def read_mtd(self) -> (etree._Element, str):
"""
Read metadata and outputs the metadata XML root and its namespace
.. code-block:: python
>>> from eoreader.reader import Reader
>>> path = r"S2A_MSIL1C_20200824T110631_N0209_R137_T30TTK_20200824T150432.SAFE.zip"
>>> prod = Reader().open(path)
>>> prod.read_mtd()
(<Element {https://psd-14.sentinel2.eo.esa.int/PSD/S2_PDI_Level-2A_Tile_Metadata.xsd}Level-2A_Tile_ID at ...>,
'{https://psd-14.sentinel2.eo.esa.int/PSD/S2_PDI_Level-2A_Tile_Metadata.xsd}')
Returns:
(etree._Element, str): Metadata XML root and its namespace
"""
# Get MTD XML file
if self.is_archived:
root = files.read_archived_xml(self.path, ".*GRANULE.*\.xml")
else:
# Open metadata file
try:
mtd_file = glob.glob(
os.path.join(self.path, "GRANULE", "*", "*.xml"))[0]
# pylint: disable=I1101:
# Module 'lxml.etree' has no 'parse' member, but source is unavailable.
xml_tree = etree.parse(mtd_file)
root = xml_tree.getroot()
except IndexError as ex:
raise InvalidProductError(
f"Metadata file not found in {self.path}") from ex
# Get namespace
idx = root.tag.rindex("}")
namespace = root.tag[:idx + 1]
return root, namespace
def read_mtd(self) -> (etree._Element, str):
"""
Read metadata and outputs the metadata XML root and its namespace
.. code-block:: python
>>> from eoreader.reader import Reader
>>> path = r"LC08_L1GT_023030_20200518_20200527_01_T2"
>>> prod = Reader().open(path)
>>> prod.read_mtd()
(<Element {http://www.rsi.ca/rs2/prod/xml/schemas}product at 0x1c0efbd37c8>,
'{http://www.rsi.ca/rs2/prod/xml/schemas}')
Returns:
(etree._Element, str): Metadata XML root and its namespace
"""
# Get MTD XML file
if self.is_archived:
root = files.read_archived_xml(self.path, ".*product\.xml")
else:
# Open metadata file
try:
mtd_file = glob.glob(os.path.join(self.path, "product.xml"))[0]
# pylint: disable=I1101:
# Module 'lxml.etree' has no 'parse' member, but source is unavailable.
xml_tree = etree.parse(mtd_file)
root = xml_tree.getroot()
except IndexError as ex:
raise InvalidProductError(
f"Metadata file (product.xml) not found in {self.path}"
) from ex
# Get namespace
idx = root.tag.rindex("}")
namespace = root.tag[:idx + 1]
return root, namespace
def _read_mtd(self, mtd_from_path: str, mtd_archived: str = None):
"""
Read metadata and outputs the metadata XML root and its namespaces as a dicts as a dict
Args:
mtd_from_path (str): Metadata regex (glob style) to find from extracted product
mtd_archived (str): Metadata regex (re style) to find from archived product
Returns:
(etree._Element, dict): Metadata XML root and its namespaces
"""
if self.is_archived:
root = files.read_archived_xml(self.path, mtd_archived)
else:
# ONLY FOR COLLECTION 2
try:
mtd_file = glob.glob(os.path.join(self.path, mtd_from_path),
recursive=True)[0]
# pylint: disable=I1101:
# Module 'lxml.etree' has no 'parse' member, but source is unavailable.
xml_tree = etree.parse(mtd_file)
root = xml_tree.getroot()
except IndexError as ex:
raise InvalidProductError(
f"Metadata file ({mtd_from_path}) not found in {self.path}"
) from ex
# Get namespaces map (only useful ones)
nsmap = {key: f"{{{ns}}}" for key, ns in root.nsmap.items()}
pop_list = ["xsi", "xs", "xlink"]
for ns in pop_list:
if ns in nsmap.keys():
nsmap.pop(ns)
return root, nsmap
def read_mtd(self) -> (etree._Element, str):
"""
Read metadata and outputs the metadata XML root and its namespace
.. code-block:: python
>>> from eoreader.reader import Reader
>>> path = r"S1A_IW_GRDH_1SDV_20191215T060906_20191215T060931_030355_0378F7_3696.zip"
>>> prod = Reader().open(path)
>>> prod.read_mtd()
(<Element product at 0x1832895d788>, '')
Returns:
(etree._Element, str): Metadata XML root and its namespace
"""
# Get MTD XML file
if self.is_archived:
root = files.read_archived_xml(self.path, ".*annotation.*\.xml")
else:
# Open metadata file
try:
mtd_file = glob.glob(os.path.join(self.path, "annotation", "*.xml"))[0]
# pylint: disable=I1101:
# Module 'lxml.etree' has no 'parse' member, but source is unavailable.
xml_tree = etree.parse(mtd_file)
root = xml_tree.getroot()
except IndexError as ex:
raise InvalidProductError(
f"Metadata file (product.xml) not found in {self.path}"
) from ex
# Get namespace
namespace = ""
return root, namespace
def read_mtd(self) -> (etree._Element, str):
"""
Read metadata and outputs the metadata XML root and its namespace
.. code-block:: python
>>> from eoreader.reader import Reader
>>> path = r"SENTINEL2B_20190401-105726-885_L2A_T31UEQ_D_V2-0.zip"
>>> prod = Reader().open(path)
>>> prod.read_mtd()
(<Element Muscate_Metadata_Document at 0x252d2071e88>, '')
Returns:
(etree._Element, str): Metadata XML root and its namespace
"""
# Get MTD XML file
if self.is_archived:
root = files.read_archived_xml(self.path, ".*MTD_ALL\.xml")
else:
# Open metadata file
try:
mtd_xml = glob.glob(os.path.join(self.path, "*MTD_ALL.xml"))[0]
# pylint: disable=I1101:
# Module 'lxml.etree' has no 'parse' member, but source is unavailable.
xml_tree = etree.parse(mtd_xml)
root = xml_tree.getroot()
except IndexError as ex:
raise InvalidProductError(
f"Metadata file not found in {self.path}"
) from ex
# Get namespace
namespace = ""
return root, namespace
def read_mtd(
self,
force_pd=False) -> Union[pd.DataFrame, Tuple[etree._Element, str]]:
"""
Read Landsat metadata as:
- a `pandas.DataFrame` whatever its collection is (by default for collection 1)
- a XML root + its namespace if the product is retrieved from the 2nd collection (by default for collection 2)
.. code-block:: python
>>> from eoreader.reader import Reader
>>> path = r"LC08_L1GT_023030_20200518_20200527_01_T2"
>>> prod = Reader().open(path)
>>> # COLLECTION 1 : Open metadata as panda DataFrame
>>> prod.read_mtd()
NAME ORIGIN ... RESAMPLING_OPTION
value "Image courtesy of the U.S. Geological Survey" ... "CUBIC_CONVOLUTION"
[1 rows x 197 columns]
>>> # COLLECTION 2 : Open metadata as XML
>>> path = r"LC08_L1TP_200030_20201220_20210310_02_T1" # Collection 2
>>> prod = Reader().open(path)
>>> prod.read_mtd()
(<Element LANDSAT_METADATA_FILE at 0x19229016048>, '')
>>> # COLLECTION 2 : Force to pandas.DataFrame
>>> prod.read_mtd(force_pd=True)
NAME ORIGIN ... RESAMPLING_OPTION
value "Image courtesy of the U.S. Geological Survey" ... "CUBIC_CONVOLUTION"
[1 rows x 263 columns]
Args:
force_pd (bool): If collection 2, return a pandas.DataFrame instead of a XML root + namespace
Returns:
pd.DataFrame: Metadata as a Pandas DataFrame
"""
# WARNING: always use force_pd in this class !
as_pd = (self._collection == LandsatCollection.COL_1) or force_pd
if as_pd:
mtd_name = f"{self.name}_MTL.txt"
if self.is_archived:
# We need to extract the file in memory to be used with pandas
tar_ds = tarfile.open(self.path, "r")
info = [
f.name for f in tar_ds.getmembers() if mtd_name in f.name
][0]
mtd_path = tar_ds.extractfile(info)
else:
# FOR COLLECTION 1 AND 2
tar_ds = None
mtd_path = os.path.join(self.path, mtd_name)
if not os.path.isfile(mtd_path):
raise FileNotFoundError(
f"Unable to find the metadata file associated with {self.path}"
)
# Parse
mtd_data = pd.read_table(
mtd_path,
sep="\s=\s",
names=["NAME", "value"],
skipinitialspace=True,
engine="python",
)
# Workaround an unexpected behaviour in pandas !
if any(mtd_data.NAME == "="):
mtd_data = pd.read_table(
mtd_path,
sep="=",
names=["NAME", "=", "value"],
usecols=[0, 2],
skipinitialspace=True,
)
# Remove useless rows
mtd_data = mtd_data[~mtd_data["NAME"].
isin(["GROUP", "END_GROUP", "END"])]
# Set index
mtd_data = mtd_data.set_index("NAME").T
# Close if needed
if tar_ds:
tar_ds.close()
else:
if self.is_archived:
root = files.read_archived_xml(self.path,
f".*{self.name}_MTL.xml")
else:
# ONLY FOR COLLECTION 2
try:
mtd_file = glob.glob(
os.path.join(self.path, f"{self.name}_MTL.xml"))[0]
# pylint: disable=I1101:
# Module 'lxml.etree' has no 'parse' member, but source is unavailable.
xml_tree = etree.parse(mtd_file)
root = xml_tree.getroot()
except IndexError as ex:
raise InvalidProductError(
f"Metadata file ({self.name}.xml) not found in {self.path}"
) from ex
# Get namespace
namespace = "" # No namespace here
mtd_data = (root, namespace)
return mtd_data