Xeron Du

13 papers A* 5Journal 3Unranked 5
YearRankTypeTitle / Venue / Authors
2026 — conf
COIG-P: A High-Quality and Large-Scale Chinese Preference Dataset for Alignment with Human Values.
EACL (Findings)
Siwei Wu, Jincheng Ren, Xeron Du, Shuyue Guo, Xingwei Qu, Yiming Liang, Jie Liu, Yunwen Li, Tyler Loakman, Tianyu Zheng, Boyu Feng, Huaqing Yuan, Zili Wang, Jiaheng Liu, Wenhao Huang, Chenglin Cai, Haoran Que, Jian Yang, Yuelin Bai, Zekun Moore Wang, Zhouliang Yu, Qunshu Lin, Ding Pan, Yuchen Eleanor Jiang, Tiannan Wang, Wangchunshu Zhou, Shenzhi Wang, Xingyuan Bu, Minghao Liu, Guoyin Wang, Ge Zhang, Chenghua Lin
2026 J jnl
Retrieval-Infused Reasoning Sandbox: A Benchmark for Decoupling Retrieval and Reasoning Capabilities.
CoRR
Shuangshuang Ying, Zheyu Wang, Yunjian Peng, Jin Chen, Yuhao Wu, Hongbin Lin, Dingyu He, Siyi Liu, Gengchen Yu, Yinzhu Piao, Yuchen Wu, Xin Gui, Zhongyuan Peng, Xin Li, Xeron Du, Libo Qin, Yixin Cao, Ge Zhang, Stephen Huang
2025 J jnl
Beyond Correctness: Evaluating Subjective Writing Preferences Across Cultures.
CoRR
Shuangshuang Ying, Yunwen Li, Xingwei Qu, Xin Li, Sheng Jin, Minghao Liu, Zhoufutu Wen, Xeron Du, Tianyu Zheng, Yichi Zhang, Letian Ni, Yuyang Cheng, Qiguang Chen, Jingzhe Ding, Shengda Long, Wangchunshu Zhou, Jiazhan Feng, Wanjun Zhong, Libo Qin, Ge Zhang, Wenhao Huang, Wanxiang Che, Chenghua Lin
2025 — conf
COIG-CQIA: Quality is All You Need for Chinese Instruction Fine-tuning.
NAACL (Findings)
Yuelin Bai, Xeron Du, Yiming Liang, Leo Jin, Junting Zhou, Ziqiang Liu, Feiteng Fang, Mingshan Chang, Tianyu Zheng, Xincheng Zhang, Nuo Ma, Zekun Moore Wang, Ruibin Yuan, Haihong Wu, Hongquan Lin, Wenhao Huang, Jiajun Zhang, Chenghua Lin, Jie Fu, Min Yang, Shiwen Ni, Ge Zhang
2025 J jnl
COIG-Writer: A High-Quality Dataset for Chinese Creative Writing with Thought Processes.
CoRR
Yunwen Li, Shuangshuang Ying, Xingwei Qu, Xin Li, Sheng Jin, Minghao Liu, Zhoufutu Wen, Tianyu Zheng, Xeron Du, Qiguang Chen, Jiajun Shi, Wangchunshu Zhou, Jiazhan Feng, Wanjun Zhong, Libo Qin, Stephen Huang, Wanxiang Che, Chenghua Lin, Eli Zhang
2025 — conf
Can MLLMs Understand the Deep Implication Behind Chinese Images?
ACL (1)
Chenhao Zhang, Xi Feng, Yuelin Bai, Xeron Du, Jinchang Hou, Kaixin Deng, Guangzeng Han, Qinrui Li, Bingli Wang, Jiaheng Liu, Xingwei Qu, Yifei Zhang, Qixuan Zhao, Yiming Liang, Ziqiang Liu, Feiteng Fang, Min Yang, Wenhao Huang, Chenghua Lin, Ge Zhang, Shiwen Ni
2025 — conf
KARPA: A Training-free Method of Adapting Knowledge Graph as References for Large Language Model's Reasoning Path Aggregation.
ACL (Findings)
Siyuan Fang, Kaijing Ma, Tianyu Zheng, Xeron Du, Ningxuan Lu, Ge Zhang, Qingkun Tang
2025 A* conf
KOR-Bench: Benchmarking Language Models on Knowledge-Orthogonal Reasoning Tasks.
ICLR
Kaijing Ma, Xeron Du, Yunran Wang, Haoran Zhang, Zhoufutu Wen, Xingwei Qu, Jian Yang, Jiaheng Liu, Minghao Liu, Xiang Yue, Wenhao Huang, Ge Zhang
2025 A* conf
MuPT: A Generative Symbolic Music Pretrained Transformer.
ICLR
Xingwei Qu, Yuelin Bai, Yinghao Ma, Ziya Zhou, Ka Man Lo, Jiaheng Liu, Ruibin Yuan, Lejun Min, Xueling Liu, Tianyu Zhang, Xeron Du, Shuyue Guo, Yiming Liang, Yizhi Li, Shangda Wu, Junting Zhou, Tianyu Zheng, Ziyang Ma, Fengze Han, Wei Xue, Gus Xia, Emmanouil Benetos, Xiang Yue, Chenghua Lin, Xu Tan, Wenhao Huang, Jie Fu, Ge Zhang
2025 A* conf
OmniEdit: Building Image Editing Generalist Models Through Specialist Supervision.
ICLR
Cong Wei, Zheyang Xiong, Weiming Ren, Xeron Du, Ge Zhang, Wenhu Chen
2025 — conf
Quantification of Large Language Model Distillation.
ACL (1)
Sunbowen Lee, Junting Zhou, Chang Ao, Kaige Li, Xeron Du, Sirui He, Haihong Wu, Tianci Liu, Jiaheng Liu, Hamid Alinejad-Rokny, Min Yang, Yitao Liang, Zhoufutu Wen, Shiwen Ni
2025 A* conf
TableBench: A Comprehensive and Complex Benchmark for Table Question Answering.
AAAI
Xianjie Wu, Jian Yang, Linzheng Chai, Ge Zhang, Jiaheng Liu, Xeron Du, Di Liang, Daixin Shu, Xianfu Cheng, Tianzhen Sun, Tongliang Li, Zhoujun Li, Guanglin Niu
2024 A* conf
II-Bench: An Image Implication Understanding Benchmark for Multimodal Large Language Models.
NeurIPS
Ziqiang Liu, Feiteng Fang, Xi Feng, Xeron Du, Chenhao Zhang, Noah Wang, Yuelin Bai, Qixuan Zhao, Liyang Fan, Chengguang Gan, Hongquan Lin, Jiaming Li, Yuansheng Ni, Haihong Wu, Yaswanth Narsupalli, Zhigang Zheng, Chengming Li, Xiping Hu, Ruifeng Xu, Xiaojun Chen, Min Yang, Jiaheng Liu, Ruibo Liu, Wenhao Huang, Ge Zhang, Shiwen Ni
tests/unit/test_decompile_analysis.py
← Index tests/unit/test_decompile_analysis.py python 
"""Unit tests (mocked Binary Ninja) for analysis modules:
- bninja/analysis/cfg.py — CFGAnalysis
- bninja/analysis/disassembly.py — DisassemblyAnalysis
- bninja/analysis/low_level.py — LowLevelAnalysis
"""
import sys
import pytest
from unittest.mock import MagicMock

from tests.unit.conftest_binja_stubs import (
    install_binja_stubs,
    BranchType,
    InstructionTextTokenType,
    MockBasicBlock,
    MockEdge,
    MockFunction,
    MockToken,
    MockDisassemblyLine,
    MockBinaryView,
    MockSymbol,
    SymbolType,
    LowLevelILOperation,
)

install_binja_stubs()

from redb.extractors.decompiler.bninja.analysis.cfg import CFGAnalysis
from redb.extractors.decompiler.bninja.analysis.disassembly import DisassemblyAnalysis
from redb.extractors.decompiler.bninja.arch.x86 import Arch_x86


# ============================================================================
# 9a. CFGAnalysis
# ============================================================================


class TestCFGCyclomaticComplexity:
    def test_cyclomatic_complexity_linear(self):
        """Single block, no edges: E - N + 2 = 0 - 1 + 2 = 1."""
        block = MockBasicBlock(start=0x1000, end=0x1010, outgoing_edges=[])
        func = MockFunction(start=0x1000, basic_blocks=[block])
        cfg = CFGAnalysis(func)
        result = cfg.extract_function_cfg()
        assert result["cyclomatic_complexity"] == 1

    def test_cyclomatic_complexity_branch(self):
        """Diamond: 4 blocks, 4 edges -> 4 - 4 + 2 = 2."""
        entry = MockBasicBlock(start=0x1000, end=0x1010)
        true_b = MockBasicBlock(start=0x1010, end=0x1020)
        false_b = MockBasicBlock(start=0x1020, end=0x1030)
        merge = MockBasicBlock(start=0x1030, end=0x1040)

        entry.outgoing_edges = [MockEdge(target=true_b), MockEdge(target=false_b)]
        true_b.outgoing_edges = [MockEdge(target=merge)]
        false_b.outgoing_edges = [MockEdge(target=merge)]
        merge.outgoing_edges = []

        func = MockFunction(start=0x1000, basic_blocks=[entry, true_b, false_b, merge])
        cfg = CFGAnalysis(func)
        result = cfg.extract_function_cfg()
        assert result["cyclomatic_complexity"] == 2

    def test_cyclomatic_complexity_loop(self):
        """Loop: 3 blocks, 3 edges -> 3 - 3 + 2 = 2."""
        header = MockBasicBlock(start=0x1000, end=0x1010)
        body = MockBasicBlock(start=0x1010, end=0x1020)
        exit_b = MockBasicBlock(start=0x1020, end=0x1030)

        header.outgoing_edges = [MockEdge(target=body), MockEdge(target=exit_b)]
        body.outgoing_edges = [MockEdge(target=header)]
        exit_b.outgoing_edges = []

        func = MockFunction(start=0x1000, basic_blocks=[header, body, exit_b])
        cfg = CFGAnalysis(func)
        result = cfg.extract_function_cfg()
        assert result["cyclomatic_complexity"] == 2


class TestCFGExtractFunctionCFG:
    def _make_simple_cfg(self):
        """Create a simple two-block CFG for testing structure."""
        entry = MockBasicBlock(start=0x1000, end=0x1010)
        exit_b = MockBasicBlock(start=0x1010, end=0x1020)

        entry.outgoing_edges = [MockEdge(source=entry, target=exit_b, edge_type=BranchType.UnconditionalBranch)]
        exit_b.incoming_edges = [MockEdge(source=entry, target=exit_b)]
        exit_b.outgoing_edges = []
        entry.incoming_edges = []

        func = MockFunction(start=0x1000, basic_blocks=[entry, exit_b])
        return func

    def test_extract_function_cfg_structure(self):
        func = self._make_simple_cfg()
        cfg = CFGAnalysis(func)
        result = cfg.extract_function_cfg()
        assert "function_address" not in result
        # New schema: no "blocks" or "measures" nesting
        assert "blocks" not in result
        assert "measures" not in result

    def test_function_cfg_new_keys(self):
        """Assert all expected keys are present in the new output dict."""
        func = self._make_simple_cfg()
        cfg = CFGAnalysis(func)
        result = cfg.extract_function_cfg()
        expected_keys = [
            "cfg_topology_hash",
            "block_count",
            "edge_count",
            "llil_total_operations",
            "call_count",
            "cyclomatic_complexity",
            "loop_count",
            "max_depth",
            "max_fan_out",
            "md_index_topdown",
            "md_index_bottomup",
            "prime_product_llil",
            "cfg_feature_tlsh",
            "wl_minhash",
            "bb_features",
            "cfg_adjacency",
        ]
        for key in expected_keys:
            assert key in result, f"Missing key: {key}"

    def test_returns_none_for_empty_blocks(self):
        func = MockFunction(start=0x1000, basic_blocks=[])
        cfg = CFGAnalysis(func)
        assert cfg.extract_function_cfg() is None


class TestCFGTopologyHash:
    def _make_two_block_cfg(self):
        entry = MockBasicBlock(start=0x1000, end=0x1010)
        exit_b = MockBasicBlock(start=0x1010, end=0x1020)
        entry.outgoing_edges = [MockEdge(target=exit_b)]
        exit_b.outgoing_edges = []
        return MockFunction(start=0x1000, basic_blocks=[entry, exit_b])

    def test_topology_hash_is_16_bytes(self):
        func = self._make_two_block_cfg()
        cfg = CFGAnalysis(func)
        result = cfg.extract_function_cfg()
        assert isinstance(result["cfg_topology_hash"], bytes)
        assert len(result["cfg_topology_hash"]) == 16

    def test_topology_hash_deterministic(self):
        func = self._make_two_block_cfg()
        r1 = CFGAnalysis(func).extract_function_cfg()
        r2 = CFGAnalysis(func).extract_function_cfg()
        assert r1["cfg_topology_hash"] == r2["cfg_topology_hash"]


class TestCFGLoopCount:
    def test_no_loops(self):
        entry = MockBasicBlock(start=0x1000, end=0x1010)
        exit_b = MockBasicBlock(start=0x1010, end=0x1020)
        entry.outgoing_edges = [MockEdge(target=exit_b)]
        exit_b.outgoing_edges = []
        func = MockFunction(start=0x1000, basic_blocks=[entry, exit_b])
        result = CFGAnalysis(func).extract_function_cfg()
        assert result["loop_count"] == 0

    def test_single_loop(self):
        header = MockBasicBlock(start=0x1000, end=0x1010)
        body = MockBasicBlock(start=0x1010, end=0x1020)
        exit_b = MockBasicBlock(start=0x1020, end=0x1030)
        header.outgoing_edges = [MockEdge(target=body), MockEdge(target=exit_b)]
        body.outgoing_edges = [MockEdge(target=header)]
        exit_b.outgoing_edges = []
        func = MockFunction(start=0x1000, basic_blocks=[header, body, exit_b])
        result = CFGAnalysis(func).extract_function_cfg()
        assert result["loop_count"] == 1


class TestCFGMaxDepth:
    def test_max_depth_linear(self):
        entry = MockBasicBlock(start=0x1000, end=0x1010)
        b1 = MockBasicBlock(start=0x1010, end=0x1020)
        b2 = MockBasicBlock(start=0x1020, end=0x1030)
        entry.outgoing_edges = [MockEdge(target=b1)]
        b1.outgoing_edges = [MockEdge(target=b2)]
        b2.outgoing_edges = []
        func = MockFunction(start=0x1000, basic_blocks=[entry, b1, b2])
        result = CFGAnalysis(func).extract_function_cfg()
        assert result["max_depth"] == 2

    def test_max_depth_single_block(self):
        block = MockBasicBlock(start=0x1000, end=0x1010, outgoing_edges=[])
        func = MockFunction(start=0x1000, basic_blocks=[block])
        result = CFGAnalysis(func).extract_function_cfg()
        assert result["max_depth"] == 0


class TestCFGCollectBlockLlilOps:
    """Test that _collect_block_llil_ops correctly maps LLIL data to native blocks."""

    def test_llil_fields_nonzero_with_mock_llil(self):
        """When LLIL is available, llil_total_operations and call_count should be non-zero."""
        # Two native blocks
        entry = MockBasicBlock(start=0x1000, end=0x1010)
        exit_b = MockBasicBlock(start=0x1010, end=0x1020)
        entry.outgoing_edges = [MockEdge(target=exit_b)]
        exit_b.outgoing_edges = []

        # LLIL instructions: SET_REG, CALL in first block; STORE, RET in second
        llil_instrs_1 = [
            MockLLILInstruction(LowLevelILOperation.LLIL_SET_REG),
            MockLLILInstruction(LowLevelILOperation.LLIL_CALL),
        ]
        llil_instrs_2 = [
            MockLLILInstruction(LowLevelILOperation.LLIL_STORE),
            MockLLILInstruction(LowLevelILOperation.LLIL_RET),
        ]

        # LLIL basic blocks map back to native blocks via source_block
        llil_bb1 = MockLLILBasicBlock(llil_instrs_1, source_block=entry)
        llil_bb2 = MockLLILBasicBlock(llil_instrs_2, source_block=exit_b)
        llil_func = MockLLILFunction([llil_bb1, llil_bb2])

        func = MockFunction(start=0x1000, basic_blocks=[entry, exit_b], llil=llil_func)
        result = CFGAnalysis(func, llil_function=llil_func).extract_function_cfg()

        assert result["llil_total_operations"] == 4
        assert result["call_count"] == 1
        assert result["prime_product_llil"] != 0

    def test_llil_none_gives_zero_fields(self):
        """Without LLIL, LLIL-dependent fields should be zero."""
        block = MockBasicBlock(start=0x1000, end=0x1010, outgoing_edges=[])
        func = MockFunction(start=0x1000, basic_blocks=[block])
        result = CFGAnalysis(func).extract_function_cfg()

        assert result["llil_total_operations"] == 0
        assert result["call_count"] == 0
        assert result["prime_product_llil"] == 0

    def test_bb_features_with_llil(self):
        """bb_features should reflect LLIL instruction categories when LLIL is available."""
        block = MockBasicBlock(start=0x1000, end=0x1010, outgoing_edges=[])

        llil_instrs = [
            MockLLILInstruction(LowLevelILOperation.LLIL_ADD),
            MockLLILInstruction(LowLevelILOperation.LLIL_LOAD),
            MockLLILInstruction(LowLevelILOperation.LLIL_CALL),
        ]
        llil_bb = MockLLILBasicBlock(llil_instrs, source_block=block)
        llil_func = MockLLILFunction([llil_bb])

        func = MockFunction(start=0x1000, basic_blocks=[block])
        result = CFGAnalysis(func, llil_function=llil_func).extract_function_cfg()

        feats = result["bb_features"]
        assert len(feats) == 1
        assert feats[0][0] == 3  # instruction count = 3
        # At least one non-zero category count (not all OTHER)
        category_counts = feats[0][1:7]
        assert sum(category_counts) > 0


# ============================================================================
# 9b. DisassemblyAnalysis
# ============================================================================


class TestDisassemblyAnalysisGetJson:
    def _make_analysis(self, instructions=None, basic_blocks=None):
        arch = Arch_x86()
        if instructions is None:
            instructions = [
                (
                    [
                        MockToken("push", InstructionTextTokenType.InstructionToken),
                        MockToken(" ", InstructionTextTokenType.TextToken),
                        MockToken("rbp", InstructionTextTokenType.RegisterToken),
                    ],
                    0x1000,
                ),
                (
                    [
                        MockToken("mov", InstructionTextTokenType.InstructionToken),
                        MockToken(" ", InstructionTextTokenType.TextToken),
                        MockToken("rsp", InstructionTextTokenType.RegisterToken),
                    ],
                    0x1003,
                ),
            ]
        if basic_blocks is None:
            basic_blocks = [MockBasicBlock(
                start=0x1000, end=0x1010,
                disassembly_text=[MockDisassemblyLine([MockToken("push rbp")])]
            )]

        func = MockFunction(
            name="test_func",
            start=0x1000,
            basic_blocks=basic_blocks,
            instructions=instructions,
            symbol=MockSymbol(symbol_type=SymbolType.FunctionSymbol, name="test_func"),
            stack_adjustment=MagicMock(value=-8),
            mlil=None,
        )
        bv = MockBinaryView()
        logger = MagicMock()
        return DisassemblyAnalysis(arch, func, bv, logger)

    def test_get_json_basic_structure(self):
        da = self._make_analysis()
        result, errors = da.get_json()
        expected_keys = [
            "disassembled_function_hash",
            "disassembled_function",
            "disassembled_function_no_addresses",
            "disassembled_function_name",
            "disassembled_function_address",
            "instructions_count",
            "function_type",
            "instructions_types",
            "control_flow_count",
            "memory_access_pattern",
            "register_usage",
            "data_references_count",
        ]
        for key in expected_keys:
            assert key in result, f"Missing key: {key}"

    def test_get_json_hash_deterministic(self):
        da = self._make_analysis()
        r1, _ = da.get_json()
        da2 = self._make_analysis()
        r2, _ = da2.get_json()
        assert r1["disassembled_function_hash"] == r2["disassembled_function_hash"]


class TestDisassemblyCollectInstructionTypes:
    def test_collect_instruction_types(self):
        arch = Arch_x86()
        instructions = [
            ([MockToken("MOV", InstructionTextTokenType.InstructionToken)], 0x1000),
            ([MockToken("ADD", InstructionTextTokenType.InstructionToken)], 0x1001),
            ([MockToken("MOV", InstructionTextTokenType.InstructionToken)], 0x1002),
        ]
        func = MockFunction(start=0x1000, instructions=instructions, symbol=MockSymbol())
        da = DisassemblyAnalysis(arch, func, MockBinaryView(), MagicMock())
        types = da.collect_instruction_types()
        assert "DATA_MOVEMENT" in types
        assert "ARITHMETIC" in types

    def test_collect_instruction_types_empty(self):
        arch = Arch_x86()
        func = MockFunction(start=0x1000, instructions=[], symbol=MockSymbol())
        da = DisassemblyAnalysis(arch, func, MockBinaryView(), MagicMock())
        types = da.collect_instruction_types()
        assert types == {}


class TestDisassemblyMemoryPatterns:
    def _make_memory_instruction(self, tokens):
        return ([t for t in tokens], 0x1000)

    def test_collect_memory_patterns_stack(self):
        arch = Arch_x86()
        tokens = [
            MockToken("[", InstructionTextTokenType.BeginMemoryOperandToken),
            MockToken("RSP", InstructionTextTokenType.RegisterToken),
            MockToken("+0x8", InstructionTextTokenType.TextToken),
            MockToken("]", InstructionTextTokenType.EndMemoryOperandToken),
        ]
        instructions = [self._make_memory_instruction(tokens)]
        func = MockFunction(start=0x1000, instructions=instructions, symbol=MockSymbol())
        da = DisassemblyAnalysis(arch, func, MockBinaryView(), MagicMock())
        patterns = da.collect_memory_patterns()
        assert "MEM_STACK" in patterns

    def test_collect_memory_patterns_direct(self):
        arch = Arch_x86()
        tokens = [
            MockToken("[", InstructionTextTokenType.BeginMemoryOperandToken),
            MockToken("0x402000", InstructionTextTokenType.TextToken),
            MockToken("]", InstructionTextTokenType.EndMemoryOperandToken),
        ]
        instructions = [self._make_memory_instruction(tokens)]
        func = MockFunction(start=0x1000, instructions=instructions, symbol=MockSymbol())
        da = DisassemblyAnalysis(arch, func, MockBinaryView(), MagicMock())
        patterns = da.collect_memory_patterns()
        assert "MEM_DIRECT" in patterns

    def test_collect_memory_patterns_scaled(self):
        arch = Arch_x86()
        tokens = [
            MockToken("[", InstructionTextTokenType.BeginMemoryOperandToken),
            MockToken("RAX+RCX*4", InstructionTextTokenType.TextToken),
            MockToken("]", InstructionTextTokenType.EndMemoryOperandToken),
        ]
        instructions = [self._make_memory_instruction(tokens)]
        func = MockFunction(start=0x1000, instructions=instructions, symbol=MockSymbol())
        da = DisassemblyAnalysis(arch, func, MockBinaryView(), MagicMock())
        patterns = da.collect_memory_patterns()
        assert "MEM_SCALED_INDEX" in patterns

    def test_collect_memory_patterns_base_offset(self):
        arch = Arch_x86()
        tokens = [
            MockToken("[", InstructionTextTokenType.BeginMemoryOperandToken),
            MockToken("RAX+0x10", InstructionTextTokenType.TextToken),
            MockToken("]", InstructionTextTokenType.EndMemoryOperandToken),
        ]
        instructions = [self._make_memory_instruction(tokens)]
        func = MockFunction(start=0x1000, instructions=instructions, symbol=MockSymbol())
        da = DisassemblyAnalysis(arch, func, MockBinaryView(), MagicMock())
        patterns = da.collect_memory_patterns()
        assert "MEM_BASE_OFFSET" in patterns


class TestDisassemblyRegisterUsage:
    def test_collect_register_usage_gpr(self):
        arch = Arch_x86()
        instructions = [
            ([MockToken("RAX", InstructionTextTokenType.RegisterToken)], 0x1000),
        ]
        func = MockFunction(start=0x1000, instructions=instructions, symbol=MockSymbol())
        da = DisassemblyAnalysis(arch, func, MockBinaryView(), MagicMock())
        regs = da.collect_register_usage()
        assert "GPR" in regs

    def test_collect_register_usage_simd(self):
        arch = Arch_x86()
        instructions = [
            ([MockToken("XMM0", InstructionTextTokenType.RegisterToken)], 0x1000),
        ]
        func = MockFunction(start=0x1000, instructions=instructions, symbol=MockSymbol())
        da = DisassemblyAnalysis(arch, func, MockBinaryView(), MagicMock())
        regs = da.collect_register_usage()
        assert "SIMD" in regs

    def test_collect_register_usage_fpu(self):
        arch = Arch_x86()
        instructions = [
            ([MockToken("ST0", InstructionTextTokenType.RegisterToken)], 0x1000),
        ]
        func = MockFunction(start=0x1000, instructions=instructions, symbol=MockSymbol())
        da = DisassemblyAnalysis(arch, func, MockBinaryView(), MagicMock())
        regs = da.collect_register_usage()
        assert "FPU" in regs


class TestDisassemblyMisc:
    def test_count_data_references(self):
        arch = Arch_x86()
        func = MockFunction(start=0x1000, instructions=[], symbol=MockSymbol(), mlil=None)
        da = DisassemblyAnalysis(arch, func, MockBinaryView(), MagicMock())
        assert da.count_data_references() == 0

    def test_compute_max_block_size(self):
        arch = Arch_x86()
        blocks = [
            MockBasicBlock(disassembly_text=[MockDisassemblyLine([]) for _ in range(3)]),
            MockBasicBlock(disassembly_text=[MockDisassemblyLine([]) for _ in range(5)]),
        ]
        func = MockFunction(start=0x1000, basic_blocks=blocks, instructions=[], symbol=MockSymbol())
        da = DisassemblyAnalysis(arch, func, MockBinaryView(), MagicMock())
        assert da.compute_max_block_size() == 5

    def test_compute_num_calls(self):
        arch = Arch_x86()
        instructions = [
            ([MockToken("CALL", InstructionTextTokenType.InstructionToken)], 0x1000),
            ([MockToken("MOV", InstructionTextTokenType.InstructionToken)], 0x1005),
            ([MockToken("CALL", InstructionTextTokenType.InstructionToken)], 0x1010),
        ]
        func = MockFunction(start=0x1000, instructions=instructions, symbol=MockSymbol())
        da = DisassemblyAnalysis(arch, func, MockBinaryView(), MagicMock())
        assert da.compute_num_calls() == 2

    def test_estimate_stack_size_value(self):
        arch = Arch_x86()
        stack = MagicMock()
        stack.value = -16
        func = MockFunction(start=0x1000, instructions=[], symbol=MockSymbol(), stack_adjustment=stack)
        da = DisassemblyAnalysis(arch, func, MockBinaryView(), MagicMock())
        assert da.estimate_stack_size() == -16

    def test_estimate_stack_size_int(self):
        arch = Arch_x86()
        func = MockFunction(start=0x1000, instructions=[], symbol=MockSymbol(), stack_adjustment=-8)
        da = DisassemblyAnalysis(arch, func, MockBinaryView(), MagicMock())
        assert da.estimate_stack_size() == -8

    def test_normalize_opcode(self):
        arch = Arch_x86()
        func = MockFunction(start=0x1000, instructions=[], symbol=MockSymbol())
        da = DisassemblyAnalysis(arch, func, MockBinaryView(), MagicMock())
        assert da.normalize_opcode("mov") == "MOV"
        assert da.normalize_opcode("PUSH") == "PUSH"


# ============================================================================
# 9c. LowLevelAnalysis (basic tests with mocked LLIL)
# ============================================================================


class MockLLILInstruction:
    """Mock LLIL instruction for low_level.py tests."""
    def __init__(self, operation, operands=None, address=0):
        self.operation = operation
        self.operands = operands or []
        self.address = address

    def __str__(self):
        return f"LLIL_{self.operation}"


class MockLLILBasicBlock:
    def __init__(self, instructions, source_block=None):
        self._instructions = instructions
        self.source_block = source_block

    def __iter__(self):
        return iter(self._instructions)


class MockLLILFunction:
    def __init__(self, basic_blocks):
        self.basic_blocks = basic_blocks
        self._instructions = []
        for bb in basic_blocks:
            self._instructions.extend(bb._instructions)

    @property
    def instructions(self):
        return iter(self._instructions)

    @property
    def source_function(self):
        mock = MagicMock()
        mock.start = 0x1000
        return mock


class TestLowLevelAnalysisCountControlFlow:
    def test_count_control_flow_instructions(self):
        from redb.extractors.decompiler.bninja.analysis.low_level import LowLevelAnalysis
        instrs = [
            MockLLILInstruction(LowLevelILOperation.LLIL_IF),
            MockLLILInstruction(LowLevelILOperation.LLIL_SET_REG),
            MockLLILInstruction(LowLevelILOperation.LLIL_CALL),
            MockLLILInstruction(LowLevelILOperation.LLIL_GOTO),
        ]
        bb = MockLLILBasicBlock(instrs)
        llil_func = MockLLILFunction([bb])

        func = MockFunction(start=0x1000, llil=llil_func, symbol=MockSymbol())
        func.low_level_il = None
        bv = MockBinaryView()
        bv.arch = MagicMock()
        bv.arch.stack_pointer = "sp"
        la = LowLevelAnalysis(func, bv, MagicMock())
        assert la.count_control_flow_instructions() == 3  # IF, CALL, GOTO


class TestLowLevelAnalysisNumCalls:
    def test_compute_num_calls_llil(self):
        from redb.extractors.decompiler.bninja.analysis.low_level import LowLevelAnalysis
        instrs = [
            MockLLILInstruction(LowLevelILOperation.LLIL_CALL),
            MockLLILInstruction(LowLevelILOperation.LLIL_TAILCALL),
            MockLLILInstruction(LowLevelILOperation.LLIL_SET_REG),
        ]
        bb = MockLLILBasicBlock(instrs)
        llil_func = MockLLILFunction([bb])

        func = MockFunction(start=0x1000, llil=llil_func, symbol=MockSymbol())
        func.low_level_il = None
        bv = MockBinaryView()
        la = LowLevelAnalysis(func, bv, MagicMock())
        assert la.compute_num_calls() == 2


class TestLowLevelAnalysisCollectNormalization:
    def test_collect_low_level(self):
        from redb.extractors.decompiler.bninja.analysis.low_level import LowLevelAnalysis
        instrs = [
            MockLLILInstruction(LowLevelILOperation.LLIL_SET_REG, address=0x1000),
            MockLLILInstruction(LowLevelILOperation.LLIL_STORE, address=0x1004),
        ]
        bb = MockLLILBasicBlock(instrs)
        llil_func = MockLLILFunction([bb])

        func = MockFunction(start=0x1000, llil=llil_func, symbol=MockSymbol())
        func.low_level_il = None
        bv = MockBinaryView()
        la = LowLevelAnalysis(func, bv, MagicMock())
        result, _ = la._collect_low_level_and_with_addr()
        assert len(result) == 2
        # Each item is a list of operation ints
        assert isinstance(result[0], list)

    def test_collect_low_level_with_addr_offset_clamping(self):
        from redb.extractors.decompiler.bninja.analysis.low_level import LowLevelAnalysis
        instrs = [
            MockLLILInstruction(LowLevelILOperation.LLIL_SET_REG, address=0x0FFF),  # Before function start
        ]
        bb = MockLLILBasicBlock(instrs)
        llil_func = MockLLILFunction([bb])

        func = MockFunction(start=0x1000, llil=llil_func, symbol=MockSymbol())
        func.low_level_il = None
        bv = MockBinaryView()
        la = LowLevelAnalysis(func, bv, MagicMock())
        _, result = la._collect_low_level_and_with_addr()
        assert len(result) == 1
        offset, _ = result[0]
        assert offset == 0  # Clamped to 0