Files
UAV-CO2/src/gasflux/reporting.py

136 lines
4.8 KiB
Python

"""This module provides functions for generating mass balance reports."""
from pathlib import Path
import plotly.graph_objects as go
from jinja2 import Template
from plotly.io import to_html
from datetime import datetime, timedelta
import yaml
import logging
from . import plotting
import json
import numpy as np
logger = logging.getLogger(__name__)
def mass_balance_report(
krig_params: dict,
wind_fig: go.Figure,
background_fig: go.Figure,
threed_fig: go.Figure,
krig_fig: go.Figure,
windrose_fig: go.Figure,
) -> str:
"""Generate a mass balance report."""
template_path = Path(__file__).parent / "templates" / "mass_balance_template.html"
# Convert the figures to HTML
plot_htmls = {}
for name, fig in zip(
["3D", "krig", "windrose", "wind", "background"],
[threed_fig, krig_fig, windrose_fig, wind_fig, background_fig],
strict=False,
):
if fig:
plot_htmls[name] = to_html(fig, full_html=False)
else:
plot_htmls[name] = plotting.blank_figure()
summary_data = {
"Estimated flux": f"{krig_params.get('volume', 0):.3f} kgh⁻¹",
}
with Path.open(template_path) as f:
template_content = f.read()
template = Template(template_content)
return template.render(
title="Mass Balance Report",
summary_data=summary_data,
threeD=plot_htmls["3D"],
krig=plot_htmls["krig"],
windrose=plot_htmls["windrose"],
wind=plot_htmls["wind"],
background=plot_htmls["background"],
)
def generate_reports(name: str, processor, config: dict, output_dir: Path):
"""
Generates reports, configuration files, and processed output variables for gasflux processing runs.
Parameters:
name (str): The name identifier for the current processing run (task_id).
processor (object): The processing object containing report data and output variables.
config (dict): Configuration dictionary used for processing.
output_dir (Path): Output directory path (already includes task_id) from INI configuration.
"""
processing_time = datetime.now() + timedelta(hours=8) # Beijing time (UTC+8)
# Save directly to the output directory (already includes task_id)
output_path = output_dir
output_path.mkdir(parents=True, exist_ok=True)
# Save reports
for gas, report in processor.reports.items():
timestamp_str = processing_time.strftime("%Y%m%d_%H%M%S")
report_path = output_path / f"{gas}_report_{timestamp_str}.html"
with open(report_path, "w", encoding="utf-8") as file:
file.write(report)
# Save config
header = f"# Gasflux output config for task {name} from processing run at {processing_time}\n"
timestamp_str = processing_time.strftime("%Y%m%d_%H%M%S")
config_path = output_path / f"config_{timestamp_str}.yaml"
with open(config_path, "w") as file:
file.write(header)
yaml.safe_dump(config, file)
# Save DataFrame to Excel
if hasattr(processor, 'df') and processor.df is not None:
timestamp_str = processing_time.strftime("%Y%m%d_%H%M%S")
excel_path = output_path / f"processed_data_{timestamp_str}.xlsx"
processor.df.to_excel(excel_path, index=False, engine='openpyxl')
logger.info(f"DataFrame saved to {excel_path}")
# Save output variables
output_vars = processor.output_vars
# output_vars = delete_large_arrays(output_vars, threshold_size=50)
header = (
f"# Gasflux output variables for task {name} from processing run at {processing_time}\n"
)
timestamp_str = processing_time.strftime("%Y%m%d_%H%M%S")
filename = output_path / f"output_vars_{timestamp_str}.json"
with open(filename, "w") as file:
file.write(header)
json.dump(
output_vars, file, default=lambda item: item.tolist() if isinstance(item, np.ndarray) else item, indent=4
)
logger.info(f"Task {name} results saved to {output_path}")
def delete_large_arrays(output_vars: dict, threshold_size: int) -> dict:
"""
Iterate through the output_vars dictionary and replace large numpy arrays
with their metadata (e.g., shape and data type).
Parameters:
output_vars (dict): The dictionary containing output data including potential numpy arrays.
threshold_size (int): The number of elements above which an array is considered large.
"""
del_keys = []
for key, value in output_vars.items():
if isinstance(value, dict):
output_vars[key] = delete_large_arrays(value, threshold_size) # recursive
elif isinstance(value, np.ndarray):
if value.size > threshold_size:
del_keys.append(key)
for key in del_keys:
del output_vars[key]
return output_vars