This application index tweets from various Twitter handles and makes them searchable using the Apache Solr library.
The front end is built using Twitter Bootstrap and JQuery while the back end is coded using Node.js
###Setup instructions
After cloning the repo, run the following command:
npm install
This will install all the requisite Node.js dependencies.
In the cloned repo you will a solr-5.5.0 directory. This has Solr set up with configuration files and schema files for two cores, namely tweets and images. You should run your Solr instance from this directory or use the configuration files (solrconfig.xml and schema.xml) from this directory.
This application also utilises the LIRE Plugin for Solr in order to support image search. All the necessary files are included in the liresolr directory of the repo. The solr-5.5.0 directory is already preconfigured to work with this liresolr directory. The plugin will be automatically called from this directory by the application while indexing and searching images.
Finally, a .env file is required for this application to run. This .env file is supposed to contain Twitter API credentials which will be used by the application to fetch the tweets. Refer to the .env.sample to understand the format of the .env file. After cloning the repo, create a .env file under the root of the application and populate according to the .env.sample file.
Note:
The 'images' folder is the one that will store and serve the image files when tweets are fetched and indexed.
Note:
The application assumes that your Solr instance runs on host 'localhost' and port '8983'. In order to change this default configuration, the file 'lib/solr.js' will need to be changed.
This repo already has more than 14500 tweets indexed in Solr. So as soon you run the application, these tweets will be searchable.
There is another Solr core in this project called images. This core stores image data in order to facilitate image search. The core is this repo currently contains data for around 1000 images. So the image search will also work when the project is run. However, in order to serve the actual images on the web app UI, the image files need to be downloaded. They can be found here. This is a zip file containing the image files. Unzip this files, extract the images and place those image files under the images directory of the repo. Once this is done, the image search will also work.
The above steps should be sufficient to get the user up and running with this application.
First, start your Solr instance using the following command:
<path_to_repo>/solr-5.5.0/bin/solr restart
Next, start the node server:
node <path_to_repo>/index.js
Go to http://localhost:3001 on your browser. The application should be visible at this URL.