two parameters instead of just one table because the single table feature was added later.

There is also a general request call. The first parameter is the method.

> response = requests.request("GET", 'http://httpbin.org/get')


Using HTTPS is as simple as changing the URL to be 'https' instead of 'http'

> response = requests.get('https://httpbing.org/get')

Basic Response

The http response contains all of the response data in different fields.

The response body is contained in response.text

> response = response.get('http://httpbin.org/robots.txt')
> print(response.text)
User-agent: *
Disallow: /deny

The response headers are contained in response.headers

> response = requests.get('http://httpbin.org/robots.txt')
> print(inspect(response.headers))
  ["access-control-allow-credentials"] = "true",
  ["access-control-allow-origin"] = "*",
  connection = "close",
  ["content-length"] = "30",
  ["content-type"] = "text/plain",
  date = "Tue, 07 Apr 2015 01:43:26 GMT",
  server = "nginx"

The response status code is contained in response.status_code

> response = requests.get('http://httpbin.org/robot.txt')
> print(response.status_code)

URL Parameters

It is common for URL's that need to have some sort of query string. For example, http://httpbin.org/response-headers?key1=val1&key2=val2. Adding parameters to a URL query is as simple as passing a table into the params field of the second argument.

> query_parameters = { key1 = 'val1', key2 = 'val2' }
> response = requests.get{'http://httpbin.org/response-headers', params = query_parameters}
> print(response.url)

For keys that contain a list of values just make the value into a table.

> query_parameters = { key1 = 'val2', key2 = {'val2', 'val3'}}
> response = requests.get{'http://httpbin.org/response-headers', params = query_parameters}
> print(response.url)

Sending Data

Sending data is possible with any command. Just pass the data you want to send into the data field of the second argument.

> data = "Example data"
> response = requests.post{'http://httpbin.org/post', data = data}

If a table is passed in to data it is automatically encoded as JSON.

> data = {Data = "JSON"}
> response = requests.post{'http://httpbin.org/post', data = data}

Custom headers

Custom headers can be added to any request method. Just pass a table into the headers field of the second argument.

> headers = {['Content-Type'] = 'application/json'}
> response = requests.get{'http://httpbin.org/headers', headers = headers}


Timeout in seconds can be passed as a parameter. If the host has not responded in timeout seconds then through an error.

> url = 'http://httpbin.org/delay/2'
> response = requests.get{url, timeout = 1}
requests.lua:261: error in GET request: timeout

Basic Authentication

Basic authentication can be added to any request.

> auth = requests.HTTPBasicAuth('user', 'passwd')
> response = requests.get{'http://httpbin.org/basic-auth/user/passwd', auth = auth}
> print(response.status_code)

Digest Authentication

Digest authentication can be added to any request.

> auth = requests.HTTPDigestAuth('user', 'passwd')
> response = requests.get{'http://httpbin.org/digest-auth/auth/user/passwd', auth = auth}
> print(response.status_code)

To continue using the same digest authentication just pass response.auth into the next request.

> response = requests.get{'http://httpbin.org/digest-auth/auth/user/passwd', auth = response.auth}
> print(response.status_code)

By reusing the response.auth you can save time by not needing to reauthenticate again. response.cookies contains cookies that the server requested to be set for authentication.


Cookies can be added to any request by setting the cookies field.

> response = requests.get{'http://httpbin.org/get', cookies = 'cookie!'}

JSON Response

JSON response's can be parsed into a Lua table using response.json(). JSON encoding and decoding is done with lua-cjson.

> response = requests.get{'http://httpbin.org/get', params =  {stuff=true}}
> json_body, error = response.json()
> print(json_body.args.stuff)

XML Response

XML response's can be parsed into a Lua table using response.xml(). XML encoding and decoding is done with xml which is based on RapidXML.

> response = requests.get('http://httpbin.org/xml')
> xml_body, error = response.xml()
> print(xml_body[1][1][1])
Wake up to WonderWidgets!

The returned xml table can be tricky to parse. I recommend using inspect to help the first time to help see the table structure.


A proxy server can be added as an argument to a request.

> response = requests.get{'http://httpbin.org/get', proxy = ''}


301 and 302 redirects are enabled by default for most requests. To disable redirects set allow_redirects = false.

> response = requests.get('http://httpbin.org/redirect-to?url=google.com', {allow_redirects = false})