Need to tweet periodically and in an automated fashion? Then Twuckoo is for you. Still not convinced? :) Read on.
Twuckoo fulfills the task of periodically fetching a message and tweeting it (Twuckoo = Twitter + Cuckoo). Since we all prefer code to words, I'll show you its main loop:
def run ... while next_tweet do tweet(next_tweet) wait if wait_between_tweets? next_tweet = self.next notify if next_tweet.nil? end end
It is very simple in its design but opens vast possibilities due to its modular approach.
A simple API
Twuckoo takes a modular approach. It defines a very simple API and expects modules to implement these API methods. The methods are:
Loads the possible tweets that twuckoo will use when selecting the next one. This could be reading all lines from a file, for example.
Fetches the next message to be tweeted. Usually the bulk of the "business logic" is implemented here. This could be scraping a web page and extracting a specific snippet of html.
Hands the last tweeted message to be stored. In the case of the one_line_from_file module, this stores the tweet in a file so that no message is used multiple times.
Twuckoo needs to be passed the name of the module to be used. At the moment there are two provided modules:
Loads all the lines from a file (lines.txt) and tweets one line randomly each time.
Scraps Wikipedia's main page and extracts the name and link of Today's Featured Article. This is not as general a module as
file. I am thinking about how to best make a "web" module out of this.
gem install twuckoo --source http://gemcutter.org
Twuckoo depends on twibot and a couple of other gems which will also get installed with the above command.
In order for twibot to be able to connect to your twitter account, a config/bot.yml file need to be present at where you launch twuckoo from. It has to contain the following lines at least:
login: my_login password: my_password
For more options, see twibot's README
Twuckoo's configuration options have to be written into
config/twuckoo.yml. The current options are:
By default, it is "1d", so the script will "relax" for 24 hours after tweeting. The value should be given in a human-comprehensible form. You can use any combination of weeks, days, hours, minutes and seconds, so 1w3d13h27m19s will work, too, although you probably do not want to be this precise :)
When twuckoo is out of messages to tweet, it quits. Most of the times you'll want to get notified of this, though, so you can take action (e.g refill it and relaunch). If the above options are set, twuckoo sends an email to
password credentials for gmail authentication. (Yes, currently only gmail is supported, I am planning to change that in a next version.)
You have to indicate the module you wish to use:
$ twuckoo file
Will make use of the
file module. There has to be a lines.txt file in the directory where you run twuckoo from that contains the possible tweets, one tweet per line. The tweeted lines will be stored in a file called used_lines.txt so you should have write permission to the directory.
$ twuckoo wikipedia_tfa
You can also provide a
name for the twuckoo instance which appears in the notifications (if you do not provide a name explicitly, the name of the directory in which your twuckoo instance runs will get used):
$ twuckoo -n pragthinklearn file
Examples out there
In the course of the development of this gem I "ate my own dogfood" so you'll find at least one twitter account for each of the modules:
An concise advice from the excellent Pragmatic Thinking and Learning book
Serves you an Oblique Strategy per day.
An interesting article from the English Wikipedia delivered right to your twitter feed.
If you set up something with twuckoo, I would like to know about it, so that I might include it here.
Credits & License
This software is released under the MIT license (see attached). A link back to this page would be appreciated if you release something based on it.
Original idea and development: Balint Erdi