SquirrelJME/SquirrelJME

View on GitHub
assets/developer-notes/stephanie-gawroriski/2018/09/01.mkd

Summary

Maintainability
Test Coverage
# 2018/09/01

## 10:59

Well, my new job has been going well working at IOpipe. I love it. It does
take a bunch of my time since it is full-time. Also recently I removed a
distraction from me and honestly it seems really peaceful now like I can think
and such. Maybe I was addicted. But anyway, going to continue work on
SquirrelJME. I got a new logo designed and Lex is being redesigned by
[CYMKat](https://twitter.com/CMYKhia).

## 11:06

Need to figure out the best way to initialize classes. But honestly I think
something that just works will be good enough. Because I hit this problem
before where I tried to solve it without understanding it. SpringCoat is
supposed to be dirty and not the best way to implement. I mean once
everything is working much better I can recycle and reiterate to SummerCoat
and slowly develop everything.

## 11:43

Would be easiest to invoke a static main entry point, will not need to deal
with initializing objects and such.

## 19:07

I think having services would be a bit too complex, something more direct
might work out better in the end. But I will avoid thinking about this until I
actually write SpringCoat. Hopefully I can write SpringCoat.

## 19:33

Actually an idea, it is inefficient but it could work. Basically when I load
and initialize a class I load all of its information such as fields, methods,
and other things. Then I maintain everything that is actually available for
the class. So there will be duplicated information but it will remove the
need to do recursive handling of classes and such. I can load everything as it
needs to be for a class and just use back references and such. But I need the
basic class hierarchy to exist first before I can process fields and methods.
But basically say I load class `A`, that will load `Object`. No class can
recursively depend on each other, so there is no issue there.