Commit Graph

111 Commits

Author SHA1 Message Date
Torsten Ruger
671512b96c avoid fixnum overflow for hashes 2016-12-31 19:53:43 +02:00
Torsten Ruger
5f3744d6af remove unused attributes functions 2016-12-31 15:20:02 +02:00
Torsten Ruger
7bb1c361a5 small word test 2016-12-31 15:17:45 +02:00
Torsten Ruger
47c0f62212 error message 2016-12-31 15:10:07 +02:00
Torsten Ruger
fb414cecd6 use helper 2016-12-31 14:54:53 +02:00
Torsten Ruger
b9073d0c88 redid the type hashing
account for the class the type is for
by adding it to the hash code, so even the ivars are identical the
types are not
2016-12-31 14:51:06 +02:00
Torsten Ruger
0fa7f54bcc found and fixed hashing bug 2016-12-31 14:04:04 +02:00
Torsten Ruger
d6c503ecf4 inlined indexed into list
which was the only one using it.
A whole lot simpler, considering it will have to be bootstrapped someday
2016-12-30 21:00:18 +02:00
Torsten Ruger
7f06e00ccd making types private in space
turned out to be unnecessary, but still better
2016-12-30 20:47:28 +02:00
Torsten Ruger
ef66a87527 fixed silly dictionary bug
also finally duplicating keys and values and not handing them out
2016-12-30 20:46:18 +02:00
Torsten Ruger
a7935db107 cleaner way to generate argument types
possibly more correct
2016-12-30 19:17:59 +02:00
Torsten Ruger
83b6628f1a whitespace and code style 2016-12-30 19:17:15 +02:00
Torsten Ruger
db798d1a4a no default arg on init
also making sure the type is in the hash
but also that the order of type does not matter
2016-12-30 18:41:36 +02:00
Torsten Ruger
a82c9e8fb1 minor whitespace and renaming 2016-12-30 18:39:49 +02:00
Torsten Ruger
f0350601a7 Move the space instance to the parfait module
A better fit, maybe even a pattern for singletons
2016-12-30 14:10:49 +02:00
Torsten Ruger
a00f6be3ba remove the space instance from register machine
bad design, probably from the booting
2016-12-30 14:04:59 +02:00
Torsten Ruger
ef872edd7a rename instance_methods to just methods 2016-12-30 13:33:07 +02:00
Torsten Ruger
e69d9a492e mostly moving to ruby instance variables 2016-12-29 18:51:24 +02:00
Torsten Ruger
25f44949e4 removing the fake memory form object
just in word and list now
2016-12-29 18:49:03 +02:00
Torsten Ruger
6214040888 changing to ruby instance variables
from the fake memory
2016-12-29 18:47:45 +02:00
Torsten Ruger
b5f04ec718 mostly renames from the type change
also making setters as explicit set_xx methods
2016-12-29 18:45:32 +02:00
Torsten Ruger
f4b3c645e5 rework type with separate arrays for names and types
Means no more indexed
also using ruby instances
small rename for names
2016-12-29 18:42:38 +02:00
Torsten Ruger
4c3007e6c0 removing meta class (for now) 2016-12-29 18:39:59 +02:00
Torsten Ruger
80237e5033 rename position accessor
for future
2016-12-28 21:40:06 +02:00
Torsten Ruger
c60949fe24 small refactor on arg loading 2016-12-28 21:10:14 +02:00
Torsten Ruger
e551732f18 tighter integration with factory methods for adding code
define methods to collapse the code Register.
in add_code Register.factory_method
most instructions done, except op and branch that are rare
2016-12-28 20:37:54 +02:00
Torsten Ruger
af31774074 implement assignment type check 2016-12-28 19:20:16 +02:00
Torsten Ruger
9cf56b3aa6 use factory functions consistently to create instructions 2016-12-28 19:01:58 +02:00
Torsten Ruger
4cf732d395 use transfer factory 2016-12-28 18:37:15 +02:00
Torsten Ruger
57c038f13b use load_constant factory method 2016-12-28 18:31:29 +02:00
Torsten Ruger
4412eda105 small refactor and rename 2016-12-28 18:16:39 +02:00
Torsten Ruger
25ae6e3d26 remove next_list from NamedList
Was legacy from the days of (linked) Frames
Now only messages are linked
2016-12-28 18:08:07 +02:00
Torsten Ruger
8867b13240 yep, opal != OPAL
engine has to be in small letters it seems
OPAL came from the original commit in 2013, but alas
2016-12-28 14:35:07 +02:00
Torsten Ruger
9226de4502 testing to_s 2016-12-28 14:11:05 +02:00
Torsten Ruger
266764be06 testing to_s on Code 2016-12-28 13:59:02 +02:00
Torsten Ruger
a8f3d5a634 had to fix the string hash for opal 2016-12-28 13:53:57 +02:00
Torsten Ruger
4940bc41a3 keep positions globally, not in the object
hopefully cleaner switch after bootstrapping
2016-12-28 12:51:18 +02:00
Torsten Ruger
903fc3e4cf much ripples from changing the calling convention
See previous commit
Now args and locals are arrays in the Message
2016-12-27 20:39:39 +02:00
Torsten Ruger
f1cfd3c379 change calling convention
Message carries arrays for arguments and locals
This means an extra load for any access
Also fix silly bug: using the wrong method (the one we’re in, not the
one we’re calling)
2016-12-27 20:37:53 +02:00
Torsten Ruger
aaa94f0743 small fixes 2016-12-27 20:34:11 +02:00
Torsten Ruger
f648bf7bd5 rename also get_slot, to slot_to_reg
makes source and target clear
2016-12-25 18:05:39 +02:00
Torsten Ruger
35adf9a5e6 rename set_slot
set_slot was clear about the target, but not the source.
Better with reg_to_slot (and soon it’s inverse slot_to_reg)
2016-12-25 18:02:39 +02:00
Torsten Ruger
1b8d6149dd adding to_s to code for debugging 2016-12-23 21:31:31 +02:00
Torsten Ruger
6c9bd7e476 remove unused code
also just came through opal
2016-12-22 21:16:56 +02:00
Torsten Ruger
3b888b203d crude rework of the hashing to work in opal 2016-12-22 21:16:36 +02:00
Torsten Ruger
2adec3226c opal gods require explicit requires 2016-12-22 21:15:20 +02:00
Torsten Ruger
c3c52d36b8 seperate args from message (just like frame) 2016-12-21 22:35:36 +02:00
Torsten Ruger
ed12108425 rename locals from type to role in message 2016-12-21 19:01:42 +02:00
Torsten Ruger
0e99ff33ae use frame type rather than the indexed construct 2016-12-21 18:59:07 +02:00
Torsten Ruger
93ba5543b3 more renaming of frame 2016-12-21 18:51:22 +02:00