DISCLAIMER: These notes are from the defunct k8 project which precedes SquirrelJME. The notes for SquirrelJME start on 2016/02/26! The k8 project was effectively a Java SE 8 operating system and as such all of the notes are in the context of that scope. That project is no longer my goal as SquirrelJME is the spiritual successor to it.


So after this diversion I am now back at the the recompilation of the meta macros and such. But first, allocator must be written. The assembler has to also be modified to use fragments instead.


I kind of have this feeling that I just want to hold that off and do some other code moving. Such as clipping out vmjrt into vmjar with compact profile classes and such.


Due to the nature of realmv, I will have to spread out moving the code around. But I wonder what I should call the runtime packages. Perhaps something simple like: java-library-profile1 and such. That can work. That way when I build the kernel I can just build that without requiring tons and tons of stuff.


Using simple sort and such, this will the most deep directories first then after that will follow the shallower ones.


Instead of profile I might just call it level, although that sounds cooler it does not exactly match Java per se since profile is used. I believe I will add an extra dash between profile and the number.


Perhaps just library then the number, since it is rather obvious as to what the number after it is (the profile). Although I should add a mark for Java 8 in the case that I ever want to up to Java 9, the names will not be in the way. Although that could prove troublesome in the future as I would basically have a fork of the entire class library and such. So just java-library in this case.


Now that everything is moved, now to do the final touches.


Completion needs to be changed to handle the new library layout.


Completion will also no longer need a source path at all also.


What I need now for completion is some kind of progress meter that uses forwarding objects that wrap the stream methods.


It appears as if the Java compiler loads all the input files before hand and then does annotation processing on them all at once. This means lots of memory usage although it is faster in the long run provided swap is not used when there is not enough memory.