req-fast
v0.2.21
Published
This module is designed to be the fast, lightweight way to fetch the web content(HTML stream).
Downloads
11,212
Maintainers
Readme
req-fast
This module is designed to be the fast, lightweight way to fetch the web content(HTML stream) from specific server. it supports:
- Follow Redirects
- Automatic Decoding Content Encodings(Avoid Messy Codes, Especially Chinese)
- Cookies
- JSON Response Auto Handling
- Gzip/Deflate Encoding(Automatic Decompress)
- Proxy
Installation
$ npm install req-fast --production
Debug
$ DEBUG=reqfast.* node ...
Usage
var req = require('req-fast');
req(options, callback);
Options
When options is instance of String, it means the URL of server that to be requested.
req('http://www.google.com', function(err, resp){
// code goes here...
});
Otherwise it should be an object, including:
uri || url Url to which the request is sent.
method Http method,
GET
as default, but ifdata
was set and this value wasundefined
, it will bePOST
. And it could be one of OPTIONS, GET, HEAD, POST, PUT, PATCH, DELETE, TRACE and CONNECT.timeout Set a timeout (in milliseconds) for the request,
60000
(60 seconds) by default.dataType Type of data that you are expecting send to server, this property effects on POST, PUT, PATCH
method
only. It could be below values:- json
content-type
equalsapplication/json
. - form
content-type
equalsapplication/x-www-form-urlencoded
.
- json
data Data to be sent to the server, it should be key/value pairs. If the method is not set to
POST
, it will be converted to a query string, and appended to theurl
.agent A value indicating whether automatic generating browser-like
user-agent
or not, i.e.:Mozilla/5.0 (Macintosh; Intel Mac OS X 10_8_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/38.0.2125.101 Safari/537.36
,true
as default.Once
user-agent
was generated, theProcess finished with exit code 0
thing will not happen unless triggered manually, i.e.: COMMAND+C orprocess.exit(0)
.charset Set charset of content encodings if necessary.
This option takes top priority of decoding chunks, if not set, the
charset
inresponse.headers['content-type']
will be used at first, then thecharset
on<meta ... />
.disableRedirect A value indicating whether disable following redirect or not, if this value was set to
true
, themaxRedirects
will has no effect.maxRedirects The maximum number of redirects to follow(3 as default).
disableGzip Request compressed content from server and automatic decompress response content, if this option sets to
true
, this feature will be disabled.trackCookie A value indicating whether gathering all the cookies when following redirect or not,
false
by default,false
means gathering the cookie of last request only.cookies It should be key/value pairs.
headers Http headers, it should be key/value pairs, and by default:
{ 'connection': 'keep-alive', 'accept': 'text/html, text/javascript, application/json, application/xhtml+xml, application/xml;q=0.9, */*;q=0.8', 'pragma': 'no-cache', 'cache-control': 'no-cache' }
You can override the aboving in
headers
.proxy The proxy including all the options from tunnel proxy:
- host A domain name or IP address of the server to issue the proxy request to.
- port Port of remote proxy server..
- localAddress Local interface if necessary.
- proxyAuth Basic authorization for proxy server if necessary, i.e.
username:password
. - headers An object containing request headers.
Callback
Function to be called if the request succeeds or fails. The function gets passed two argument:
- error The
Error
instance. if succeeds, this value should benull
. If status is not okay,error.message
should be one of http.STATUSCODES. - response the response object, including:
- body The response body. If
response.headers['content-type']
equalsapplication/json
, the data(response.body) back from server will be parsed asJSON
automatic, otherwise isString
. - cookies The response cookies(key/value pairs).
- headers The response headers(key/value pairs).
- redirects The urls redirect(Array).
- statusCode The response status code.
- body The response body. If
see test or examples folder for a complete example
Streaming
Stream is amazing in node.js, if you are interesting on it, read John's Blog. You can add listeners on the returning Stream if you want.
var rs = req([options]);
rs.on('data', function(chunk){
// ...
});
rs.on('end', function(resp){
// ...
});
rs.on('error', function(error, response){
// ...
});
rs.on('abort', function(){
// ...
});
Pipe to file
In my project downloading millions of files from servers, using pipe
could improving performance, the file downloading from server chunk by chunk, but not read whole file to memory then download once, it sucks.
var fs = require('fs');
req('http://example.com/beauty.gif').pipe(fs.createWriteStream('download/001.gif'));
Http Status
All the http statuses will be handled, but you'd better check status carefully.
req('http://example.com', function(err, resp){
if(err){
// get status error;
}
// statusCode always exist except STREAM `error` was caught.
var status = resp && resp.statusCode;
})
Proxy
req({
url: 'http://example.com',
proxy: {
host: '127.0.0.1', // host
port: 8082, // port
proxyAuth: 'user:password' // authentication if necessary.
}
}, function(err, resp){
// code goes here
});
Benchmark
It's comparing with request
module, in order to avoid the influence of network, all the requests are sent to localhost.
The test cases are just for referencing, it's not trustworthy ^^.
Run Server
node --harmony benchmark/server.js
Elapsed Time
node --harmony benchmark/elapsed_time.js
A sample of 1000 cases:
module avg min max
request 0.005ms 0ms 2ms
reqfast 0.001ms 0ms 1ms
completed
Memory Usage
node --harmony benchmark/memory_usage.js
A sample of 1000 cases:
module avg min max
request 204.8b 0b 110592b
reqfast 8.192b 0b 4096b
completed
GC effects these a lot, and I do not believe the result of
process.memoryUsage().rss
,request
should performances better maybe.
Tests
Most tests' requests are sent to httpbin, so if you wanna run the test, please make sure you can resolve the host(httpbin). Run test:
npm test
Thanks
Appreciate to andris9. I've used fetch for a long time, it's very fast and simple to use.
my ES Spider needs speed up, request is very powerful, but too heavy/slow to me, and can not automatic decode encodings, especially Chinese.
Unfortunately andris9 could not maintain his repository any more, it have bugs, also I can fix them in my project, but it's fussy. One more, I need a PROXY feature.
License
Copyright 2014 Tjatse
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.