import xsar 
import glob

get product info from a list of safes path

safe_list = glob.glob("/home/oarcher/SAFE/S1A*.SAFE")
print('about to get product information on %d SAFEs' % len(safe_list))

about to get product information on 44 SAFEs

df = xsar.product_info(safe_list)
df

	ipf	platform	swath	product	pols	multi	meta
SENTINEL1_DS:/home/oarcher/SAFE/S1A_IW_GRDH_1SDV_20201123T001313_20201123T001338_035368_0421EE_BF23.SAFE:IW	3.31	SENTINEL-1A	IW	GRD	VV VH	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_IW_GRDH_1SDV_20201123T001108_20201123T001133_035368_0421EE_2218.SAFE:IW	3.31	SENTINEL-1A	IW	GRD	VV VH	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_IW_GRDH_1SDV_20201123T001248_20201123T001313_035368_0421EE_A1E3.SAFE:IW	3.31	SENTINEL-1A	IW	GRD	VV VH	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_IW_GRDH_1SDV_20170907T102951_20170907T103020_018268_01EB76_44BE.SAFE:IW	2.84	SENTINEL-1A	IW	GRD	VV VH	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_IW_GRDH_1SDV_20200406T071419_20200406T071444_032003_03B25D_FFFF.SAFE:IW	3.2	SENTINEL-1A	IW	GRD	VV VH	False	single SentinelMeta object
...	...	...	...	...	...	...	...
SENTINEL1_DS:/home/oarcher/SAFE/S1A_EW_GRDM_1SDV_20201029T205631_20201029T205735_035016_0415BC_00C3.SAFE:EW	3.31	SENTINEL-1A	EW	GRD	VV VH	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_IW_GRDH_1SDV_20201123T001133_20201123T001158_035368_0421EE_707B.SAFE:IW	3.31	SENTINEL-1A	IW	GRD	VV VH	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_IW_GRDH_1SDV_20201123T000928_20201123T000953_035368_0421EE_3F2B.SAFE:IW	3.31	SENTINEL-1A	IW	GRD	VV VH	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_EW_GRDM_1SDV_20200316T020222_20200316T020322_031694_03A788_FFFF.SAFE:EW	3.2	SENTINEL-1A	EW	GRD	VV VH	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_EW_GRDM_1SDV_20191105T195834_20191105T195916_029780_03651B_4A35.SAFE:EW	3.1	SENTINEL-1A	EW	GRD	VV VH	False	single SentinelMeta object

71 rows × 7 columns

Each line represent metadata information for all datasets founds in safe_list.

There are more dataset than paths, because some SAFEs can have multiple subdatasets (like WV or IW_SLC)

xsar.product_info returns all datasets, including nested one and multi ones

the 'multi' columns allows to see witch safe are multi datasets

df[df['multi']]

	ipf	platform	swath	product	pols	multi	meta
SENTINEL1_DS:/home/oarcher/SAFE/S1A_IW_SLC__1SDV_20170907T102951_20170907T103021_018268_01EB76_AE7B.SAFE:	2.84	SENTINEL-1A	IW	SLC	VV VH	True	multi (3) SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:	2.7	SENTINEL-1A	WV	SLC	VV	True	multi (24) SentinelMeta object

a multi dataset cannot be directly opened as an xarray dataset with xsar.open_dataset. The user must provide a single dataset to xsar.open_dataset

Here are the openable WV datasets:

df[(df['swath'] == 'WV') & ~df['multi'] ]

	ipf	platform	swath	product	pols	multi	meta
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_001	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_010	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_011	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_012	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_013	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_014	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_015	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_016	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_017	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_018	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_019	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_002	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_020	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_021	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_022	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_023	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_024	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_003	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_004	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_005	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_006	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_007	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_008	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object
SENTINEL1_DS:/home/oarcher/SAFE/S1A_WV_SLC__1SSV_20160510T101603_20160510T102143_011195_010EA1_7038.SAFE:WV_009	2.7	SENTINEL-1A	WV	SLC	VV	False	single SentinelMeta object

There are 3 ways to open a dataset with xsar.open_dataset:

1. by path This will fail if the safe path as subdatasets
2. by gdal dataset id
3. by meta object

# those commands are equivatents:
ds1 = xsar.open_dataset('/home/oarcher/SAFE/S1A_IW_GRDH_1SDV_20201123T001313_20201123T001338_035368_0421EE_BF23.SAFE')
ds2 = xsar.open_dataset('SENTINEL1_DS:/home/oarcher/SAFE/S1A_IW_GRDH_1SDV_20201123T001313_20201123T001338_035368_0421EE_BF23.SAFE:IW')
meta = df.loc['SENTINEL1_DS:/home/oarcher/SAFE/S1A_IW_GRDH_1SDV_20201123T001313_20201123T001338_035368_0421EE_BF23.SAFE:IW']['meta']
ds3 = xsar.open_dataset(meta) 

extended columns for xsar.product_info

with 'columns="spatial"' keyword, the dataframe has also 'time_range' and 'footprint' columns

df = xsar.product_info(safe_list,columns='spatial')
import geopandas as gpd
import matplotlib.pyplot as plt
fig,ax = fig, ax = plt.subplots(figsize=(20,20))
world = gpd.read_file(gpd.datasets.get_path('naturalearth_lowres'))
world_map = world.plot(color='white', edgecolor='black',ax=ax) 
df.footprint.plot(ax=world_map)

<AxesSubplot:>

using meta object

Dataset attributes are stored in ds1.attrs, but those attributes must respect the CF convention to be properly dumped to disk.

So those attributes are numbers or strings, and are not easyly handable by the python programmer

# this is a string
ds3.attrs['footprint']

'POLYGON ((86.68592421893726 21.88782026084792, 84.26341868599728 22.31126646161592, 83.97503900505524 20.80381451549505, 86.37079841612797 20.37730344779156, 86.68592421893726 21.88782026084792))'

# and this is a shapely polygon
meta.footprint

meta has also some usefull methods

help(meta.ll2coords)

Help on method ll2coords in module xsar.sentinel1.Sentinel1:

ll2coords(lon, lat) method of xsar.sentinel1.Sentinel1.SentinelMeta instance
    Get `(atracks, xtracks)` from `(lon, lat)`
    
    Parameters
    ----------
    lon: np.array
    lat: np.array
    
    `lat` and `lon` must have the same shape.
    
    Returns
    -------
    tuple of np.array
        (atracks, xtracks)
    
    Examples:
    _________
        get nearest (atrack,xtrack) from (lon,lat) = (84.81, 21.32) in ds, with out of bounds check.
        ```
        >>> (atrack, xtrack) = meta.ll2coords(lon, lat)
        >>> (atrack, xtrack)
        (9752.766349989339, 17852.571322887554)
        >>> try:
        ...     # select the nearest valid pixel in ds
        ...     ds_nearest = ds.sel(atrack=atrack,xtrack=xtrack,method='nearest',tolerance=1)
        ...     (atrack, xtrack) = (ds_nearest.atrack.values.item() , ds_nearest.xtrack.values.item())
        ... except KeyError:
        ...     # out of bounds, because of `tolerance` keyword
        ...     # if ds is resampled, tolerance should be computed from resolution/2 (for ex 5 for a resolution of 10)
        ...     (atrack, xtrack) = (np.nan, np.nan)
        ```
    
    Notes:
    ------
        `atracks` and `xtracks` are float, with no out of bounds checks
    
    See Also:
    ---------
        self.coords2ll

# find (atrack,xtrack) from (lon,lat)

lon = 84.81
lat = 21.31
(atrack, xtrack) = meta.ll2coords(lon, lat)
(atrack, xtrack)

(9752.766349989339, 17852.571322887554)

# (atrack,xtrack) are float, so we are able to know the exact position, even at subpixel level
# to convert them to existing index, we can use .sel with 'nearest' method:
try:
    ds_nearest = ds1.sel(atrack=atrack,xtrack=xtrack,method='nearest',tolerance=0.5)
    (atrack, xtrack) = (ds_nearest.atrack.values.item() , ds_nearest.xtrack.values.item())
except:
    # out of bounds
    (atrack, xtrack) = (np.nan, np.nan)
(atrack, xtrack)

(9752.5, 17852.5)