loklak_scraper_js

loklak_scraper_js

Scrapers for loklak in javascript

The idea is, that loklak and its scrapers (i.e. in loklak_server, loklak_wok, a possible loklak_wok_ios and a possible scraper in a web page integration) all use the same common code base.

That code shall be placed here in the scraper subdirectory. Each file in that subdirectory should be named by the scraper target (like twitter.js)

Execution of the scrapers must output a json very like to a loklak search result - without the enrichments that the scrapers also implement. I.e. there is no need to integrate a link de-shortener, that must be implemented in the loklak application itself.

Example

There is an example file scrapers/example.js which has just a static output of a json object. You would just have to implement the same behavior for the other scrapers.

I.e.:

node scrapers/example.js

outputs a json object. all other scrapers must do the same but just take a query attribute.

Installing dependencies

This project uses request library to request for web pages and cheerio library to scrape. These libraries can be installed as shown below.

cd loklak_scraper_js
npm install

The dependencies will now be installed in node_modules directory

Quora profile scraper (quora.js)

The Quora profile scraper takes a single argument that is the name of the profile to scrape and outputs the scrapped data as a json object.

node scrapers/quora.js <profile_name>

For example:

…