Shariff is used to determine how often a page is shared in social media, but without generating requests from the displaying page to the social sites.
This document describes the Perl backend. The following backends are also available:
This software is based on Mojolicious. Make sure, you've installed it, e.g. by running cpanm Mojolicious.
To get started with the shariff backend, unzip the current release zip file into a local folder. Start the server application by using the hypnotoad script.
$ hypnotoad script/shariff
Server available at http://127.0.0.1:8080.You need to proxy requests from your webserver to the hypnotoad server. See Mojolicious Cookbook for some Nginx or Apache recipes.
This project is bundled with a configuration file shariff.conf. The following configuration options are available:
| Key | Type | Description |
|---|---|---|
cache |
hashref |
Cache settings described below |
domain |
regexp |
A regular expression describing one or more domain(s) for which share counts may be requested |
hypnotoad |
hashref |
Hypnotoad webserver settings as described in Mojolicious docs. You may want to change the listen setting. |
services |
hashref |
Service specific settings |
Cache settings:
| Key | Type | Description | Default |
|---|---|---|---|
class |
scalar |
Cache class name to use. | Heise::Shariff::Cache. |
expires |
scalar |
Time in seconds to hold fetched data in in-memory cache. | 60 |
options |
arrayref |
A list of options to be passed to the cache class constructor. Setting options to [{foo => 'bar'}, 'baz'] leads to call $cache->new({foo => 'bar'}, 'baz'). |
[] |
If the backend runs under http://example.com/my-shariff-backend/, calling the URL http://example.com/my-shariff-backend/?url=http%3A%2F%2Fwww.example.com should return a JSON structure with numbers in it, e.g.:
{"facebook":1452,"googleplus":23}