import tables, re, os
import numpy as np
from datetime import datetime
import warnings
from tables.exceptions import DataTypeWarning
warnings.filterwarnings('ignore', category=DataTypeWarning)
import polsartools
def identify_nisar_product(file_path):
with tables.open_file(file_path, mode="r") as h5file:
band = "Unknown"
if "/science/LSAR" in h5file:
band = "LSAR"
elif "/science/SSAR" in h5file:
band = "SSAR"
product_type = None
supported_products = ["RSLC", "GSLC", "GCOV"]
if band != "Unknown":
for prod in supported_products:
path = f"/science/{band}/{prod}"
if path in h5file:
product_type = prod
break
if not product_type:
product_type = "Unknown"
return band, product_type
def print_rslc_meta(file_path, band, product_type):
"""Prints a summary of metadata for a given NISAR HDF5 product file.
This function checks the given NASA-ISRO NISAR product file,
identifies its radar band (L-band or S-band) and product type (RSLC, GSLC,
or GCOV), and outputs a detailed text summary.
The printed report includes:
- Product identification (Band, Product Type, Polarizations)
- Acquisition bandwidth and orbit pass direction
- Acquisition start/end timestamps and calculated duration
- Spatial grid and pixel spacing parameters (along-track, slant/ground range,
or UTM coordinate spacings)
- Bounding polygon geographic corners (NW, NE, SE, SW)
- Near, mid, and far-range incidence angle
Args:
`product_file` (str or pathlib.Path): Path to the NISAR HDF5 data product file.
Raises:
FileNotFoundError: If the specified `product_file` does not exist.
KeyError: If expected HDF5 internal paths or datasets are missing.
ValueError: If the product type or band cannot be correctly identified
or is unsupported.
Example:
>>> nisar_product_info("path/to/NISAR_PRODUCT.h5")
"""
if band == "LSAR":
band_prefix = "L"
elif band == "SSAR":
band_prefix = "S"
else:
print("Unknown band. Cannot proceed.")
return
with tables.open_file(file_path, mode="r") as h5file:
# 1. Read spacing values
along_track = h5file.get_node(f"/science/{band_prefix}SAR/RSLC/swaths/frequencyA/sceneCenterAlongTrackSpacing").read()
ground_range = h5file.get_node(f"/science/{band_prefix}SAR/RSLC/swaths/frequencyA/sceneCenterGroundRangeSpacing").read()
incidence_angle = h5file.get_node(f"/science/{band_prefix}SAR/RSLC/metadata/geolocationGrid/incidenceAngle").read()
orbit_pass_direction = h5file.get_node(f"/science/{band_prefix}SAR/identification/orbitPassDirection").read()
slant_range = h5file.get_node(f"/science/{band_prefix}SAR/RSLC/swaths/frequencyA/slantRangeSpacing").read()
zero_doppler_end_time = h5file.get_node(f"/science/{band_prefix}SAR/identification/zeroDopplerEndTime").read()
zero_doppler_start_time = h5file.get_node(f"/science/{band_prefix}SAR/identification/zeroDopplerStartTime").read()
listOfPolarizations = h5file.get_node(f"/science/{band_prefix}SAR/{product_type}/swaths/frequencyA/listOfPolarizations").read()
listOfPolarizations = listOfPolarizations.astype(str).tolist()
acquiredRangeBandwidth = h5file.get_node(f"/science/{band_prefix}SAR/{product_type}/swaths/frequencyA/acquiredRangeBandwidth").read()
num_cols = incidence_angle.shape[1]
# Calculate the mean of the first, middle, and last columns
min_angle = np.mean(incidence_angle[:, 0])
center_angle = np.mean(incidence_angle[:, num_cols // 2])
max_angle = np.mean(incidence_angle[:, -1])
# 2. Read and decode the bounding polygon
polygon_node = h5file.get_node(f"/science/{band_prefix}SAR/identification/boundingPolygon")
polygon_val = polygon_node.read()
polygon_str = polygon_val.decode('utf-8') if isinstance(polygon_val, bytes) else str(polygon_val)
# Extract coordinate triplets (Longitude, Latitude, Height)
coords = re.findall(r"([-+]?\d*\.\d+|\d+)\s+([-+]?\d*\.\d+|\d+)\s+([-+]?\d*\.\d+|\d+)", polygon_str)
lons = [float(c[0]) for c in coords]
lats = [float(c[1]) for c in coords]
# 3. Derive the 4 bounding rectangle corners (Longitude, Latitude)
min_lon, max_lon = min(lons), max(lons)
min_lat, max_lat = min(lats), max(lats)
corners = {
"Top-Left (NW)": (min_lon, max_lat),
"Top-Right (NE)": (max_lon, max_lat),
"Bottom-Right (SE)": (max_lon, min_lat),
"Bottom-Left (SW)": (min_lon, min_lat)
}
val = orbit_pass_direction.item() if hasattr(orbit_pass_direction, 'item') else orbit_pass_direction
if isinstance(val, bytes):
val = val.decode('utf-8')
time_bytes = zero_doppler_end_time.item()
time_str = time_bytes.decode('utf-8')
dt_obj = datetime.fromisoformat(time_str[:26])
start_bytes = zero_doppler_start_time.item()
end_bytes = zero_doppler_end_time.item()
start_str = start_bytes.decode('utf-8')
end_str = end_bytes.decode('utf-8')
# 2. Parse into datetime objects (slicing to [:26] avoids nanosecond overflow errors)
start_dt = datetime.fromisoformat(start_str[:26])
end_dt = datetime.fromisoformat(end_str[:26])
# 3. Calculate the time difference (duration)
duration = end_dt - start_dt
print(f"{band.split('SAR')[0]}-band {product_type} with polarizations: {', '.join(listOfPolarizations)}")
print(f"Acquired Range Bandwidth: {acquiredRangeBandwidth/1e6:.2f} MHz")
print(f"Orbit Pass Direction: {val}")
print("Acquisition Date/Time (UTC):", dt_obj.strftime("%Y-%m-%d %H:%M:%S"))
print(f"Acquisition Duration: {duration.total_seconds()} seconds")
print("-" * 40)
print(f"Along-Track Spacing: {along_track:3.3f} m")
print(f"Ground Range Spacing: {ground_range:3.3f} m")
print(f"Slant Range Spacing: {slant_range:3.3f} m")
print("-" * 40)
print("Coverage Coordinates (Longitude, Latitude):")
for name, (lon, lat) in corners.items():
print(f" {name:<17}: {lon:.6f}, {lat:.6f}")
print("-" * 40)
print(f"Near-range Incidence Angle: {min_angle:.6f}°")
print(f"mid-range Incidence Angle: {center_angle:.6f}°")
print(f"far-range Incidence Angle: {max_angle:.6f}°")
def print_gslc_meta(file_path, band, product_type):
if band == "LSAR":
band_prefix = "L"
elif band == "SSAR":
band_prefix = "S"
else:
print("Unknown band. Cannot proceed.")
return
with tables.open_file(file_path, mode="r") as h5file:
# 1. Read spacing values
if band == "LSAR":
along_track = h5file.get_node(f"/science/{band_prefix}SAR/{product_type}/metadata/sourceData/swaths/frequencyA/sceneCenterAlongTrackSpacing").read()
if product_type == "GSLC":
slant_range = h5file.get_node(f"/science/{band_prefix}SAR/{product_type}/grids/frequencyA/slantRangeSpacing").read()
elif product_type == "GCOV":
slant_range = h5file.get_node(f"/science/{band_prefix}SAR/{product_type}/metadata/sourceData/swaths/frequencyA/slantRangeSpacing").read()
# ground_range = h5file.get_node(f"/science/{band_prefix}SAR/{product_type}/metadata/sourceData/swaths/frequencyA/sceneCenterGroundRangeSpacing").read()
x_coordinate_spacing = h5file.get_node(f"/science/{band_prefix}SAR/{product_type}/grids/frequencyA/xCoordinateSpacing").read()
y_coordinate_spacing = h5file.get_node(f"/science/{band_prefix}SAR/{product_type}/grids/frequencyA/yCoordinateSpacing").read()
listOfPolarizations = h5file.get_node(f"/science/{band_prefix}SAR/{product_type}/grids/frequencyA/listOfPolarizations").read()
listOfPolarizations = listOfPolarizations.astype(str).tolist()
incidence_angle = h5file.get_node(f"/science/{band_prefix}SAR/{product_type}/metadata/radarGrid/incidenceAngle").read()
orbit_pass_direction = h5file.get_node(f"/science/{band_prefix}SAR/identification/orbitPassDirection").read()
if band == "LSAR":
acquiredRangeBandwidth = h5file.get_node(f"/science/{band_prefix}SAR/{product_type}/metadata/sourceData/swaths/frequencyA/acquiredRangeBandwidth").read()
elif band == "SSAR" and product_type == "GCOV":
acquiredRangeBandwidth = h5file.get_node(f"/science/{band_prefix}SAR/{product_type}/metadata/sourceData/swaths/frequencyA/rangeBandwidth").read()
elif band == "SSAR" and product_type == "GSLC":
acquiredRangeBandwidth = h5file.get_node(f"/science/{band_prefix}SAR/{product_type}/grids/frequencyA/rangeBandwidth").read()
zero_doppler_end_time = h5file.get_node(f"/science/{band_prefix}SAR/identification/zeroDopplerEndTime").read()
zero_doppler_start_time = h5file.get_node(f"/science/{band_prefix}SAR/identification/zeroDopplerStartTime").read()
num_cols = incidence_angle.shape[1]
# Calculate the mean of the first, middle, and last columns
min_angle = np.mean(incidence_angle[:, 0])
center_angle = np.mean(incidence_angle[:, num_cols // 2])
max_angle = np.mean(incidence_angle[:, -1])
# 2. Read and decode the bounding polygon
polygon_node = h5file.get_node(f"/science/{band_prefix}SAR/identification/boundingPolygon")
polygon_val = polygon_node.read()
polygon_str = polygon_val.decode('utf-8') if isinstance(polygon_val, bytes) else str(polygon_val)
# Extract coordinate triplets (Longitude, Latitude, Height)
coords = re.findall(r"([-+]?\d*\.\d+|\d+)\s+([-+]?\d*\.\d+|\d+)\s+([-+]?\d*\.\d+|\d+)", polygon_str)
lons = [float(c[0]) for c in coords]
lats = [float(c[1]) for c in coords]
# 3. Derive the 4 bounding rectangle corners (Longitude, Latitude)
min_lon, max_lon = min(lons), max(lons)
min_lat, max_lat = min(lats), max(lats)
corners = {
"Top-Left (NW)": (min_lon, max_lat),
"Top-Right (NE)": (max_lon, max_lat),
"Bottom-Right (SE)": (max_lon, min_lat),
"Bottom-Left (SW)": (min_lon, min_lat)
}
val = orbit_pass_direction.item() if hasattr(orbit_pass_direction, 'item') else orbit_pass_direction
if isinstance(val, bytes):
val = val.decode('utf-8')
time_bytes = zero_doppler_end_time.item()
time_str = time_bytes.decode('utf-8')
dt_obj = datetime.fromisoformat(time_str[:26])
start_bytes = zero_doppler_start_time.item()
end_bytes = zero_doppler_end_time.item()
start_str = start_bytes.decode('utf-8')
end_str = end_bytes.decode('utf-8')
# 2. Parse into datetime objects (slicing to [:26] avoids nanosecond overflow errors)
start_dt = datetime.fromisoformat(start_str[:26])
end_dt = datetime.fromisoformat(end_str[:26])
# 3. Calculate the time difference (duration)
duration = end_dt - start_dt
print(f"{band.split('SAR')[0]}-band {product_type} with polarizations: {', '.join(listOfPolarizations)}")
print(f"Acquired Range Bandwidth: {acquiredRangeBandwidth/1e6:.2f} MHz")
print(f"Orbit Pass Direction: {val}")
print("Acquisition Date/Time (UTC):", dt_obj.strftime("%Y-%m-%d %H:%M:%S"))
print(f"Acquisition Duration: {duration.total_seconds()} seconds")
print("-" * 40)
if band == "LSAR":
print(f"Along-Track Spacing: {along_track:3.3f} m")
print(f"Slant Range Spacing: {slant_range:3.3f} m")
print(f"X Coordinate Spacing (ground): {x_coordinate_spacing:3.1f} m")
print(f"Y Coordinate Spacing (ground): {abs(y_coordinate_spacing):3.1f} m")
print("-" * 40)
print("Coverage Coordinates (Longitude, Latitude):")
for name, (lon, lat) in corners.items():
print(f" {name:<17}: {lon:.6f}, {lat:.6f}")
print("-" * 40)
print(f"Near-range Incidence Angle: {min_angle:.6f}°")
print(f"mid-range Incidence Angle: {center_angle:.6f}°")
print(f"far-range Incidence Angle: {max_angle:.6f}°")
[docs]
def nisar_product_info(product_file):
band, product_type = identify_nisar_product(product_file)
print(os.path.basename(product_file).split('.h5')[0])
print('FrequencyA metadata')
print("-" * 40)
if product_type == "RSLC":
print_rslc_meta(product_file, band, product_type)
elif product_type == "GSLC" or product_type == "GCOV":
print_gslc_meta(product_file, band, product_type)
else:
print(f"Unsupported product type: {product_type}")
print("-" * 40)
