[![NPM download][download-image]][download-url]
[![NPM quality][quality-image]][quality-url]

[quality-image]: http://npm.packagequality.com/shield/seenreq.svg?style=flat-square
[quality-url]: http://packagequality.com/#?package=seenreq
[download-image]: https://img.shields.io/npm/dm/seenreq.svg?style=flat-square
[download-url]: https://npmjs.org/package/seenreq

seenreq

A library to test if a url/request is crawled, usually used in a web crawler. Compatible with request and node-crawler. The 1.x or newer version has quite different APIs and is not compatible with 0.x versions. Please read the upgrade guide document.

Table of Contents

* Quick Start
* Installation
* Basic Usage
* Use Redis
* Use Mongodb
* Class:seenreq
* seen.initialize()
* [seen.normalize(uri|option[,options])](#seennormalizeurioptionoptions)
* [seen.exists(uri|option|array[,options])](#seenexistsurioptionarrayoptions)
* seen.dispose()
* Options

Quick Start

$3

$ npm install seenreq --save

$3

``javascript
const seenreq = require('seenreq')
, seen = new seenreq();

//url to be normalized
let url = "http://www.GOOGLE.com";
console.log(seen.normalize(url));//{ sign: "GET http://www.google.com/\r\n", options: {} }

//request options to be normalized
let option = {
uri: 'http://www.GOOGLE.com',
rupdate: false
};

console.log(seen.normalize(option));//{sign: "GET http://www.google.com/\r\n", options:{rupdate: false} }

seen.initialize().then(()=>{
return seen.exists(url);
}).then( (rst) => {
console.log(rst[0]);//false if ask for a request never see
return seen.exists(opt);
}).then( (rst) => {
console.log(rst[0]);//true if got same request
}).catch(e){
console.error(e);
};
`
When you call exists, the module will do normalization itself first and then check if exists.

$3


seenreq stores keys in memory by default, memory usage will soar as number of keys increases. Redis will solve this problem. Because seenreq uses ioredis as redis client, all ioredis' options are recived and supported. You should first install:

`javascript
npm install seenreq-repo-redis --save
`
and then set repo to redis:

`javascript
const seenreq = require('seenreq')
let seen = new seenreq({
repo:'redis',// use redis instead of memory
host:'127.0.0.1',
port:6379,
clearOnQuit:false // clear redis cache or don't when calling dispose(), default true.
});

seen.initialize().then(()=>{
//do stuff...
}).catch(e){
console.error(e);
}
`

$3

It is similar with redis above:

`javascript
npm install seenreq-repo-mongo --save
`

`javascript
const seenreq = require('seenreq')
let seen = new seenreq({
repo:'mongo',
url:'mongodb://xxx/seenreq',
collection: 'foor'
});
`




Class:seenreq

Instance of seenreq

$3

Initialize the repo, returns a promise.

$3

* uri String, option is Option of request or node-crawler
* options

Returns normalized Object: {sign,options}.

$3

* uri|option
* options

Returns a promise with an Boolean array, e.g. [true, false, true, false, false].

$3

Dispose resources of repo. If you are using repo other than memory, like Redis you should call dispose to release connection. Returns a promise.

Options

* removeKeys: Array, Ignore specified keys when doing normalization. For instance, there is a ts property in the url like http://www.xxx.com/index?ts=1442382602504 which is timestamp and it should be same whenever you visit.
* stripFragment: Boolean, Remove the fragment at the end of the URL (Default true).
* rupdate: Boolean, it is short for repo update. Store in repo so that seenreq can hit the same req next time (Default true).

RoadMap


* add mysql` repo to persist keys to disk.
* add keys life time management.