@UnownBot, filtered

This program is designed to give you a filtered view of the @UnownBot Twitter account. It filters by distance to a set of locations (so you can be notified only about Unowns that spawn near locations you usually hang out at).

Setup

Getting the code

This program is written in Node.js, which you'll need to install on your computer. If you already have Node.js installed, make sure it is at least v6.5.0; if not, update.

Once you have Node.js installed, you'll need to open a terminal (Mac documentation, Windows documentation). Then install the program by typing

npm install -g unownbot-filtered

This will take a while as it downloads this program from the internet. Once it's done, try to run it, by typing:

unownbot-filtered

This will error, telling you that you need to set up the config file. Follow those instructions, using the below section for guidance. Once you've done all that, you can type unownbot-filtered in the console again, and this time it should work!

Configuring the program

• Copy config.sample.json to config.json.
• Add your Twitter API credentials:
  • Generate a Twitter token for your account.
  • You'll probably need to create a new application while doing so.
  • Use the generated data to fill in the appropriate fields in config.json.
• Add your Twilio API credentials:
  • Sign up for an account
  • Use the resulting page to fill in the appropriate fields in config.json.
  • Go to the SMS section and click "Get your first Twilio number", and fill that field in as well.
  • FYI, if you don't pay Twilio, it will prepend "Sent from your Twilio trial account - " to all your messages.
• Set the phone number to text in config.json's numberToText field.
• Update the locations with whatever locations and radiuses you want to monitor. Radiuses are in kilometers.
  • Unown spawns so infrequently that you probably want to cast a wide net; at least 20km, or you may never even see a spawn. Unown spawns typically last 30min, so if you're driving, a 20km detour should be achievable.

Testing

You can test the program after installing it by running

unownbot-filtered-test

which will use the test data stored in this repository's test/fixtures/tweets.json file to run through the normal program, log what's going on, and text you if appropriate. This can be used to ensure you did the setup correctly.

Running

Once setup is complete, do

unownbot-filtered

to start the program. It will run forever, logging output for your information, and texting when appropriate.

You could leave this running on your computer, if it doesn't turn off often, or maybe you could get a cloud provider like EC2. If you're on a Linux system and want to run things in the background, you can use the command

nohup unownbot-filtered my-config.json >my-log.txt &

Running multiple instances

If you want to run multiple instances (e.g., set up one for you and another for your friends), you can use multiple config files.

To do that, copy config.sample.json to multiple files, e.g. config-alice.json and config-bob.json, and fill it out differently for each person. (You can share API credentials between configs, although you might get rate limited if you share them too much.) You will want to change the configLabel field with some identifying value (e.g. their name).

Once you have the configs set up, run the program passing each config file. For example, to run an instance for Alice, do

unownbot-filtered config-alice.json

Running as a web server (experimental)

If you've cloned the source repository, you can do

npm run start

to start a server which will act the same as the daemon, but visiting it will display the log file. This is helpful if deploying somewhere (like now.sh) which demands that you run a HTTP server.

Future possibilities

• If you have overlapping zones, only text about one of them
• Add walking or cab times to the messages
• Add a simple web UI that would allow you to update the locations list on the go (so e.g. if you're spending a day in Central Park you can add it from your phone)