Example Usage:

@osdc-bot insult <name>

This returns an insult joke with name as it's subject.

@<mention_name> ++
@<mention_name> --

Both the above should work and increase/decrease the value of score of the mentioned person.

Note for anyone tackling this issue: The score should be persistent and must not be reset to zero on deployment (re-running of the node chatbot.js script)!

The reasons for having a separate Gitter room for testing:

1. Manual testing - this room can be used to test out local changes freely without having to worry about the main chat being cluttered.
2. Automated testing (A minor possibility, needs further investigation) - The current tests that are being implemented in #5 are just unit tests. It would a great addition if we could add integration tests for the bot, for which this room can be used as a playground!

There are many of the services which are not tested in the current code. This issue aims at writing unit tests for those services as well. As a reference, see https://github.com/osdc/osdc-bot/blob/master/test/jokeServiceSpec.js

• Weather Service
• Locate Service
• Wiki Service
• Quote Service
• Define Service

For eg. @osdc-bot karma is the wrong command.

Below is the traceback. This causes the app to exit.

[INFO] In loop karma
/home/osdc-bot/chatbot.js:57
    } else if (msgBody[2].trim() === "++") {
                         ^

TypeError: Cannot read property 'trim' of undefined
    at getBotReply (/home/osdc-bot/chatbot.js:57:26)
    at Request._callback (/home/osdc-bot/api.js:32:9)
    at Request.self.callback (/home/osdc-bot/node_modules/request/request.js:187:22)
    at emitTwo (events.js:87:13)
    at Request.emit (events.js:172:7)
    at Request.<anonymous> (/home/osdc-bot/node_modules/request/request.js:1044:10)
    at emitOne (events.js:77:13)
    at Request.emit (events.js:169:7)
    at IncomingMessage.<anonymous> (/home/osdc-bot/node_modules/request/request.js:965:12)
    at emitNone (events.js:72:2

As running the python server is now compulsory, I think the installation part in README.md should include the steps required to configure nltk.

If nobody has a problem with this, I'd like to solve this.

Currently, jokes are only fetched through Internet Chuck Norris Database.

It would be nice if we integrate more joke platforms in this.

The current parsing flow is very rigid. It should allow for more free statements like:

• @osdc-bot joke
• @osdc-bot tell me a joke
• @osdc-bot joke suna
• @osdc-bot can i hear a joke

Parsing should be done such that all the above should yield a joke.

Currently the help doesn't show how to use the commands.
Example: it may be like this:

wiki <term>  
location <city>   
joke