pmonks / for-science Goto Github PK

When a user triggers evaluation (i.e. via !clj) and they use a clojure code fence in the subsequent content, I want the bot to recognise the code fence and pull the content to be evaluated from within it, so that the code fence formatting itself does not interfere with the evaluation of the content.

Notes:

• Un-code-fenced code should still be evaluated as it currently is today.
• Should this support multiple independent code-fenced sections in a single message, or just the first one? If so, should those be considered independent evaluations, or a single evaluation (i.e. concatenate the content of all of the fences, prior to evaluation)? The latter may be useful if issue #7 isn't implemented, and will also allow "literate" style messages (interspersed code & explanatory text).

The "if so how long" part is especially important, to prevent memory leaks. My initial thought is that this could be based on an inactivity period e.g. after X minutes of inactivity, the interpreter is reset.

Steps to Reproduce:

1. In a server where for-science is running, type the following:
2. !clj (seq "Hello, World!")

Expected Result:

Result:

(\H \e \l \l \o \, \space \W \o \r \l \d \!)

Actual Result:

Result:

(H e l l o ,   W o r l d !)

When I wish to execute some Clojure code, I want to be able to use an application ("slash") command to do so, so that the Discord UI provides me with guided assistance.