fixed analyzer E. added vtk export of api

Author: maureeungaro

analyzer/README.md

@@ -63,6 +63,17 @@ The digitized B2 output includes columns like:
 hitn, pid, tid, E, time, totEdep
 ```
 
+The true-info output includes tracking columns like:
+
+```text
+processName, avgTime, avgx, avgy, avgz, hitn, pid, tid, totalEDeposited
+```
+
+When the matching digitized CSV is available, the analyzer also adds `E` to
+true-info tables by matching rows on event, detector, hit, PID, and track ID.
+In that case `E` is the track total energy, while `totalEDeposited` remains
+the deposited energy.
+
 The ROOT streamer writes one ROOT file per worker thread. For one thread and
 `filename: b2`, the file is typically:
 

analyzer/__pycache__/__init__.cpython-314.pyc

@@ -24,6 +24,12 @@
 	"totalE": "Total Energy (MeV)",
 }
 
+VARIABLE_ALIASES: Mapping[str, tuple[str, ...]] = {
+	"E": ("totalE", "etot"),
+	"totalEDeposited": ("totEdep",),
+	"etot": ("totalE",),
+}
+
 
 def plot_variable(
 	output: GemcOutput | pd.DataFrame,
@@ -73,9 +79,10 @@ def plot_histogram(
 ) -> tuple[plt.Figure, plt.Axes]:
 	"""Plot one numeric variable as a histogram, optionally grouped by a column."""
 
-	if variable not in frame.columns:
-		available = ", ".join(frame.columns)
-		raise KeyError(f"Column '{variable}' not found. Available columns: {available}")
+	variable = _resolve_variable(frame, variable)
+
+	if frame.empty:
+		raise ValueError("Selected data table is empty.")
 
 	values = pd.to_numeric(frame[variable], errors="coerce").dropna()
 	if values.empty:
@@ -124,3 +131,13 @@ def plot_histogram(
 		plt.show()
 
 	return fig, ax
+
+
+def _resolve_variable(frame: pd.DataFrame, variable: str) -> str:
+	if variable not in frame.columns:
+		for alias in VARIABLE_ALIASES.get(variable, ()):
+			if alias in frame.columns:
+				return alias
+		available = ", ".join(frame.columns)
+		raise KeyError(f"Column '{variable}' not found. Available columns: {available}")
+	return variable

analyzer/__pycache__/__main__.cpython-314.pyc

@@ -44,9 +44,10 @@ def read_csv_output(path: str | Path) -> GemcOutput:
 	output = GemcOutput(source=str(path))
 
 	if path.suffix == ".csv":
-		frame = _read_csv(path)
+		frame = _read_csv_with_fallback(path)
 		stream = _classify_csv_stream(path)
 		if stream == "true_info":
+			frame = _with_track_energy(frame, _matching_digitized_csv(path))
 			output.true_info[path.stem] = frame
 		else:
 			output.digitized[path.stem] = frame
@@ -55,7 +56,7 @@ def read_csv_output(path: str | Path) -> GemcOutput:
 	for stream, suffix in CSV_STREAM_SUFFIXES.items():
 		candidate = Path(str(path) + suffix)
 		if candidate.exists():
-			frame = _read_csv(candidate)
+			frame = _read_csv_with_fallback(candidate)
 			if stream == "true_info":
 				output.true_info["csv"] = frame
 			else:
@@ -64,6 +65,7 @@ def read_csv_output(path: str | Path) -> GemcOutput:
 	if not output.true_info and not output.digitized:
 		raise FileNotFoundError(f"No GEMC CSV files found for '{path}'.")
 
+	_add_track_energy_columns(output)
 	return output
 
 
@@ -104,6 +106,61 @@ def _read_csv(path: Path) -> pd.DataFrame:
 	return pd.read_csv(path, sep=",", skipinitialspace=True)
 
 
+def _read_csv_with_fallback(path: Path) -> pd.DataFrame:
+	try:
+		return _read_csv(path)
+	except pd.errors.EmptyDataError:
+		fallback = _thread_csv_fallback(path)
+		if fallback is not None:
+			return _read_csv(fallback)
+		raise
+
+
+def _thread_csv_fallback(path: Path) -> Path | None:
+	"""
+	Return the run-level CSV matching an empty worker-thread CSV, if present.
+
+	Some accumulated digitizations, such as dosimeter output, write their final
+	table to ``name_digitized.csv`` while the corresponding ``name_t0_digitized.csv``
+	worker file exists but is empty.
+	"""
+	name = path.name
+	if "_t0_" not in name:
+		return None
+
+	candidate = path.with_name(name.replace("_t0_", "_", 1))
+	if candidate.exists() and candidate.stat().st_size > 0:
+		return candidate
+	return None
+
+
+def _matching_digitized_csv(path: Path) -> pd.DataFrame | None:
+	candidate = path.with_name(path.name.replace(CSV_STREAM_SUFFIXES["true_info"], CSV_STREAM_SUFFIXES["digitized"]))
+	if candidate == path or not candidate.exists() or candidate.stat().st_size == 0:
+		return None
+	return _read_csv_with_fallback(candidate)
+
+
+def _add_track_energy_columns(output: GemcOutput) -> None:
+	for name, frame in list(output.true_info.items()):
+		digitized = output.digitized.get(name)
+		if digitized is None and len(output.digitized) == 1:
+			digitized = next(iter(output.digitized.values()))
+		output.true_info[name] = _with_track_energy(frame, digitized)
+
+
+def _with_track_energy(frame: pd.DataFrame, digitized: pd.DataFrame | None) -> pd.DataFrame:
+	if "E" in frame.columns or digitized is None or "E" not in digitized.columns:
+		return frame
+
+	keys = [key for key in ("evn", "thread_id", "detector", "hitn", "pid", "tid") if key in frame.columns and key in digitized.columns]
+	if not keys:
+		return frame
+
+	energy = digitized[keys + ["E"]].drop_duplicates(subset=keys)
+	return frame.merge(energy, on=keys, how="left")
+
+
 def _classify_csv_stream(path: Path) -> str:
 	name = path.name
 	if name.endswith(CSV_STREAM_SUFFIXES["true_info"]):

analyzer/__pycache__/cli.cpython-314.pyc

@@ -19,7 +19,7 @@
 
 # python
 import sqlite3
-import os, argparse, sys
+import os, argparse, sys, json, zipfile
 
 from dataclasses import dataclass
 from typing import Optional, Tuple
@@ -78,6 +78,10 @@ def get_arguments(argv=None):
 	parser.add_argument("-pvh", "--height", type=int, default=800, help="Set plotter height")
 	parser.add_argument("-pvx", "--x", type=int, default=0, help="Set plotter x position")
 	parser.add_argument("-pvy", "--y", type=int, default=0, help="Set plotter y position")
+	parser.add_argument("-pvvtk", "--pyvista-vtksz", default=None,
+	                    help="Export PyVista scene as a VTK.js .vtksz file; .vtksz is added if omitted")
+	parser.add_argument("-pvz", "--pyvista-vtksz-zoom", type=float, default=0.25,
+	                    help="Initial VTK.js scene zoom for --pyvista-vtksz; smaller values zoom out")
 	parser.add_argument("-axes", "--add_axes_at_zero", action="store_true",
 	                    help="Add 10cm axes at (0, 0, 0)")
 
@@ -103,6 +107,7 @@ def __init__(
 			args=None,
 			enable_pyvista: Optional[bool] = None,
 			use_background_plotter: bool = None,
+			pyvista_vtksz: Optional[str] = None,
 	):
 		self.args = get_arguments()  # expose args to scripts that use this class
 		self.experiment = experiment
@@ -133,12 +138,21 @@ def __init__(
 		# pyvista
 		# CLI background flag
 		background_flag = bool(getattr(self.args, "pyvista_background", False))
+		self.pyvista_vtksz = pyvista_vtksz if pyvista_vtksz is not None else getattr(
+			self.args, "pyvista_vtksz", None
+		)
+		self.pyvista_vtksz_zoom = getattr(self.args, "pyvista_vtksz_zoom", 0.25)
+		self.show_pyvista_window = (
+			self.args.pyvista
+			or background_flag
+			or (enable_pyvista is True and not self.pyvista_vtksz)
+		)
 
 		# Decide if PyVista is wanted:
 		#   - programmatic enable_pyvista overrides CLI (True/False)
-		#   - otherwise: --pyvista OR --pvb imply pyvista
+		#   - otherwise: --pyvista, --pvb, or --pyvista-vtksz imply pyvista
 		if enable_pyvista is None:
-			wants_pyvista = self.args.pyvista or background_flag
+			wants_pyvista = self.args.pyvista or background_flag or bool(self.pyvista_vtksz)
 		else:
 			wants_pyvista = enable_pyvista
 
@@ -159,6 +173,7 @@ def __init__(
 
 		self._plotter: Optional[object] = None
 		self._camera_initialized = False
+		self._pyvista_vtksz_exported = False
 
 		# Set the initial variation and file names.
 		#
@@ -225,6 +240,92 @@ def close(self):
 			self._plotter.close()
 			self._plotter = None
 
+	def export_vtksz(self, filename: Optional[str] = None, zoom: Optional[float] = None):
+		"""Export the current PyVista scene as a VTK.js OfflineLocalView file."""
+		if not self.use_pyvista:
+			return None
+
+		output = filename if filename is not None else self.pyvista_vtksz
+		if not output:
+			return None
+		if not output.endswith(".vtksz"):
+			output = f"{output}.vtksz"
+
+		p = self.plotter
+		if p is None:
+			return None
+
+		self._configure_camera_from_bounds()
+		try:
+			p.camera.view_angle = 70.0
+		except Exception:
+			pass
+		try:
+			p.render()
+		except Exception:
+			pass
+
+		try:
+			argv = sys.argv
+			sys.argv = [argv[0]]
+			exported = p.export_vtksz(output)
+		except ImportError as e:
+			sys.exit(f"{GColors.RED}Error exporting PyVista VTK.js scene: {e}{GColors.END}")
+		finally:
+			sys.argv = argv
+
+		output_zoom = self.pyvista_vtksz_zoom if zoom is None else zoom
+		self._adjust_vtksz_camera(exported, zoom=output_zoom)
+
+		self._pyvista_vtksz_exported = True
+		print(f"  ❖ Exported PyVista VTK.js scene: {exported}")
+		return exported
+
+	def _adjust_vtksz_camera(self, filename, zoom: float = 0.25):
+		if zoom <= 0:
+			return
+
+		try:
+			with zipfile.ZipFile(filename, "r") as zf:
+				entries = {name: zf.read(name) for name in zf.namelist()}
+		except (OSError, zipfile.BadZipFile):
+			return
+
+		if "index.json" not in entries:
+			return
+
+		try:
+			data = json.loads(entries["index.json"].decode("utf-8"))
+		except (UnicodeDecodeError, json.JSONDecodeError):
+			return
+
+		def adjust_node(node):
+			if not isinstance(node, dict):
+				return
+
+			if "Camera" in str(node.get("type", "")):
+				properties = node.get("properties", {})
+				position = properties.get("position")
+				focal_point = properties.get("focalPoint")
+				if position and focal_point and len(position) == 3 and len(focal_point) == 3:
+					properties["position"] = [
+						focal_point[i] + (position[i] - focal_point[i]) / zoom
+						for i in range(3)
+					]
+				properties["viewAngle"] = 45.0
+				properties["parallelProjection"] = False
+				properties.pop("parallelScale", None)
+
+			for child in node.get("dependencies", []):
+				adjust_node(child)
+
+		adjust_node(data.get("scene"))
+		entries["index.json"] = json.dumps(data).encode("utf-8")
+
+		with zipfile.ZipFile(filename, "w", compression=zipfile.ZIP_DEFLATED) as zf:
+			for name, content in entries.items():
+				zf.writestr(name, content)
+
 
 	def ascii_storage_files(self):
 		"""
@@ -389,6 +490,9 @@ def show(self, block: bool = True):
 				except AttributeError:
 					pass  # BackgroundPlotter may not expose ren_win directly
 
+				if self.pyvista_vtksz and not self._pyvista_vtksz_exported:
+					self.export_vtksz(self.pyvista_vtksz)
+
 				if self.use_background_plotter:
 					# BackgroundPlotter path
 					if block and hasattr(p, "app"):
@@ -408,7 +512,7 @@ def show(self, block: bool = True):
 						)
 
 
-					elif block:
+					elif block and self.show_pyvista_window:
 						p.show()
 
 	def _configure_camera_from_bounds(self, margin: float = 0.8, distance_scale: float = 4.0):
@@ -496,6 +600,7 @@ def autogeometry(
 		auto_show: bool = True,
 		enable_pyvista: Optional[bool] = None,
 		use_background_plotter: Optional[bool] = None,
+		pyvista_vtksz: Optional[str] = None,
 ):
 	# in jupyter: always enable pyvista, never use atexit
 	if _in_jupyter:
@@ -507,6 +612,7 @@ def autogeometry(
 		application,
 		enable_pyvista=enable_pyvista,
 		use_background_plotter=use_background_plotter,
+		pyvista_vtksz=pyvista_vtksz,
 	)
 
 	if auto_show: