This little bash script shows how to use curl, grep, tail, sed and perl one-liners in order to compose a bleeding-edge twitter bot.
This one returns daily weather forecasts for Lisbon city based on the BBC weather forecast rss feed.
This one returns daily weather forecasts for Lisbon city based on the BBC weather forecast rss feed.
#! /bin/sh #Goto here here=/home/guillaume/Personal cd $here #BBC Lisbon weather id id=0048 #BBC weather RSS feed address feed="http://feeds.bbc.co.uk/weather/feeds/rss/5day/world/${id}.xml" #City city=lisbon #temporary file file="weather${city}" #Weather twitter bot twitbot=weatherlisbon:******* #Timestamp the log file echo .>> $file.log date >> $file.log #Read the RSS feed and filter it curl $feed | grep 'title' | tail -n 1 | perl -wlne'm/title>(.*)<\/title/i && print $1' | sed -e "s/°//g" > $file.txt #Read the forecast into a weather variable read weather < $file.txt #Twit the weather variable away curl --basic --user $twitbot --data status="$weather" http://twitter.com/statuses/update.xml >> $file.log
