Calvin and Bot

A very simple Discord bot that displays contents from Calvin and Hobbes comics.

This shares comics from GoComics and will look only on original publication dates (between 18th november 1985 and 31th december 1995).

Inviting the bot to your Discord server

Click here to invite the bot to your Discord server

How to interact with it

You can call this bot with mention, /calvin (customizable) and direct message.

Commands:

• help: You’re reading it!
• random: gives you a random strip.
• yyyy-mm-dd, yy-mm-dd, dd/mm/yyyy or dd/mm/yy: displays the strip for the given date (between 18th november 1985 and 31th december 1995) (sorry to US people, you’ll have to use logical date formats).
• any other terms will search a strip with provided terms in transcript.

Search allows wilcards (single-character te?t or multiple characters te*t), regular expressions (/[mb]oat/), fuzzy search (roam~) and any Lucene-supported terms modifier as specified here.

How to run it

Build

There is a shadowWar Gradle command to build a single executable Jar file : ./gradlew shadowJar

Run

Just run it as a standard executable Jar:

java -Xmx64m -Ddiscord4j.token={your token here} -jar calvinandbot-1.0.0-all.jar

Configure

This bot will accept following parameters as Java VM options (-Dparameter.name=value form):

Name	Type	Default	Comment
discord4j.token	String	None	Mandatory. Your Discord bot token.
bot.command	String	/calvin	The command to trigger the bot in public channels.
lucene.index.path	Path (as String)	lucene/	Path to store Lucene indexation files.
strips.dataUrl	URL (as String)	Empty	URL to store the strips data to use, before indexation. If empty, blank or non-provided, the bot will use embedded data (should be complete).
scraper.enabled	Boolean (true or false)	false	Enables the data scraper to re-download strips data from GoComics.
scraper.delayMs	Integer	1000	Delay between two strips loads, in milliseconds. Used only if scraper is enabled.
scraper.timeoutMs	Integer	30000	Timeout for strip scraping, in milliseconds. Used only if scraper is enabled.

Real project’s goal

OK, this is mainly a test for various technologies, as Kotlin coroutines, Gradle’s Kotlin DSL, Discord4J, embedded Lucene index (there was a private ElasticSearch + Spring Boot version of this bot, which required way more RAM), Koin (dependency injection) Ktor (HTTP client), JSoup (HTML parsing) and MockK (mocking).

This also explains the lack of CI, very partial test coverage (some tests are quite complicated) and deployment stuff.

Transcripts from GoComics contains lots of typos, and indexation settings are badly tuned, thus this bot will probably give quite bad results.