SPRUCE-scraper/spruce/parsers.py

"""
Pure HTML / text parsing functions for the RootView web application.

All functions are side-effect-free: string (or list[str]) in, dict/list out.
No network access, no filesystem access.
"""

import math
import re
from typing import Any
from urllib.parse import unquote

from bs4 import BeautifulSoup


# ---------------------------------------------------------------------------
# Machine descriptor
# ---------------------------------------------------------------------------


def parse_machine_option(label: str, raw_value: str) -> dict[str, Any]:
    """Decode the pipe-delimited <option> value for a machine."""
    decoded = unquote(raw_value)
    parts = decoded.split("|")
    return {
        "label": label,
        "option_value": raw_value,
        "name": parts[0] if len(parts) > 0 else label,
        "ip": parts[1] if len(parts) > 1 else "",
        "port1": parts[2] if len(parts) > 2 else "",
        "machine_id": parts[3] if len(parts) > 3 else "",
        "port2": parts[4] if len(parts) > 4 else "",
        "version": parts[5] if len(parts) > 5 else "",
    }


# ---------------------------------------------------------------------------
# Scan list row
# ---------------------------------------------------------------------------


def parse_scan_row(cells: list[str]) -> dict[str, Any] | None:
    """
    Parse one table row from the scan list into a scan dict.

    Expected columns (from the observed HTML):
      ID, Name, Scan Time, Step Units, (X,Y)-(X,Y)-(DX,DY),
      Dwell Time ms, Scan Lines, Scan Mode, Start Time, End Time,
      Cancelled, User, Scan Status, Archived, [View link]

    Returns None for header rows or rows whose first cell is not a digit.
    """
    if not cells or not cells[0].strip().isdigit():
        return None
    try:
        scan_id = int(cells[0].strip())
        return {
            "scan_id": scan_id,
            "name": cells[1].strip() if len(cells) > 1 else "",
            "scan_time": cells[2].strip() if len(cells) > 2 else "",
            "step_units": cells[3].strip() if len(cells) > 3 else "",
            "coord_str": cells[4].strip() if len(cells) > 4 else "",
            "dwell_ms": cells[5].strip() if len(cells) > 5 else "",
            "scan_lines": cells[6].strip() if len(cells) > 6 else "",
            "scan_mode": cells[7].strip() if len(cells) > 7 else "",
            "start_datetime": cells[8].strip() if len(cells) > 8 else "",
            "end_datetime": cells[9].strip() if len(cells) > 9 else "",
            "cancelled": cells[10].strip() if len(cells) > 10 else "",
            "user": cells[11].strip() if len(cells) > 11 else "",
            "status": cells[12].strip() if len(cells) > 12 else "",
            "archived": cells[13].strip() if len(cells) > 13 else "",
        }
    except (ValueError, IndexError):
        return None


# ---------------------------------------------------------------------------
# Scan view page
# ---------------------------------------------------------------------------


def parse_scan_view(html: str) -> dict[str, Any]:
    """
    Extract grid parameters from a scan view page.

    Returns a dict with keys:
      scan_id, name, scan_time, start_x, start_y, end_x, end_y,
      dx, dy, nx, ny, total_tiles, disk_space_mb, ...
    """
    result: dict[str, Any] = {}

    # Extract grid params from the show_tile() URL inside the page JS.
    # The scan view embeds them as query params in a JS string, e.g.:
    #   "include/tile_view.php?...&sX=0&sY=0&eX=310&eY=740&dX=3.01&dY=2.26&..."
    tile_url_m = re.search(r'tile_view\.php\?([^"\']+)', html)
    if tile_url_m:
        qs = tile_url_m.group(1)
        param_map = {
            "sX": "startX",
            "sY": "startY",
            "eX": "endX",
            "eY": "endY",
            "dX": "deltaX",
            "dY": "deltaY",
        }
        for qs_key, result_key in param_map.items():
            m = re.search(rf"(?:^|&){re.escape(qs_key)}=([\d.]+)", qs)
            if m:
                result[result_key] = float(m.group(1))

    # Fallback: look for standalone JS var declarations (present in tile_view.php)
    js_var_patterns = {
        "startX": r"var\s+startX\s*=\s*([\d.]+)",
        "startY": r"var\s+startY\s*=\s*([\d.]+)",
        "endX": r"var\s+endX\s*=\s*([\d.]+)",
        "endY": r"var\s+endY\s*=\s*([\d.]+)",
        "deltaX": r"var\s+deltaX\s*=\s*([\d.]+)",
        "deltaY": r"var\s+deltaY\s*=\s*([\d.]+)",
    }
    for key, pattern in js_var_patterns.items():
        if key not in result:
            m = re.search(pattern, html)
            if m:
                result[key] = float(m.group(1))

    # Extract from the data table
    soup = BeautifulSoup(html, "lxml")
    for row in soup.find_all("tr"):
        cells = [td.get_text(strip=True) for td in row.find_all("td")]
        for i, cell in enumerate(cells):
            if cell == "Scan ID:" and i + 1 < len(cells):
                try:
                    result["scan_id"] = int(cells[i + 1])
                except ValueError:
                    pass
            elif cell == "Name:" and i + 1 < len(cells):
                result["name"] = cells[i + 1]
            elif cell == "Scan Time:" and i + 1 < len(cells):
                result["scan_time"] = cells[i + 1]
            elif cell == "Starting X:" and i + 1 < len(cells):
                result["start_x_label"] = cells[i + 1]
            elif cell == "Starting Y:" and i + 1 < len(cells):
                result["start_y_label"] = cells[i + 1]
            elif cell == "Ending X:" and i + 1 < len(cells):
                result["end_x_label"] = cells[i + 1]
            elif cell == "Ending Y:" and i + 1 < len(cells):
                result["end_y_label"] = cells[i + 1]
            elif cell == "DX:" and i + 1 < len(cells):
                result["dx_label"] = cells[i + 1]
            elif cell == "DY:" and i + 1 < len(cells):
                result["dy_label"] = cells[i + 1]
            elif cell == "Scan Lines:" and i + 1 < len(cells):
                result["scan_lines"] = cells[i + 1]
            elif cell == "Scan Mode:" and i + 1 < len(cells):
                result["scan_mode"] = cells[i + 1]
            elif cell == "Start Time:" and i + 1 < len(cells):
                result["start_datetime"] = cells[i + 1]
            elif cell == "End Time:" and i + 1 < len(cells):
                result["end_datetime"] = cells[i + 1]
            elif cell == "Scan Status:" and i + 1 < len(cells):
                result["status"] = cells[i + 1]
            elif cell == "User:" and i + 1 < len(cells):
                result["user"] = cells[i + 1]
            elif cell == "Total number of images:" and i + 1 < len(cells):
                # Format: "33784 (103x328)"
                m = re.match(r"(\d+)\s*\((\d+)x(\d+)\)", cells[i + 1])
                if m:
                    result["total_tiles"] = int(m.group(1))
                    result["nx"] = int(m.group(2))
                    result["ny"] = int(m.group(3))
            elif cell == "Total Disk Space:" and i + 1 < len(cells):
                m = re.search(r"([\d.]+)\s*Mb", cells[i + 1])
                if m:
                    result["disk_space_mb"] = float(m.group(1))

    # Promote JS/URL grid param names to canonical keys
    for raw, canon in (
        ("startX", "start_x"),
        ("startY", "start_y"),
        ("endX", "end_x"),
        ("endY", "end_y"),
        ("deltaX", "dx"),
        ("deltaY", "dy"),
    ):
        if raw in result:
            result[canon] = result.pop(raw)

    # Compute nx/ny from grid params if not parsed from table
    if "nx" not in result and all(k in result for k in ("start_x", "end_x", "dx")):
        result["nx"] = _grid_count(result["start_x"], result["end_x"], result["dx"])
    if "ny" not in result and all(k in result for k in ("start_y", "end_y", "dy")):
        result["ny"] = _grid_count(result["start_y"], result["end_y"], result["dy"])
    if "total_tiles" not in result and "nx" in result and "ny" in result:
        result["total_tiles"] = result["nx"] * result["ny"]

    return result


# ---------------------------------------------------------------------------
# Grid helpers
# ---------------------------------------------------------------------------


def _grid_count(start: float, end: float, step: float) -> int:
    """Number of grid positions from start up to (but not including) end."""
    if step <= 0:
        return 0
    return math.ceil((end - start) / step)


def _grid_values(start: float, count: int, step: float) -> list[float]:
    """Generate `count` evenly-spaced grid positions, rounded to 2 dp."""
    return [round(start + i * step, 2) for i in range(count)]