2125_GCE/tests/test_text_extraction.py

#!/usr/bin/env python3
"""Tests for GX/LAIN text extraction in gx_lain/runtime.py

Tests verify:
- normalize_segments extracts text from decompressed payload via byte ranges
- _infer_lane assigns correct lanes (0-7) based on segment metadata
- _infer_semantic_role returns correct role strings
- map_lanes organizes segments into 8 lanes
- build_envelope constructs proper ExecutionEnvelope
- fuse_lanes merges lane results correctly
- compute_resonance computes per-lane resonance
- render_output_text formats text by cognitive_mode
- make_error constructs structured error dicts
"""

import sys
import json
from pathlib import Path

sys.path.insert(0, str(Path.cwd()))

from gx_lain.runtime import (
    normalize_segments,
    _infer_lane,
    _infer_semantic_role,
    map_lanes,
    build_envelope,
    fuse_lanes,
    compute_resonance,
    render_output_text,
    make_error,
)
from gx_compiler.compressor import GXCompressor


def test_normalize_segments_text_extraction():
    """Test that normalize_segments extracts correct text via byte ranges."""
    source = "def hello():\n    print('world')\n\nresult = 42"
    compressed = GXCompressor.compress(source)

    manifest = {
        "source_file": "test.py",
        "version": "1.0.0",
        "segments": [
            {"id": "seg_0", "start": 0, "end": 2, "start_byte": 0, "end_byte": 27},
            {"id": "seg_1", "start": 2, "end": 3, "start_byte": 28, "end_byte": 41},
        ],
    }

    raw_segments = manifest["segments"]
    result = normalize_segments(manifest, raw_segments, compressed)

    if len(result) != 2:
        print(f"FAIL: Expected 2 segments, got {len(result)}")
        return False

    seg0 = result[0]
    seg1 = result[1]

    if "hello" not in seg0["text"]:
        print(f"FAIL: seg_0 should contain 'hello', got: {repr(seg0['text'])}")
        return False

    if "result" not in seg1["text"]:
        print(f"FAIL: seg_1 should contain 'result', got: {repr(seg1['text'])}")
        return False

    print("PASS: Text extraction via byte ranges works correctly")
    return True


def test_normalize_segments_fallback():
    """Test normalize_segments fallback when no text is extracted."""
    manifest = {"source_file": "test.py", "version": "1.0.0", "segments": []}
    compressed = GXCompressor.compress("test")

    result = normalize_segments(manifest, [], compressed)

    if result:
        print(f"FAIL: Expected empty result for empty segments, got {len(result)}")
        return False

    print("PASS: Empty segments returns empty list")
    return True


def test_normalize_segments_corrupt_payload():
    """Test normalize_segments handles corrupt payload gracefully."""
    manifest = {
        "source_file": "test.py",
        "version": "1.0.0",
        "segments": [
            {"id": "seg_0", "start": 0, "end": 1, "start_byte": 0, "end_byte": 10}
        ],
    }

    result = normalize_segments(manifest, manifest["segments"], b"corrupt")

    if len(result) != 1:
        print(f"FAIL: Expected 1 segment even with corrupt payload, got {len(result)}")
        return False

    if result[0]["text"] and "segment" not in result[0]["text"]:
        print(f"FAIL: Expected fallback text, got: {repr(result[0]['text'])}")
        return False

    print("PASS: Corrupt payload handled gracefully with fallback text")
    return True


def test_infer_lane_structural():
    """Test lane inference for structural segments."""
    assert _infer_lane({"id": "def_main"}) == 0, "def should be lane 0"
    assert _infer_lane({"id": "class_Foo"}) == 0, "class should be lane 0"
    assert _infer_lane({"id": "struct_header"}) == 0, "struct should be lane 0"

    print("PASS: Structural segments correctly mapped to lane 0")
    return True


def test_infer_lane_default():
    """Test lane inference returns default for unknown segments."""
    lane = _infer_lane({"id": "some_random_segment"})

    if lane != 1:
        print(f"FAIL: Expected lane 1 (default), got {lane}")
        return False

    print("PASS: Unclassified segments default to lane 1")
    return True


def test_infer_lane_all_types():
    """Test all lane types are correctly inferred."""
    cases = [
        ({"id": "comment_block"}, 3),
        ({"id": "annotation_section"}, 3),
        ({"id": "hint_optimization"}, 4),
        ({"id": "meta_author"}, 6),
        ({"id": "epoch_v1"}, 7),
        ({"id": "version_2_0"}, 7),
    ]

    for segment, expected in cases:
        actual = _infer_lane(segment)
        if actual != expected:
            print(f"FAIL: {segment['id']} → lane {actual}, expected {expected}")
            return False

    print("PASS: All lane types correctly inferred")
    return True


def test_infer_semantic_role():
    """Test semantic role inference for various segment types."""
    cases = [
        ({"id": "def_calculate"}, "definition"),
        ({"id": "class_User"}, "definition"),
        ({"id": "constraint_validation"}, "constraint"),
        ({"id": "example_usage"}, "example"),
        ({"id": "meta_description"}, "meta"),
        ({"id": "comment_note"}, "meta"),
        ({"id": "unknown_block"}, "unknown"),
    ]

    for segment, expected in cases:
        actual = _infer_semantic_role(segment)
        if actual != expected:
            print(f"FAIL: {segment['id']} → {actual}, expected {expected}")
            return False

    print("PASS: All semantic roles correctly inferred")
    return True


def test_map_lanes():
    """Test mapping segments into 8 lanes."""
    segments = [
        {"id": "s1", "symbolic_lane": 0, "text": "struct"},
        {"id": "s2", "symbolic_lane": 1, "text": "flow"},
        {"id": "s3", "symbolic_lane": 5, "text": "predict"},
        {"id": "s4", "symbolic_lane": 7, "text": "epoch"},
    ]

    lanes = map_lanes(segments)

    if len(lanes) != 8:
        print(f"FAIL: Expected 8 lanes, got {len(lanes)}")
        return False

    if len(lanes[0]) != 1 or len(lanes[1]) != 1:
        print("FAIL: Segments not properly assigned")
        return False

    if lanes[0][0]["id"] != "s1":
        print("FAIL: Lane 0 has wrong segment")
        return False

    print("PASS: Segments correctly mapped to 8 lanes")
    return True


def test_map_lanes_clamp():
    """Test that lane values outside 0-7 are clamped."""
    segments = [
        {"id": "s1", "symbolic_lane": -1, "text": "neg"},
        {"id": "s2", "symbolic_lane": 10, "text": "over"},
    ]

    lanes = map_lanes(segments)
    # -1 clamped to 0, 10 clamped to 7
    if len(lanes[0]) != 1:
        print("FAIL: Negative lane not clamped to 0")
        return False

    if len(lanes[7]) != 1:
        print("FAIL: Over-max lane not clamped to 7")
        return False

    print("PASS: Lane clamping works for values outside 0-7")
    return True


def test_build_envelope():
    """Test envelope construction."""
    manifest = {"source_file": "test.py", "version": "1.0.0", "contributor": "tester"}
    lanes = {0: [{"id": "s1", "text": "test"}]}
    payload = b"test"

    envelope = build_envelope(manifest, lanes, payload, context={"cognitive_mode": "analyze"})

    if envelope["manifest"] != manifest:
        print("FAIL: Manifest not preserved")
        return False

    if envelope["lanes"] != lanes:
        print("FAIL: Lanes not preserved")
        return False

    if envelope["payload"] != payload:
        print("FAIL: Payload not preserved")
        return False

    ctx = envelope["context"]
    if ctx["cognitive_mode"] != "analyze":
        print("FAIL: cognitive_mode not passed through")
        return False

    if ctx["contributor"] != "tester":
        print("FAIL: contributor not from manifest")
        return False

    if "invocation_id" not in ctx:
        print("FAIL: missing invocation_id")
        return False

    print("PASS: Envelope constructed with all fields")
    return True


def test_fuse_lanes():
    """Test lane fusion combines results correctly."""
    lane_results = {
        0: {
            "summary": "Structural analysis",
            "key_points": ["Point A", "Point B"],
            "constraints": ["Constraint 1"],
            "open_questions": ["Question 1"],
        },
        1: {
            "summary": "Flow analysis",
            "key_points": ["Point C"],
            "constraints": [],
            "open_questions": [],
        },
    }

    fused = fuse_lanes(lane_results)

    if "Structural" not in fused["summary"] or "Flow" not in fused["summary"]:
        print(f"FAIL: Summaries not merged: {fused['summary']}")
        return False

    if len(fused["key_points"]) != 3:
        print(f"FAIL: Expected 3 key points, got {len(fused['key_points'])}")
        return False

    if "Constraint 1" not in fused["constraints"]:
        print("FAIL: Constraint not merged")
        return False

    print("PASS: Lane results fused correctly")
    return True


def test_compute_resonance():
    """Test resonance computation."""
    lane_results = {
        0: {"summary": "Has content"},
        1: {"summary": ""},
        2: {"summary": "More content"},
    }

    resonance = compute_resonance(lane_results, {})

    if resonance["lane_0"] != 1.0:
        print(f"FAIL: Lane 0 should have resonance 1.0, got {resonance['lane_0']}")
        return False

    if resonance["lane_1"] != 0.0:
        print(f"FAIL: Lane 1 should have resonance 0.0, got {resonance['lane_1']}")
        return False

    if resonance["lane_2"] != 1.0:
        print(f"FAIL: Lane 2 should have resonance 1.0, got {resonance['lane_2']}")
        return False

    print("PASS: Resonance correctly computed")
    return True


def test_render_output_text():
    """Test output text rendering for different modes."""
    fused = {
        "summary": "Test summary",
        "key_points": ["Key 1", "Key 2"],
        "constraints": ["Const 1"],
        "open_questions": ["Q1"],
    }

    text = render_output_text(fused, {"cognitive_mode": "analyze"})

    if "[ANALYZE]" not in text:
        print(f"FAIL: Missing mode header in: {text}")
        return False

    if "Test summary" not in text:
        print(f"FAIL: Missing summary in: {text}")
        return False

    if "Key 1" not in text:
        print(f"FAIL: Missing key points in: {text}")
        return False

    if "Const 1" not in text:
        print(f"FAIL: Missing constraints in: {text}")
        return False

    if "Q1" not in text:
        print(f"FAIL: Missing open questions in: {text}")
        return False

    print("PASS: Output text rendered correctly")
    return True


def test_make_error():
    """Test structured error construction."""
    err = make_error("DecodeError", "Something failed", segment_id="seg_0", lane=1, recoverable=True)

    if err["type"] != "DecodeError":
        print("FAIL: Error type mismatch")
        return False

    if err["message"] != "Something failed":
        print("FAIL: Error message mismatch")
        return False

    if err["segment_id"] != "seg_0":
        print("FAIL: Error segment_id mismatch")
        return False

    if err["lane"] != 1:
        print("FAIL: Error lane mismatch")
        return False

    if err["recoverable"] is not True:
        print("FAIL: Error recoverable mismatch")
        return False

    print("PASS: Structured error constructed correctly")
    return True


def main_test():
    print("[TEST SUITE] test_text_extraction.py")
    print()

    tests = [
        ("Text extraction via byte ranges", test_normalize_segments_text_extraction),
        ("Empty segments fallback", test_normalize_segments_fallback),
        ("Corrupt payload handling", test_normalize_segments_corrupt_payload),
        ("Structural lane inference", test_infer_lane_structural),
        ("Default lane inference", test_infer_lane_default),
        ("All lane types", test_infer_lane_all_types),
        ("Semantic role inference", test_infer_semantic_role),
        ("Map lanes", test_map_lanes),
        ("Lane clamping", test_map_lanes_clamp),
        ("Build envelope", test_build_envelope),
        ("Fuse lanes", test_fuse_lanes),
        ("Compute resonance", test_compute_resonance),
        ("Render output text", test_render_output_text),
        ("Make error", test_make_error),
    ]

    passed = 0
    failed = 0

    for name, test_func in tests:
        print(f"Running: {name}...", end=" ")
        try:
            if test_func():
                passed += 1
            else:
                failed += 1
        except Exception as e:
            print(f"FAIL: {e}")
            failed += 1

    print()
    print(f"Results: {passed} passed, {failed} failed")

    return 0 if failed == 0 else 1


if __name__ == "__main__":
    sys.exit(main_test())