import os
import re
from lxml import etree
from pathlib import Path
import dateutil.parser
from datetime import datetime, timedelta
from pystac_client import Client
from pystac_client.stac_api_io import StacApiIO
from spatialist import Vector, crsConvert
import asf_search as asf
from pyroSAR import identify_many, ID
from S1_NRB.ancillary import buffer_time
from S1_NRB.tile_extraction import aoi_from_tile, tile_from_aoi
[docs]
class ASF(ID):
"""
Simple SAR metadata handler for scenes in the ASF archive. The interface is consistent with the driver classes in
:mod:`pyroSAR.drivers` but does not implement the full functionality due to limited content of the CMR
metadata catalog. Registered attributes:
- acquisition_mode
- coordinates
- frameNumber
- orbit
- orbitNumber_abs
- orbitNumber_rel
- polarizations
- product
- projection
- sensor
- start
- stop
"""
def __init__(self, meta):
self.scene = meta['properties']['url']
self._meta = meta
self.meta = self.scanMetadata()
super(ASF, self).__init__(self.meta)
[docs]
class STACArchive(object):
"""
Search for scenes in a SpatioTemporal Asset Catalog.
Scenes are expected to be unpacked with a folder suffix .SAFE.
The interface is kept consistent with :func:`~S1_NRB.search.ASFArchive`
and :class:`pyroSAR.drivers.Archive`.
Parameters
----------
url: str
the catalog URL
collections: str or list[str]
the catalog collection(s) to be searched
timeout: int
the allowed timeout in seconds
max_retries: int or None
the number of times to retry requests. Set to None to disable retries.
See Also
--------
pystac_client.Client.open
pystac_client.stac_api_io.StacApiIO
"""
def __init__(self, url, collections, timeout=60, max_retries=20):
self.url = url
self.timeout = timeout
self.max_tries = max_retries
self._open_catalog()
if isinstance(collections, str):
self.collections = [collections]
elif isinstance(collections, list):
self.collections = collections
else:
raise TypeError("'collections' must be of type str or list")
def __enter__(self):
return self
def __exit__(self, exc_type, exc_val, exc_tb):
self.close()
@staticmethod
def _get_proc_time(scene):
with open(os.path.join(scene, 'manifest.safe'), 'rb') as f:
tree = etree.fromstring(f.read())
proc = tree.find(path='.//xmlData/safe:processing',
namespaces=tree.nsmap)
start = proc.attrib['start']
del tree, proc
return datetime.strptime(start, '%Y-%m-%dT%H:%M:%S.%f')
def _filter_duplicates(self, scenes):
tmp = sorted(scenes)
pattern = '([0-9A-Z_]{16})_([0-9T]{15})_([0-9T]{15})'
keep = []
i = 0
while i < len(tmp):
group = [tmp[i]]
match1 = re.search(pattern, os.path.basename(tmp[i])).groups()
j = i + 1
while j < len(tmp):
match2 = re.search(pattern, os.path.basename(tmp[j])).groups()
if match1 == match2:
group.append(tmp[j])
j += 1
else:
break
if len(group) > 1:
tproc = [self._get_proc_time(x) for x in group]
keep.append(group[tproc.index(max(tproc))])
else:
keep.append(group[0])
i = j
return keep
def _open_catalog(self):
stac_api_io = StacApiIO(max_retries=self.max_tries)
self.catalog = Client.open(url=self.url,
stac_io=stac_api_io,
timeout=self.timeout)
[docs]
def close(self):
del self.catalog
[docs]
def select(self, sensor=None, product=None, acquisition_mode=None,
mindate=None, maxdate=None, frameNumber=None,
vectorobject=None, date_strict=True, check_exist=True):
"""
Select scenes from the catalog. Used STAC keys:
- platform
- start_datetime
- end_datetime
- sar:instrument_mode
- sar:product_type
- s1:datatake (custom)
Parameters
----------
sensor: str or list[str] or None
S1A or S1B
product: str or list[str] or None
GRD or SLC
acquisition_mode: str or list[str] or None
IW, EW or SM
mindate: str or datetime.datetime or None
the minimum acquisition date
maxdate: str or datetime.datetime or None
the maximum acquisition date
frameNumber: int or list[int] or None
the data take ID in decimal representation.
Requires custom STAC key `s1:datatake`.
vectorobject: spatialist.vector.Vector or None
a geometry with which the scenes need to overlap
date_strict: bool
treat dates as strict limits or also allow flexible limits to incorporate scenes
whose acquisition period overlaps with the defined limit?
- strict: start >= mindate & stop <= maxdate
- not strict: stop >= mindate & start <= maxdate
check_exist: bool
check whether found files exist locally?
Returns
-------
list[str]
the locations of the scene directories with suffix .SAFE
See Also
--------
pystac_client.Client.search
"""
pars = locals()
del pars['date_strict']
del pars['check_exist']
del pars['self']
lookup = {'product': 'sar:product_type',
'acquisition_mode': 'sar:instrument_mode',
'mindate': 'start_datetime',
'maxdate': 'end_datetime',
'sensor': 'platform',
'frameNumber': 's1:datatake'}
lookup_platform = {'S1A': 'sentinel-1a',
'S1B': 'sentinel-1b'}
args = {'datetime': [None, None]}
flt = {'op': 'and', 'args': []}
dt_pattern = '%Y-%m-%dT%H:%M:%SZ'
for key in pars.keys():
val = pars[key]
if val is None:
continue
if key in ['mindate', 'maxdate']:
if isinstance(val, str):
val = dateutil.parser.parse(val)
if isinstance(val, datetime):
val = datetime.strftime(val, dt_pattern)
if key == 'mindate':
args['datetime'][0] = val
if date_strict:
arg = {'op': '>=', 'args': [{'property': 'start_datetime'}, val]}
else:
arg = {'op': '>=', 'args': [{'property': 'end_datetime'}, val]}
flt['args'].append(arg)
elif key == 'maxdate':
args['datetime'][1] = val
if date_strict:
arg = {'op': '<=', 'args': [{'property': 'end_datetime'}, val]}
else:
arg = {'op': '<=', 'args': [{'property': 'start_datetime'}, val]}
flt['args'].append(arg)
elif key == 'vectorobject':
if isinstance(val, Vector):
with val.clone() as vec:
vec.reproject(4326)
ext = vec.extent
args['bbox'] = [ext['xmin'], ext['ymin'], ext['xmax'], ext['ymax']]
else:
raise TypeError('argument vectorobject must be of type spatialist.vector.Vector')
else:
args2 = []
if isinstance(val, (str, int)):
val = [val]
for v in val:
if key == 'sensor':
value = lookup_platform[v]
elif key == 'frameNumber':
value = '{:06X}'.format(v) # convert to hexadecimal
else:
value = v
a = {'op': '=', 'args': [{'property': lookup[key]}, value]}
args2.append(a)
if len(args2) == 1:
arg = args2[0]
else:
arg = {'op': 'or', 'args': args2}
flt['args'].append(arg)
if len(flt['args']) == 0:
flt = None
if args['datetime'] == [None, None]:
args['datetime'] = None
result = self.catalog.search(collections=self.collections,
filter=flt, max_items=None,
**args)
result = list(result.items())
out = []
for item in result:
assets = item.assets
ref = assets[list(assets.keys())[0]]
href = ref.href
path = href[:re.search(r'\.SAFE', href).end()]
path = re.sub('^file://', '', path)
if Path(path).exists():
path = os.path.realpath(path)
else:
if check_exist:
raise RuntimeError('scene does not exist locally:', path)
out.append(path)
out = self._filter_duplicates(out)
return out
[docs]
class ASFArchive(object):
"""
Search for scenes in the Alaska Satellite Facility (ASF) catalog.
"""
def __enter__(self):
return self
def __exit__(self, exc_type, exc_val, exc_tb):
return
[docs]
@staticmethod
def select(sensor=None, product=None, acquisition_mode=None, mindate=None,
maxdate=None, vectorobject=None, date_strict=True, return_value='url'):
"""
Select scenes from the ASF catalog. This is a simple wrapper around the function
:func:`~S1_NRB.search.asf_select` to be consistent with the interfaces of
:func:`~S1_NRB.search.STACArchive` and :class:`pyroSAR.drivers.Archive`.
Parameters
----------
sensor: str or list[str] or None
S1A or S1B
product: str or list[str] or None
GRD or SLC
acquisition_mode: str or list[str] or None
IW, EW or SM
mindate: str or datetime.datetime or None
the minimum acquisition date
maxdate: str or datetime.datetime or None
the maximum acquisition date
vectorobject: spatialist.vector.Vector or None
a geometry with which the scenes need to overlap
date_strict: bool
treat dates as strict limits or also allow flexible limits to incorporate scenes
whose acquisition period overlaps with the defined limit?
- strict: start >= mindate & stop <= maxdate
- not strict: stop >= mindate & start <= maxdate
return_value: str or list[str]
the metadata return value; see :func:`~S1_NRB.search.asf_select` for details
See Also
--------
asf_select
Returns
-------
list[str or tuple[str] or ASF]
the scene metadata attributes as specified with `return_value`;
see :func:`~S1_NRB.search.asf_select` for details
"""
return asf_select(sensor, product, acquisition_mode, mindate, maxdate, vectorobject,
return_value=return_value, date_strict=date_strict)
[docs]
def asf_select(sensor, product, acquisition_mode, mindate, maxdate,
vectorobject=None, return_value='url', date_strict=True):
"""
Search scenes in the Alaska Satellite Facility (ASF) data catalog. This is a simple interface to the
`asf_search <https://github.com/asfadmin/Discovery-asf_search>`_ package.
Parameters
----------
sensor: str
S1A or S1B
product: str
GRD or SLC
acquisition_mode: str
IW, EW or SM
mindate: str or datetime.datetime
the minimum acquisition date
maxdate: str or datetime.datetime
the maximum acquisition date
vectorobject: spatialist.vector.Vector or None
a geometry with which the scenes need to overlap
return_value: str or list[str]
the metadata return value; if `ASF`, an :class:`~S1_NRB.search.ASF` object is returned;
further string options specify certain properties to return: `beamModeType`, `browse`,
`bytes`, `centerLat`, `centerLon`, `faradayRotation`, `fileID`, `flightDirection`, `groupID`,
`granuleType`, `insarStackId`, `md5sum`, `offNadirAngle`, `orbit`, `pathNumber`, `platform`,
`pointingAngle`, `polarization`, `processingDate`, `processingLevel`, `sceneName`, `sensor`,
`startTime`, `stopTime`, `url`, `pgeVersion`, `fileName`, `frameNumber`; all options except
`ASF` can also be combined in a list
date_strict: bool
treat dates as strict limits or also allow flexible limits to incorporate scenes
whose acquisition period overlaps with the defined limit?
- strict: start >= mindate & stop <= maxdate
- not strict: stop >= mindate & start <= maxdate
Returns
-------
list[str or tuple[str] or ASF]
the scene metadata attributes as specified with `return_value`; the return type is a list of strings,
tuples or :class:`~S1_NRB.search.ASF` objects depending on whether `return_type` is of type string, list or :class:`~S1_NRB.search.ASF`.
"""
if isinstance(return_value, list) and 'ASF' in return_value:
raise RuntimeError("'ASF' may not be a list element of 'return_value'")
if product == 'GRD':
processing_level = ['GRD_HD', 'GRD_MD', 'GRD_MS', 'GRD_HS', 'GRD_FD']
else:
processing_level = product
if acquisition_mode == 'SM':
beam_mode = ['S1', 'S2', 'S3', 'S4', 'S5', 'S6']
else:
beam_mode = acquisition_mode
if vectorobject is not None:
with vectorobject.clone() as geom:
geom.reproject(4326)
geometry = geom.convert2wkt(set3D=False)[0]
else:
geometry = None
if isinstance(mindate, str):
start = dateutil.parser.parse(mindate)
else:
start = mindate
if isinstance(maxdate, str):
stop = dateutil.parser.parse(maxdate)
else:
stop = maxdate
result = asf.search(platform=sensor.replace('S1', 'Sentinel-1'),
processingLevel=processing_level,
beamMode=beam_mode,
start=start,
end=stop,
intersectsWith=geometry).geojson()
features = result['features']
def date_extract(item, key):
value = item['properties'][key]
out = datetime.strptime(value, '%Y-%m-%dT%H:%M:%S.%fZ')
return out
if date_strict:
features = [x for x in features
if start <= date_extract(x, 'startTime')
and date_extract(x, 'stopTime') <= stop]
if return_value == 'ASF':
return [ASF(x) for x in features]
out = []
for item in features:
properties = item['properties']
if isinstance(return_value, str):
out.append(properties[return_value])
elif isinstance(return_value, list):
out.append(tuple([properties[x] for x in return_value]))
else:
raise TypeError(f'invalid type of return value: {type(return_value)}')
return sorted(out)
[docs]
def scene_select(archive, kml_file, aoi_tiles=None, aoi_geometry=None, **kwargs):
"""
Central scene search utility. Selects scenes from a database and returns their file names
together with the MGRS tile names for which to process ARD products.
The list of MGRS tile names is either identical to the list provided with `aoi_tiles`,
the list of all tiles overlapping with `aoi_geometry` or `vectorobject` (via `kwargs`),
or the list of all tiles overlapping with an initial scene search result if no geometry
has been defined via `aoi_tiles` or `aoi_geometry`. In the latter (most complex) case,
the search procedure is as follows:
- perform a first search matching all other search parameters
- derive all MGRS tile geometries overlapping with the selection
- derive the minimum and maximum acquisition times of the selection as search parameters
`mindate` and `maxdate`
- extend the `mindate` and `maxdate` search parameters by one minute
- perform a second search with the extended acquisition date parameters and the
derived MGRS tile geometries
As consequence, if one defines the search parameters to only return one scene, the neighboring
acquisitions will also be returned. This is because the scene overlaps with a set of MGRS
tiles of which many or all will also overlap with these neighboring acquisitions. To ensure
full coverage of all MGRS tiles, the neighbors of the scene in focus have to be processed too.
This function has three ways to define search geometries. In order of priority overriding others:
`aoi_tiles` > `aoi_geometry` > `vectorobject` (via `kwargs`). In the latter two cases, the search
geometry is extended to the bounding box of all MGRS tiles overlapping with the initial geometry
to ensure full coverage of all tiles.
Parameters
----------
archive: pyroSAR.drivers.Archive or STACArchive or ASFArchive
an open scene archive connection
kml_file: str
the KML file containing the MGRS tile geometries.
aoi_tiles: list[str] or None
a list of MGRS tile names for spatial search
aoi_geometry: str or None
the name of a vector geometry file for spatial search
kwargs
further search arguments passed to :meth:`pyroSAR.drivers.Archive.select`
or :meth:`STACArchive.select` or :meth:`ASFArchive.select`
Returns
-------
tuple[list[str], list[str]]
- the list of scenes
- the list of MGRS tiles
"""
args = kwargs.copy()
for key in ['acquisition_mode']:
if key not in args.keys():
args[key] = None
if args['acquisition_mode'] == 'SM':
args['acquisition_mode'] = ('S1', 'S2', 'S3', 'S4', 'S5', 'S6')
if isinstance(archive, ASFArchive):
args['return_value'] = 'ASF'
vec = None
selection = []
if aoi_tiles is not None:
vec = aoi_from_tile(kml=kml_file, tile=aoi_tiles)
elif aoi_geometry is not None:
with Vector(aoi_geometry) as geom:
vec = tile_from_aoi(vector=geom, kml=kml_file,
return_geometries=True)
elif 'vectorobject' in args.keys() and args['vectorobject'] is not None:
vec = tile_from_aoi(vector=args['vectorobject'], kml=kml_file,
return_geometries=True)
if vec is not None:
if not isinstance(vec, list):
vec = [vec]
if aoi_tiles is None:
aoi_tiles = [x.mgrs for x in vec]
# derive geometries and tiles from scene footprints
if vec is None:
selection_tmp = archive.select(**args)
if len(selection_tmp) == 0:
return [], []
if not isinstance(selection_tmp[0], ID):
scenes = identify_many(scenes=selection_tmp, sortkey='start')
else:
scenes = selection_tmp
scenes_geom = [x.geometry() for x in scenes]
# select all tiles overlapping with the scenes for further processing
vec = tile_from_aoi(vector=scenes_geom, kml=kml_file,
return_geometries=True)
if not isinstance(vec, list):
vec = [vec]
aoi_tiles = [x.mgrs for x in vec]
del scenes_geom
args['mindate'] = min([datetime.strptime(x.start, '%Y%m%dT%H%M%S') for x in scenes])
args['maxdate'] = max([datetime.strptime(x.stop, '%Y%m%dT%H%M%S') for x in scenes])
del scenes
else:
if isinstance(args['mindate'], str):
args['mindate'] = dateutil.parser.parse(args['mindate'])
if isinstance(args['maxdate'], str):
args['maxdate'] = dateutil.parser.parse(args['maxdate'])
# extend the time range to fully cover all tiles
# (one additional scene needed before and after each data take group)
args['mindate'] -= timedelta(minutes=1)
args['maxdate'] += timedelta(minutes=1)
if isinstance(archive, ASFArchive):
args['return_value'] = 'url'
for item in vec:
args['vectorobject'] = item
selection.extend(archive.select(**args))
del vec, args
return sorted(list(set(selection))), aoi_tiles
[docs]
def collect_neighbors(archive, scene):
"""
Collect a scene's neighboring acquisitions in a data take
Parameters
----------
archive: pyroSAR.drivers.Archive or STACArchive or ASFArchive
an open scene archive connection
scene: pyroSAR.drivers.ID
the Sentinel-1 scene to be checked
Returns
-------
list[str]
the file names of the neighboring scenes
"""
start, stop = buffer_time(scene.start, scene.stop, seconds=2)
neighbors = archive.select(mindate=start, maxdate=stop, date_strict=False,
sensor=scene.sensor, product=scene.product,
acquisition_mode=scene.acquisition_mode)
del neighbors[neighbors.index(scene.scene)]
return neighbors
[docs]
def check_acquisition_completeness(archive, scenes):
"""
Check presence of neighboring acquisitions.
Check that for each scene a predecessor and successor can be queried
from the database unless the scene is at the start or end of the data take.
This ensures that no scene that could be covering an area of interest is missed
during processing. In case a scene is suspected to be missing, the Alaska Satellite Facility (ASF)
online catalog is cross-checked.
An error will only be raised if the locally missing scene is present in the ASF catalog.
Parameters
----------
archive: pyroSAR.drivers.Archive or STACArchive
an open scene archive connection
scenes: list[pyroSAR.drivers.ID]
a list of scenes
Returns
-------
Raises
------
RuntimeError
See Also
--------
S1_NRB.search.asf_select
"""
messages = []
for scene in scenes:
slice = scene.meta['sliceNumber']
n_slices = scene.meta['totalSlices']
groupsize = 3
has_successor = True
has_predecessor = True
start, stop = buffer_time(scene.start, scene.stop, seconds=2)
ref = None
if slice == 0 or n_slices == 0:
# NRT slicing mode
ref = asf_select(sensor=scene.sensor,
product=scene.product,
acquisition_mode=scene.acquisition_mode,
mindate=start,
maxdate=stop,
return_value='sceneName')
match = [re.search(scene.pattern, x + '.SAFE').groupdict() for x in ref]
ref_start_min = min([x['start'] for x in match])
ref_stop_max = max([x['stop'] for x in match])
if ref_start_min == scene.start:
groupsize -= 1
has_predecessor = False
if ref_stop_max == scene.stop:
groupsize -= 1
has_successor = False
else:
if slice == 1: # first slice in the data take
groupsize -= 1
has_predecessor = False
if slice == n_slices: # last slice in the data take
groupsize -= 1
has_successor = False
# Do another database selection to get the scene in question as well as its potential
# predecessor and successor by adding an acquisition time buffer of two seconds.
group = archive.select(sensor=scene.sensor,
product=scene.product,
acquisition_mode=scene.acquisition_mode,
mindate=start,
maxdate=stop,
date_strict=False)
group = identify_many(group)
# if the number of selected scenes is lower than the expected group size,
# check whether the predecessor, the successor or both are missing by
# cross-checking with the ASF database.
if len(group) < groupsize:
if ref is None:
ref = asf_select(sensor=scene.sensor,
product=scene.product,
acquisition_mode=scene.acquisition_mode,
mindate=start,
maxdate=stop,
return_value='sceneName')
match = [re.search(scene.pattern, x + '.SAFE').groupdict() for x in ref]
ref_start_min = min([x['start'] for x in match])
ref_stop_max = max([x['stop'] for x in match])
start_min = min([x.start for x in group])
stop_max = max([x.stop for x in group])
missing = []
if ref_start_min < start < start_min and has_predecessor:
missing.append('predecessor')
if stop_max < stop < ref_stop_max and has_successor:
missing.append('successor')
if len(missing) > 0:
base = os.path.basename(scene.scene)
messages.append(f'{" and ".join(missing)} acquisition for scene {base}')
if len(messages) != 0:
text = '\n - '.join(messages)
raise RuntimeError(f'missing the following scenes:\n - {text}')