#!/usr/bin/env bash
#
#
#THIS PROGRAM FETCHES NEWSLINKS FROM EUROPACIFIC CAPITAL AND 
#COVERTS THEM TO READABLE JAVASCRIPT CODE FOR INSERTING IN BLOGGER'S
#HTML/JAVASCRIPT WEB MODULE.
#
#First we use curl to download the webpage.
curl -s http://www.europac.com/research_analysis/newswire > newswire.txt;
#
#Here we use egrep and sed on the file we just downloaded. We first create urls.txt file.
egrep -o 'url=([^'\'']*)' newswire.txt | sed 's|url=|<a href="|g' | sed 's|$|">|g' > urls.txt;
#Now let's grab the titles and create titles.txt file.
egrep -o "redirect\?url([^<]*)</a>" newswire.txt | sed -r 's|redirect\?url=[^>]*>||g' | sed -r 's|$|<p>|g' > titles.txt;
#Merge urls and titles files and create t_fetch_news.txt
paste urls.txt titles.txt > t_fetch_news.txt;
#remove lines and create fetch_news.txt
tr --delete '\n' < t_fetch_news.txt > fetch_news.txt;
#convert for javascript
sed -r 's|"|'\''|g' fetch_news.txt | sed -r 's|^|document.write("|g' | sed -r 's|$|");|g' > fetch_news.js