mlog-arithmetic-runner

MlogArithmeticRunner

A Mindustry Logic emulator built for automated testing of compilers (like MlogEvo).

Install

pip install https://github.com/UMRnInside/MlogArithmeticRunner or

pip install mlog_arithmetic_runner

Usage

$ python3 -m mlog_arithmetic_runner --help
usage: mlog_arithmetic_runner [-h] [--limit LIMIT] [--continue-if-past-the-end] [--ipt IPT]
                              [--memory-banks MEMORY_BANKS] [--memory-cells MEMORY_CELLS]
                              [--json-indent JSON_INDENT] [--json-dump-memory-blocks]

Reads some mlog code from stdin, runs it, and generates JSON report

optional arguments:
  -h, --help            show this help message and exit
  --limit LIMIT         Max instructions/cycles allowed, something like TimeLimit
  --continue-if-past-the-end
  --ipt IPT             Instructions per tick. 2 for micro-processor, 8 for logic-processor,
                        25 for hyper-processor
  --memory-banks MEMORY_BANKS
                        Memory bank count
  --memory-cells MEMORY_CELLS
                        Memory cell count
  --json-indent JSON_INDENT
                        JSON indent, set 0 to disable
  --json-dump-memory-blocks
                        dump all memory content in JSON report

Supported Instructions

• set
• jump
• end
• All ops since Mindustry V7 Beta (beta 140), except op noise
• read and write memory cells and banks
• getlink to get memory blocks (cells/banks)
• set @counter or op @counter as unconditional jumps

Features

• @tick and @time increases with processor runs
• Stop emulation once some instruction jumps to itself
• (Optional, Default) Stop emulation once @counter past the end
• (Optional) Dump memory cells and banks
• Dump variables into JSON report

Limitations

• Does NOT have @this yet, so do @thisx and @thisy
• No comment in mlog code
• (not sure)

Python Usage

from mlog_arithmetic_runner import MlogProcessor
processor = MlogProcessor(ipt=2, memory_cells=0, memory_banks=0)
code = """\
set a 90
op sin b a 0
"""
processor.assemble_code(code)
processor.run_with_limit(1000)
# When comparing 2 float-point numbers, remember there could be float precision errors.
# Use abs(a-b) < 1e-6 instead of a == b
print("a =", processor.get_variable("a"))
print("b =", processor.get_variable("b"))

JSON Report Sample

set a 1
set a 2
set b @tick
set c @time

This generates:

{
    "cycles": 4,
    "success": true,
    "reason": "",
    "variables": {
        "a": 2.0,
        "b": 1,
        "c": 0.016666666666666666
    },
    "memory_blocks": {}
}