oscura 0.5.1__py3-none-any.whl → 0.7.0__py3-none-any.whl

oscura/automotive/lin/analyzer.py

@@ -0,0 +1,612 @@
+"""LIN protocol analyzer with enhanced checksum and LDF generation.
+
+This module provides comprehensive LIN 2.x protocol analysis including
+protected ID calculation, enhanced checksum validation, diagnostic frame
+parsing, and LDF (LIN Description File) generation from captured traffic.
+
+References:
+    LIN Specification 2.2A
+    ISO 17987 (LIN standard)
+"""
+
+from __future__ import annotations
+
+from dataclasses import dataclass, field
+from pathlib import Path
+from typing import Any, ClassVar
+
+__all__ = [
+    "LINAnalyzer",
+    "LINFrame",
+    "LINScheduleEntry",
+    "LINSignal",
+]
+
+
+@dataclass
+class LINFrame:
+    """LIN frame representation.
+
+    Attributes:
+        timestamp: Frame timestamp in seconds.
+        frame_id: Frame ID (0-63, 6-bit).
+        data: Frame data bytes (1-8 bytes).
+        checksum: Received checksum byte.
+        checksum_valid: True if checksum is valid.
+        checksum_type: Checksum type ("classic" or "enhanced").
+        parity_bits: 2-bit parity from protected ID.
+        is_diagnostic: True if diagnostic frame (0x3C or 0x3D).
+        decoded_signals: Decoded signal values.
+    """
+
+    timestamp: float
+    frame_id: int
+    data: bytes
+    checksum: int
+    checksum_valid: bool
+    checksum_type: str = "enhanced"
+    parity_bits: int = 0
+    is_diagnostic: bool = False
+    decoded_signals: dict[str, Any] = field(default_factory=dict)
+
+    def __repr__(self) -> str:
+        """Human-readable representation."""
+        status = "✓" if self.checksum_valid else "✗"
+        return (
+            f"LINFrame(ID=0x{self.frame_id:02X}, t={self.timestamp:.6f}s, "
+            f"data={self.data.hex().upper()}, checksum={status} {self.checksum_type})"
+        )
+
+
+@dataclass
+class LINSignal:
+    """LIN signal definition.
+
+    Attributes:
+        name: Signal name.
+        frame_id: Frame ID containing this signal (0-63).
+        start_bit: Starting bit position (0-63 within 8 bytes).
+        bit_length: Signal length in bits (1-64).
+        init_value: Initial/default value.
+        publisher: Node publishing this signal ("Master" or slave name).
+    """
+
+    name: str
+    frame_id: int
+    start_bit: int
+    bit_length: int
+    init_value: int = 0
+    publisher: str = "Master"
+
+
+@dataclass
+class LINScheduleEntry:
+    """LIN schedule table entry.
+
+    Attributes:
+        frame_id: Frame ID to transmit.
+        delay_ms: Delay before next frame in milliseconds.
+    """
+
+    frame_id: int
+    delay_ms: float
+
+
+class LINAnalyzer:
+    """Enhanced LIN protocol analyzer with LDF generation.
+
+    Supports comprehensive LIN 2.x protocol analysis including:
+    - Protected ID calculation with parity bits
+    - Classic and enhanced checksum validation
+    - Diagnostic frame parsing (0x3C master request, 0x3D slave response)
+    - Signal decoding with bit-level extraction
+    - Schedule table inference from frame timing
+    - LDF (LIN Description File) generation
+
+    Example:
+        >>> analyzer = LINAnalyzer()
+        >>> # Parse LIN frame with enhanced checksum
+        >>> frame = analyzer.parse_frame(
+        ...     data=b'\\x55\\x80\\x01\\x02\\x03\\xFA',
+        ...     timestamp=1.0
+        ... )
+        >>> print(f"Frame ID: {frame.frame_id}")
+        Frame ID: 0
+        >>> # Add signal definition
+        >>> analyzer.add_signal(LINSignal(
+        ...     name="Speed",
+        ...     frame_id=0,
+        ...     start_bit=0,
+        ...     bit_length=16,
+        ...     publisher="Master"
+        ... ))
+        >>> # Generate LDF from captured traffic
+        >>> analyzer.generate_ldf(Path("output.ldf"), baudrate=19200)
+    """
+
+    # Diagnostic frame IDs per LIN 2.x specification
+    MASTER_REQUEST_FRAME: ClassVar[int] = 0x3C  # 60
+    SLAVE_RESPONSE_FRAME: ClassVar[int] = 0x3D  # 61
+
+    # Diagnostic services (subset of UDS adapted for LIN)
+    DIAGNOSTIC_SERVICES: ClassVar[dict[int, str]] = {
+        0xB0: "AssignFrameIdRange",
+        0xB1: "AssignNAD",
+        0xB2: "ConditionalChangeNAD",
+        0xB3: "DataDump",
+        0xB4: "SaveConfiguration",
+        0xB5: "AssignFrameId",
+        0xB6: "ReadById",
+        0xB7: "TargetedReset",
+    }
+
+    def __init__(self) -> None:
+        """Initialize LIN analyzer."""
+        self.frames: list[LINFrame] = []
+        self.signals: list[LINSignal] = []
+        self.schedule: list[LINScheduleEntry] = []
+        self.detected_frame_ids: set[int] = set()
+
+    def parse_frame(
+        self,
+        data: bytes,
+        timestamp: float = 0.0,
+        checksum_type: str = "enhanced",
+    ) -> LINFrame:
+        """Parse LIN frame including sync, protected ID, data, and checksum.
+
+        LIN Frame Format:
+        - Break field (dominant, >= 13 bit times) - not in data
+        - Sync field (0x55) - first byte
+        - Protected ID (frame ID + parity bits) - second byte
+        - Data field (1-8 bytes) - variable
+        - Checksum field (1 byte) - last byte
+
+        Args:
+            data: Raw frame bytes including sync, protected ID, data, checksum.
+            timestamp: Frame timestamp in seconds.
+            checksum_type: Checksum type ("classic" or "enhanced").
+
+        Returns:
+            Parsed LINFrame object.
+
+        Raises:
+            ValueError: If frame is invalid or too short.
+
+        Example:
+            >>> analyzer = LINAnalyzer()
+            >>> # Frame: sync=0x55, protected_id=0x80 (ID=0), data=0x01,0x02, checksum=0xFA
+            >>> frame = analyzer.parse_frame(
+            ...     data=b'\\x55\\x80\\x01\\x02\\xFA',
+            ...     timestamp=1.0,
+            ...     checksum_type="enhanced"
+            ... )
+            >>> print(f"Frame ID: {frame.frame_id}, Valid: {frame.checksum_valid}")
+            Frame ID: 0, Valid: True
+        """
+        if len(data) < 4:  # Minimum: sync + protected_id + 1 data byte + checksum
+            raise ValueError(f"LIN frame too short: {len(data)} bytes (minimum 4)")
+
+        # Parse sync byte (should be 0x55)
+        sync_byte = data[0]
+        if sync_byte != 0x55:
+            raise ValueError(f"Invalid sync byte: 0x{sync_byte:02X} (expected 0x55)")
+
+        # Parse protected ID
+        protected_id = data[1]
+        frame_id = protected_id & 0x3F  # Lower 6 bits
+        parity_bits = (protected_id >> 6) & 0x03  # Upper 2 bits
+
+        # Validate protected ID parity
+        expected_protected_id = self._calculate_protected_id(frame_id)
+        if protected_id != expected_protected_id:
+            raise ValueError(
+                f"Invalid protected ID parity: 0x{protected_id:02X} "
+                f"(expected 0x{expected_protected_id:02X} for frame ID {frame_id})"
+            )
+
+        # Parse data and checksum
+        frame_data = data[2:-1]
+        received_checksum = data[-1]
+
+        # Calculate and validate checksum
+        if checksum_type == "classic":
+            expected_checksum = self._calculate_classic_checksum(frame_data)
+        else:  # enhanced
+            expected_checksum = self._calculate_enhanced_checksum(protected_id, frame_data)
+
+        checksum_valid = received_checksum == expected_checksum
+
+        # Check if diagnostic frame
+        is_diagnostic = frame_id in (self.MASTER_REQUEST_FRAME, self.SLAVE_RESPONSE_FRAME)
+
+        # Parse diagnostic frame if applicable
+        decoded_signals: dict[str, Any] = {}
+        if is_diagnostic:
+            decoded_signals = self._parse_diagnostic_frame(frame_id, frame_data)
+        else:
+            # Decode regular signals
+            decoded_signals = self.decode_signals(frame_id, frame_data)
+
+        frame = LINFrame(
+            timestamp=timestamp,
+            frame_id=frame_id,
+            data=frame_data,
+            checksum=received_checksum,
+            checksum_valid=checksum_valid,
+            checksum_type=checksum_type,
+            parity_bits=parity_bits,
+            is_diagnostic=is_diagnostic,
+            decoded_signals=decoded_signals,
+        )
+
+        self.frames.append(frame)
+        self.detected_frame_ids.add(frame_id)
+
+        return frame
+
+    def _calculate_protected_id(self, frame_id: int) -> int:
+        """Calculate protected ID with parity bits.
+
+        Protected ID Format (8 bits):
+        - Bits 0-5: Frame ID (6 bits, range 0-63)
+        - Bit 6 (P0): ID0 XOR ID1 XOR ID2 XOR ID4
+        - Bit 7 (P1): NOT(ID1 XOR ID3 XOR ID4 XOR ID5)
+
+        Args:
+            frame_id: Frame ID (0-63).
+
+        Returns:
+            Protected ID with parity bits (8 bits).
+
+        Raises:
+            ValueError: If frame ID exceeds 63.
+
+        Example:
+            >>> analyzer = LINAnalyzer()
+            >>> protected_id = analyzer._calculate_protected_id(0)
+            >>> print(f"Protected ID: 0x{protected_id:02X}")
+            Protected ID: 0x80
+        """
+        if frame_id > 0x3F:
+            raise ValueError(f"Frame ID {frame_id} exceeds 6 bits (max 63)")
+
+        # Extract individual ID bits
+        id0 = (frame_id >> 0) & 1
+        id1 = (frame_id >> 1) & 1
+        id2 = (frame_id >> 2) & 1
+        id3 = (frame_id >> 3) & 1
+        id4 = (frame_id >> 4) & 1
+        id5 = (frame_id >> 5) & 1
+
+        # Calculate parity bits
+        p0 = id0 ^ id1 ^ id2 ^ id4
+        p1 = (id1 ^ id3 ^ id4 ^ id5) ^ 1  # Inverted
+
+        # Construct protected ID
+        protected_id = frame_id | (p0 << 6) | (p1 << 7)
+
+        return protected_id
+
+    def _calculate_classic_checksum(self, data: bytes) -> int:
+        """Calculate classic checksum (LIN 1.x).
+
+        Classic checksum = inverted modulo-256 sum of data bytes only.
+
+        Args:
+            data: Frame data bytes.
+
+        Returns:
+            Classic checksum byte.
+
+        Example:
+            >>> analyzer = LINAnalyzer()
+            >>> checksum = analyzer._calculate_classic_checksum(b'\\x01\\x02\\x03')
+            >>> print(f"Checksum: 0x{checksum:02X}")
+            Checksum: 0xF9
+        """
+        checksum_sum = sum(data)
+
+        # Handle carry (modulo-256 with carry propagation)
+        while checksum_sum > 255:
+            checksum_sum = (checksum_sum & 0xFF) + (checksum_sum >> 8)
+
+        checksum = (~checksum_sum) & 0xFF
+
+        return checksum
+
+    def _calculate_enhanced_checksum(self, protected_id: int, data: bytes) -> int:
+        """Calculate enhanced checksum (LIN 2.x).
+
+        Enhanced checksum = inverted modulo-256 sum of protected ID + data bytes.
+
+        Args:
+            protected_id: Protected ID byte (frame ID with parity).
+            data: Frame data bytes.
+
+        Returns:
+            Enhanced checksum byte.
+
+        Example:
+            >>> analyzer = LINAnalyzer()
+            >>> checksum = analyzer._calculate_enhanced_checksum(0x80, b'\\x01\\x02\\x03')
+            >>> print(f"Checksum: 0x{checksum:02X}")
+            Checksum: 0x79
+        """
+        checksum_sum = protected_id
+
+        for byte in data:
+            checksum_sum += byte
+            # Handle carry (modulo-256 with carry propagation)
+            if checksum_sum > 255:
+                checksum_sum = (checksum_sum & 0xFF) + 1
+
+        checksum = (~checksum_sum) & 0xFF
+
+        return checksum
+
+    def _parse_diagnostic_frame(self, frame_id: int, data: bytes) -> dict[str, Any]:
+        """Parse diagnostic frame (Master Request 0x3C or Slave Response 0x3D).
+
+        Diagnostic Frame Format:
+        - NAD (Node Address for Diagnostics) - 1 byte
+        - PCI (Protocol Control Information) - 1 byte
+        - SID (Service Identifier) - 1 byte
+        - Service data - variable (up to 5 bytes)
+
+        Args:
+            frame_id: Frame ID (0x3C or 0x3D).
+            data: Frame data bytes.
+
+        Returns:
+            Dictionary of decoded diagnostic fields.
+
+        Example:
+            >>> analyzer = LINAnalyzer()
+            >>> decoded = analyzer._parse_diagnostic_frame(
+            ...     0x3C,
+            ...     b'\\x01\\x06\\xB6\\x00\\x01\\x00\\x00\\x00'
+            ... )
+            >>> print(decoded["service_name"])
+            ReadById
+        """
+        if len(data) < 3:
+            return {"error": "Diagnostic frame too short"}
+
+        nad = data[0]  # Node address
+        pci = data[1]  # Protocol control info (single frame = 0x06)
+        sid = data[2]  # Service ID
+
+        service_name = self.DIAGNOSTIC_SERVICES.get(sid, f"Unknown (0x{sid:02X})")
+        service_data = data[3:]
+
+        result: dict[str, Any] = {
+            "nad": nad,
+            "pci": pci,
+            "service_id": sid,
+            "service_name": service_name,
+            "service_data": service_data.hex().upper(),
+            "frame_type": "MasterRequest"
+            if frame_id == self.MASTER_REQUEST_FRAME
+            else "SlaveResponse",
+        }
+
+        # Decode specific services
+        if sid == 0xB6 and len(service_data) >= 2:  # ReadById
+            identifier = int.from_bytes(service_data[0:2], "big")
+            result["identifier"] = identifier
+
+        return result
+
+    def add_signal(self, signal: LINSignal) -> None:
+        """Add signal definition for decoding.
+
+        Args:
+            signal: Signal definition to add.
+
+        Example:
+            >>> analyzer = LINAnalyzer()
+            >>> analyzer.add_signal(LINSignal(
+            ...     name="EngineSpeed",
+            ...     frame_id=0x10,
+            ...     start_bit=0,
+            ...     bit_length=16,
+            ...     publisher="Master"
+            ... ))
+        """
+        self.signals.append(signal)
+
+    def decode_signals(self, frame_id: int, data: bytes) -> dict[str, Any]:
+        """Decode signals from frame data.
+
+        Args:
+            frame_id: Frame ID.
+            data: Frame data bytes.
+
+        Returns:
+            Dictionary mapping signal names to decoded values.
+
+        Example:
+            >>> analyzer = LINAnalyzer()
+            >>> analyzer.add_signal(LINSignal("Speed", frame_id=0, start_bit=0, bit_length=16))
+            >>> decoded = analyzer.decode_signals(0, b'\\x10\\x27')
+            >>> print(f"Speed: {decoded['Speed']}")
+            Speed: 10000
+        """
+        decoded: dict[str, Any] = {}
+
+        # Find signals for this frame
+        frame_signals = [s for s in self.signals if s.frame_id == frame_id]
+
+        # Convert data to bit array
+        if len(data) == 0:
+            return decoded
+
+        data_bits = int.from_bytes(data, "little")
+
+        for signal in frame_signals:
+            # Extract signal bits
+            mask = (1 << signal.bit_length) - 1
+            value = (data_bits >> signal.start_bit) & mask
+            decoded[signal.name] = value
+
+        return decoded
+
+    def infer_schedule_table(self) -> list[LINScheduleEntry]:
+        """Infer schedule table from captured frame timing.
+
+        Analyzes frame timestamps to determine typical transmission schedule.
+
+        Returns:
+            List of schedule entries ordered by typical transmission sequence.
+
+        Example:
+            >>> analyzer = LINAnalyzer()
+            >>> # ... parse frames ...
+            >>> schedule = analyzer.infer_schedule_table()
+            >>> for entry in schedule:
+            ...     print(f"Frame 0x{entry.frame_id:02X} delay {entry.delay_ms:.1f}ms")
+        """
+        if len(self.frames) < 2:
+            return []
+
+        # Group frames by ID and calculate average inter-frame delays
+        frame_delays: dict[int, list[float]] = {}
+
+        for i in range(1, len(self.frames)):
+            prev_frame = self.frames[i - 1]
+            curr_frame = self.frames[i]
+
+            # Calculate delay in milliseconds
+            delay_ms = (curr_frame.timestamp - prev_frame.timestamp) * 1000.0
+
+            # Group by current frame ID
+            if curr_frame.frame_id not in frame_delays:
+                frame_delays[curr_frame.frame_id] = []
+            frame_delays[curr_frame.frame_id].append(delay_ms)
+
+        # Calculate average delay for each frame ID
+        schedule_entries = []
+        for frame_id in sorted(self.detected_frame_ids):
+            if frame_id in frame_delays and len(frame_delays[frame_id]) > 0:
+                avg_delay = sum(frame_delays[frame_id]) / len(frame_delays[frame_id])
+                schedule_entries.append(LINScheduleEntry(frame_id=frame_id, delay_ms=avg_delay))
+
+        self.schedule = schedule_entries
+        return schedule_entries
+
+    def generate_ldf(self, output_path: Path, baudrate: int = 19200) -> None:
+        """Generate LDF (LIN Description File) from captured traffic.
+
+        LDF Format per LIN 2.x specification:
+        - Header (protocol version, language version, speed)
+        - Nodes (master and slaves)
+        - Signals (name, size, init value, publisher)
+        - Frames (ID, publisher, size, signals)
+        - Schedule tables (frame sequence and timing)
+
+        Args:
+            output_path: Path to output LDF file.
+            baudrate: LIN bus baudrate in bps (default 19200).
+
+        Raises:
+            ValueError: If no frames have been captured.
+
+        Example:
+            >>> analyzer = LINAnalyzer()
+            >>> # ... parse frames and add signals ...
+            >>> analyzer.generate_ldf(Path("vehicle.ldf"), baudrate=19200)
+        """
+        if len(self.frames) == 0:
+            raise ValueError("No frames captured - cannot generate LDF")
+
+        ldf_lines: list[str] = []
+        self._add_ldf_header(ldf_lines, baudrate)
+        self._add_ldf_nodes(ldf_lines)
+        self._add_ldf_signals(ldf_lines)
+        self._add_ldf_frames(ldf_lines)
+        self._add_ldf_schedule(ldf_lines)
+
+        output_path.write_text("\n".join(ldf_lines) + "\n", encoding="utf-8")
+
+    def _add_ldf_header(self, ldf_lines: list[str], baudrate: int) -> None:
+        """Add LDF header section."""
+        ldf_lines.append("LIN_description_file;")
+        ldf_lines.append('LIN_protocol_version = "2.1";')
+        ldf_lines.append('LIN_language_version = "2.1";')
+        ldf_lines.append(f"LIN_speed = {baudrate / 1000:.1f} kbps;")
+        ldf_lines.append("")
+
+    def _add_ldf_nodes(self, ldf_lines: list[str]) -> None:
+        """Add LDF nodes section (master and slaves)."""
+        ldf_lines.append("Nodes {")
+        ldf_lines.append("  Master: Master, 5 ms, 0.1 ms;")
+
+        slaves = sorted({s.publisher for s in self.signals if s.publisher != "Master"})
+        if len(slaves) > 0:
+            ldf_lines.append(f"  Slaves: {', '.join(slaves)};")
+        ldf_lines.append("}")
+        ldf_lines.append("")
+
+    def _add_ldf_signals(self, ldf_lines: list[str]) -> None:
+        """Add LDF signals section."""
+        if len(self.signals) == 0:
+            return
+
+        ldf_lines.append("Signals {")
+        for signal in self.signals:
+            ldf_lines.append(
+                f"  {signal.name}: {signal.bit_length}, {signal.init_value}, {signal.publisher};"
+            )
+        ldf_lines.append("}")
+        ldf_lines.append("")
+
+    def _add_ldf_frames(self, ldf_lines: list[str]) -> None:
+        """Add LDF frames section."""
+        ldf_lines.append("Frames {")
+        for frame_id in sorted(self.detected_frame_ids):
+            if frame_id in (self.MASTER_REQUEST_FRAME, self.SLAVE_RESPONSE_FRAME):
+                continue
+
+            frame_signals = [s for s in self.signals if s.frame_id == frame_id]
+            publisher, dlc = self._determine_frame_metadata(frame_id, frame_signals)
+
+            ldf_lines.append(f"  Frame_{frame_id:02X}: {frame_id}, {publisher}, {dlc} {{")
+            if len(frame_signals) > 0:
+                for signal in frame_signals:
+                    ldf_lines.append(f"    {signal.name}, {signal.start_bit};")
+            ldf_lines.append("  }")
+
+        ldf_lines.append("}")
+        ldf_lines.append("")
+
+    def _determine_frame_metadata(
+        self, frame_id: int, frame_signals: list[LINSignal]
+    ) -> tuple[str, int]:
+        """Determine frame publisher and data length."""
+        if len(frame_signals) > 0:
+            publisher = frame_signals[0].publisher
+            max_bit = max((s.start_bit + s.bit_length) for s in frame_signals)
+            dlc = (max_bit + 7) // 8
+        else:
+            frame_data_lengths = [len(f.data) for f in self.frames if f.frame_id == frame_id]
+            dlc = max(frame_data_lengths) if frame_data_lengths else 1
+            publisher = "Master"
+        return publisher, dlc
+
+    def _add_ldf_schedule(self, ldf_lines: list[str]) -> None:
+        """Add LDF schedule tables section."""
+        if len(self.schedule) == 0:
+            self.infer_schedule_table()
+
+        if len(self.schedule) == 0:
+            return
+
+        ldf_lines.append("Schedule_tables {")
+        ldf_lines.append("  NormalTable {")
+        for entry in self.schedule:
+            if entry.frame_id in (self.MASTER_REQUEST_FRAME, self.SLAVE_RESPONSE_FRAME):
+                continue
+            ldf_lines.append(f"    Frame_{entry.frame_id:02X} delay {entry.delay_ms:.1f} ms;")
+        ldf_lines.append("  }")
+        ldf_lines.append("}")

oscura/automotive/loaders/blf.py

@@ -32,7 +32,7 @@ def load_blf(file_path: Path | str) -> CANMessageList:
         >>> print(f"Unique IDs: {len(messages.unique_ids())}")
     """
     try:
-        import can  # type: ignore[import-untyped]
+        import can
     except ImportError as e:
         raise ImportError(
             "python-can is required for BLF file support. "
@@ -56,8 +56,19 @@ def load_blf(file_path: Path | str) -> CANMessageList:
                     ch = msg.channel
                     if isinstance(ch, int):
                         channel = ch
+                    elif isinstance(ch, str):
+                        channel = int(ch)
                     elif ch is not None:
-                        channel = int(ch)  # type: ignore[arg-type]
+                        # Handle other types (e.g., sequences), default to 0 if conversion fails
+                        try:
+                            # Convert sequences to single value
+                            from collections.abc import Sequence as ABCSequence
+
+                            if isinstance(ch, ABCSequence):
+                                channel = int(ch[0]) if ch else 0
+                            # Note: other cases caught by exception handler
+                        except (TypeError, ValueError, IndexError):
+                            channel = 0
 
                 can_msg = CANMessage(
                     arbitration_id=msg.arbitration_id,