2014/10/01

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.

00:00

New month, this means statistic generation time. Made a bunch of removals and a bunch of additions, but very even scaled. I spent most of the month attempting to get my compiler to compile (two iterations, the last one is final). However I stopped doing that to work on the compiler-wrapper which works quite well so far. I need to get into kernel land soon with actual booting of my own code, and leveraging a preexisting compiler will work despite not being able to self host when I get to that point before the compiler is written. For this month of October I will have to get the compiler-wrapper working, have some kind of compilation system to native code and hopefully have booting done. I will also need a testing system where I can produce and reproduce results, I can perhaps use compiler-wrapper, but the test system will help in conformity and bug searching.

[x-date-201409 --] x-date-201410]
Subtractions: 27013, Additions: 38953

Totals grouped by language (dominant language first):
java:         16013 (93.00%)
sh:            1206 (7.00%)

02:39

Now output files are generated, but after output it appears it is trying to delete a file, and I wonder which.

03:34

ByteArrayFileChannel's write was very broken but now that is fixed and works now. Also the compiler invokes fully and compiles code at least for the k8-progress package. I also need to change the writing stuff for the shadowing a bit so I can access the stuff easier, and where the shadow reflects modified changes of the source.

07:00

I have all of that code and the shadow cache stuff now (except for writing JAR files), and it appears that the shadow cache stuff is never called.

07:14

Actually, implCloseChannel() must be called on said objects otherwise the normal close in the channel will just call the same method. And now it works. So the only thing to do at this point, is to actually generate JAR files. Then once those are generated, go up to another dependency and build that and use the JAR file and such.

22:42

Naming the JAR is easy, however I have colons in the version number which complicates things a bit. I need to switch to a new character, perhaps "--" instead of just ":", that would complicate parsing version numbers a bit but not too much hopefully.

23:54

JAR files are created, now they just to be put in the classpath for the compiler to use. One thing that needs fixing is sharing of InternalDoBuilds because otherwise tons are created needlessly for the same package.