# -*- coding: utf-8 -*-
# Copyright 2018, CS GROUP - France, https://www.csgroup.eu/
#
# This file is part of EODAG project
# https://www.github.com/CS-SI/EODAG
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""Miscellaneous utilities to be used throughout eodag.
Everything that does not fit into one of the specialised categories of utilities in
this package should go here
"""
from __future__ import annotations
import ast
import datetime
import errno
import functools
import hashlib
import inspect
import logging as py_logging
import os
import re
import shutil
import string
import types
import unicodedata
import warnings
from collections import defaultdict
from copy import deepcopy as copy_deepcopy
from datetime import datetime as dt
from email.message import Message
from glob import glob
from importlib.metadata import metadata
from itertools import repeat, starmap
from pathlib import Path
from tempfile import mkdtemp
from typing import (
TYPE_CHECKING,
Any,
Callable,
Dict,
Iterator,
List,
Optional,
Tuple,
Type,
Union,
)
# All modules using these should import them from utils package
from urllib.parse import ( # noqa; noqa
parse_qs,
parse_qsl,
quote,
unquote,
urlencode,
urljoin,
urlparse,
urlsplit,
)
from urllib.request import url2pathname
try:
from typing import Annotated, get_args, get_origin # noqa
except ImportError:
# for python < 3.9
from typing_extensions import Annotated, get_args, get_origin # type: ignore # noqa
import click
import orjson
import shapefile
import shapely.wkt
import yaml
from dateutil.parser import isoparse
from dateutil.tz import UTC
from jsonpath_ng import jsonpath
from jsonpath_ng.ext import parse
from jsonpath_ng.jsonpath import Child, Fields, Index, Root, Slice
from requests import HTTPError
from shapely.geometry import Polygon, shape
from shapely.geometry.base import BaseGeometry
from tqdm.auto import tqdm
from eodag.utils import logging as eodag_logging
from eodag.utils.exceptions import MisconfiguredError
if TYPE_CHECKING:
from jsonpath_ng import JSONPath
from eodag.api.product import EOProduct
logger = py_logging.getLogger("eodag.utils")
DEFAULT_PROJ = "EPSG:4326"
GENERIC_PRODUCT_TYPE = "GENERIC_PRODUCT_TYPE"
eodag_version = metadata("eodag")["Version"]
USER_AGENT = {"User-Agent": f"eodag/{eodag_version}"}
HTTP_REQ_TIMEOUT = 5 # in seconds
DEFAULT_STREAM_REQUESTS_TIMEOUT = 60 # in seconds
# default wait times in minutes
DEFAULT_DOWNLOAD_WAIT = 2 # in minutes
DEFAULT_DOWNLOAD_TIMEOUT = 20 # in minutes
JSONPATH_MATCH = re.compile(r"^[\{\(]*\$(\..*)*$")
WORKABLE_JSONPATH_MATCH = re.compile(r"^\$(\.[a-zA-Z0-9-_:\.\[\]\"\(\)=\?\*]+)*$")
ARRAY_FIELD_MATCH = re.compile(r"^[a-zA-Z0-9-_:]+(\[[0-9\*]+\])+$")
# pagination defaults
DEFAULT_PAGE = 1
DEFAULT_ITEMS_PER_PAGE = 20
# Default maximum number of items per page requested by search_all. 50 instead of 20
# (DEFAULT_ITEMS_PER_PAGE) to increase it to the known and current minimum value (mundi)
DEFAULT_MAX_ITEMS_PER_PAGE = 50
def _deprecated(reason: str = "", version: Optional[str] = None) -> Callable[..., Any]:
"""Simple decorator to mark functions/methods/classes as deprecated.
Warning: Does not work with staticmethods!
@deprecate(reason="why", version="1.2")
def foo():
pass
foo()
DeprecationWarning: Call to deprecated function/method foo (why) -- Deprecated since v1.2
"""
def decorator(callable: Callable[..., Any]) -> Any:
if inspect.isclass(callable):
ctype = "class"
else:
ctype = "function/method"
cname = callable.__name__
reason_ = f"({reason})" if reason else ""
version_ = f" -- Deprecated since v{version}" if version else ""
@functools.wraps(callable)
def wrapper(*args: Any, **kwargs: Any) -> Any:
with warnings.catch_warnings():
warnings.simplefilter("always", DeprecationWarning)
warnings.warn(
f"Call to deprecated {ctype} {cname} {reason_}{version_}",
category=DeprecationWarning,
stacklevel=2,
)
return callable(*args, **kwargs)
return wrapper
return decorator
class FloatRange(click.types.FloatParamType):
"""A parameter that works similar to :data:`click.FLOAT` but restricts the
value to fit into a range. Fails if the value doesn't fit into the range.
"""
name = "percentage"
def __init__(
self, min: Optional[float] = None, max: Optional[float] = None
) -> None:
self.min = min
self.max = max
def convert(
self,
value: Any,
param: Optional[click.core.Parameter],
ctx: Optional[click.core.Context],
) -> Any:
"""Convert value"""
rv = click.types.FloatParamType.convert(self, value, param, ctx)
if (
self.min is not None
and rv < self.min
or self.max is not None
and rv > self.max
):
if self.min is None:
self.fail(
"%s is bigger than the maximum valid value " "%s." % (rv, self.max),
param,
ctx,
)
elif self.max is None:
self.fail(
"%s is smaller than the minimum valid value "
"%s." % (rv, self.min),
param,
ctx,
)
else:
self.fail(
"%s is not in the valid range of %s to %s."
% (rv, self.min, self.max),
param,
ctx,
)
return rv
def __repr__(self) -> str:
return "FloatRange(%r, %r)" % (self.min, self.max)
def slugify(value: Any, allow_unicode: bool = False) -> str:
"""Copied from Django Source code, only modifying last line (no need for safe
strings).
source: https://github.com/django/django/blob/master/django/utils/text.py
Convert to ASCII if 'allow_unicode' is False. Convert spaces to hyphens.
Remove characters that aren't alphanumerics, underscores, or hyphens.
Convert to lowercase. Also strip leading and trailing whitespace.
"""
value = str(value)
if allow_unicode:
value = unicodedata.normalize("NFKC", value)
else:
value = (
unicodedata.normalize("NFKD", value)
.encode("ascii", "ignore")
.decode("ascii")
)
value = re.sub(r"[^\w\s-]", "", value).strip().lower()
return re.sub(r"[-\s]+", "-", value)
def sanitize(value: str) -> str:
"""Sanitize string to be used as a name of a directory.
>>> sanitize('productName')
'productName'
>>> sanitize('name with multiple spaces')
'name_with_multiple_spaces'
>>> sanitize('âtre fête île alcôve bûche çà génèse où Noël ovoïde capharnaüm')
'atre_fete_ile_alcove_buche_ca_genese_ou_Noel_ovoide_capharnaum'
>>> sanitize('replace,ponctuation:;signs!?byunderscorekeeping-hyphen.dot_and_underscore') # noqa
'replace_ponctuation_signs_byunderscorekeeping-hyphen.dot_and_underscore'
"""
# remove accents
rv = strip_accents(value)
# replace punctuation signs and spaces by underscore
# keep hyphen, dot and underscore from punctuation
tobereplaced = re.sub(r"[-_.]", "", string.punctuation)
# add spaces to be removed
tobereplaced += r"\s"
rv = re.sub(r"[" + tobereplaced + r"]+", "_", rv)
return str(rv)
def strip_accents(s: str) -> str:
"""Strip accents of a string.
>>> strip_accents('productName')
'productName'
>>> strip_accents('génèse')
'genese'
>>> strip_accents('preserve-punct-special-chars:;,?!§%$£œ')
'preserve-punct-special-chars:;,?!§%$£œ'
"""
# Mn stands for a nonspacing combining mark
return "".join(
c for c in unicodedata.normalize("NFD", s) if unicodedata.category(c) != "Mn"
)
def uri_to_path(uri: str) -> str:
"""
Convert a file URI (e.g. 'file:///tmp') to a local path (e.g. '/tmp')
"""
if not uri.startswith("file"):
raise ValueError("A file URI must be provided (e.g. 'file:///tmp'")
_, _, path, _, _ = urlsplit(uri)
# On Windows urlsplit returns the path starting with a slash ('/C:/User)
path = url2pathname(path)
# url2pathname removes it
return path
def path_to_uri(path: str) -> str:
"""Convert a local absolute path to a file URI"""
return Path(path).as_uri()
def mutate_dict_in_place(func: Callable[[Any], Any], mapping: Dict[Any, Any]) -> None:
"""Apply func to values of mapping.
The mapping object's values are modified in-place. The function is recursive,
allowing to also modify values of nested dicts that may be level-1 values of
mapping.
:param func: A function to apply to each value of mapping which is not a dict object
:type func: func
:param mapping: A Python dict object
:type mapping: dict
:returns: None
"""
for key, value in mapping.items():
if isinstance(value, dict):
mutate_dict_in_place(func, value)
else:
mapping[key] = func(value)
def merge_mappings(mapping1: Dict[Any, Any], mapping2: Dict[Any, Any]) -> None:
"""Merge two mappings with string keys, values from `mapping2` overriding values
from `mapping1`.
Do its best to detect the key in `mapping1` to override. For example::
>>> mapping2 = {"keya": "new"}
>>> mapping1 = {"keyA": "obsolete"}
>>> merge_mappings(mapping1, mapping2)
>>> mapping1
{'keyA': 'new'}
If mapping2 has a key that cannot be detected in mapping1, this new key is added
to mapping1 as is.
:param mapping1: The mapping containing values to be overridden
:type mapping1: dict
:param mapping2: The mapping containing values that will override the
first mapping
:type mapping2: dict
"""
# A mapping between mapping1 keys as lowercase strings and original mapping1 keys
m1_keys_lowercase = {key.lower(): key for key in mapping1}
for key, value in mapping2.items():
if isinstance(value, dict):
try:
merge_mappings(mapping1[key], value)
except KeyError:
# If the key from mapping2 is not in mapping1, it is either key is
# the lowercased form of the corresponding key in mapping1 or because
# key is a new key to be added in mapping1
current_value = mapping1.setdefault(m1_keys_lowercase.get(key, key), {})
if not current_value:
current_value.update(value)
else:
merge_mappings(current_value, value)
else:
# Even for "scalar" values (a.k.a not nested structures), first check if
# the key from mapping1 is not the lowercase version of a key in mapping2.
# Otherwise, create the key in mapping1. This is the meaning of
# `m1_keys_lowercase.get(key, key)`
current_value = mapping1.get(m1_keys_lowercase.get(key, key), None)
if current_value is not None:
current_value_type = type(current_value)
new_value_type = type(value)
try:
# If current or new value is a list (search queryable parameter), simply replace current with new
if (
new_value_type == list
and current_value_type != list
or new_value_type != list
and current_value_type == list
):
mapping1[m1_keys_lowercase.get(key, key)] = value
else:
mapping1[m1_keys_lowercase.get(key, key)] = cast_scalar_value(
value, current_value_type
)
except (TypeError, ValueError):
# Ignore any override value that does not have the same type
# as the default value
logger.debug(
f"Ignored '{key}' setting override from '{current_value}' to '{value}', "
f"(could not cast {new_value_type} to {current_value_type})"
)
pass
else:
mapping1[key] = value
def maybe_generator(obj: Any) -> Iterator[Any]:
"""Generator function that get an arbitrary object and generate values from it if
the object is a generator."""
if isinstance(obj, types.GeneratorType):
for elt in obj:
yield elt
else:
yield obj
def get_timestamp(date_time: str) -> float:
"""Return the Unix timestamp of an ISO8601 date/datetime in seconds.
If the datetime has no offset, it is assumed to be an UTC datetime.
:param date_time: The datetime string to return as timestamp
:type date_time: str
:returns: The timestamp corresponding to the date_time string in seconds
:rtype: float
"""
dt = isoparse(date_time)
if not dt.tzinfo:
dt = dt.replace(tzinfo=UTC)
return dt.timestamp()
def datetime_range(start: dt, end: dt) -> Iterator[dt]:
"""Generator function for all dates in-between start and end date."""
delta = end - start
for nday in range(delta.days + 1):
yield start + datetime.timedelta(days=nday)
class DownloadedCallback:
"""Example class for callback after each download in :meth:`~eodag.api.core.EODataAccessGateway.download_all`"""
def __call__(self, product: EOProduct) -> None:
"""Callback
:param product: The downloaded EO product
:type product: :class:`~eodag.api.product._product.EOProduct`
"""
logger.debug("Download finished for the product %s", product)
[docs]class ProgressCallback(tqdm):
"""A callable used to render progress to users for long running processes.
It inherits from `tqdm.auto.tqdm`, and accepts the same arguments on
instantiation: `iterable`, `desc`, `total`, `leave`, `file`, `ncols`,
`mininterval`, `maxinterval`, `miniters`, `ascii`, `disable`, `unit`,
`unit_scale`, `dynamic_ncols`, `smoothing`, `bar_format`, `initial`,
`position`, `postfix`, `unit_divisor`.
It can be globally disabled using `eodag.utils.logging.setup_logging(0)` or
`eodag.utils.logging.setup_logging(level, no_progress_bar=True)`, and
individually disabled using `disable=True`.
"""
def __init__(self, *args: Any, **kwargs: Any) -> None:
self.kwargs = kwargs.copy()
if "unit" not in kwargs:
kwargs["unit"] = "B"
if "unit_scale" not in kwargs:
kwargs["unit_scale"] = True
if "desc" not in kwargs:
kwargs["desc"] = ""
if "position" not in kwargs:
kwargs["position"] = 0
if "disable" not in kwargs:
kwargs["disable"] = eodag_logging.disable_tqdm
if "dynamic_ncols" not in kwargs:
kwargs["dynamic_ncols"] = True
super(ProgressCallback, self).__init__(*args, **kwargs)
def __call__(self, increment: int, total: Optional[int] = None) -> None:
"""Update the progress bar.
:param increment: Amount of data already processed
:type increment: int
:param total: (optional) Maximum amount of data to be processed
:type total: int
"""
if total is not None and total != self.total:
self.reset(total=total)
self.update(increment)
def copy(self, *args: Any, **kwargs: Any) -> ProgressCallback:
"""Returns another progress callback using the same initial
keyword-arguments.
Optional `args` and `kwargs` parameters will be used to create a
new `~eodag.utils.ProgressCallback` instance, overriding initial
`kwargs`.
"""
return ProgressCallback(*args, **dict(self.kwargs, **kwargs))
@_deprecated(reason="Use ProgressCallback class instead", version="2.2.1")
class NotebookProgressCallback(tqdm):
"""A custom progress bar to be used inside Jupyter notebooks"""
pass
@_deprecated(reason="Use ProgressCallback class instead", version="2.2.1")
def get_progress_callback() -> tqdm:
"""Get progress_callback"""
return tqdm()
def repeatfunc(func: Callable[..., Any], n: int, *args: Any) -> starmap:
"""Call `func` `n` times with `args`"""
return starmap(func, repeat(args, n))
def makedirs(dirpath: str) -> None:
"""Create a directory in filesystem with parents if necessary"""
try:
os.makedirs(dirpath)
except OSError as err:
# Reraise the error unless it's about an already existing directory
if err.errno != errno.EEXIST or not os.path.isdir(dirpath):
raise
def rename_subfolder(dirpath: str, name: str) -> None:
"""Rename first subfolder found in dirpath with given name,
raise RuntimeError if no subfolder can be found
:param dirpath: path to the directory containing the subfolder
:type dirpath: str
:param name: new name of the subfolder
:type name: str
:raises: RuntimeError
Example:
>>> import os
>>> import tempfile
>>> with tempfile.TemporaryDirectory() as tmpdir:
... somefolder = os.path.join(tmpdir, "somefolder")
... otherfolder = os.path.join(tmpdir, "otherfolder")
... os.makedirs(somefolder)
... assert os.path.isdir(somefolder) and not os.path.isdir(otherfolder)
... rename_subfolder(tmpdir, "otherfolder")
... assert not os.path.isdir(somefolder) and os.path.isdir(otherfolder)
Before:
$ tree <tmp-folder>
<tmp-folder>
└── somefolder
└── somefile
After:
$ tree <tmp-folder>
<tmp-folder>
└── otherfolder
└── somefile
"""
try:
subdir, *_ = (p for p in glob(os.path.join(dirpath, "*")) if os.path.isdir(p))
except ValueError:
raise RuntimeError(f"No subfolder was found in {dirpath}")
os.rename(
subdir,
os.path.join(dirpath, name),
)
def format_dict_items(
config_dict: Dict[str, Any], **format_variables: Any
) -> Dict[Any, Any]:
r"""Recursive apply string.format(\**format_variables) to dict elements
>>> format_dict_items(
... {"foo": {"bar": "{a}"}, "baz": ["{b}?", "{b}!"]},
... **{"a": "qux", "b": "quux"},
... ) == {"foo": {"bar": "qux"}, "baz": ["quux?", "quux!"]}
True
:param config_dict: Dictionnary having values that need to be parsed
:type config_dict: dict
:param format_variables: Variables used as args for parsing
:type format_variables: dict
:returns: Updated dict
:rtype: dict
"""
return dict_items_recursive_apply(config_dict, format_string, **format_variables)
def jsonpath_parse_dict_items(
jsonpath_dict: Dict[str, Any], values_dict: Dict[str, Any]
) -> Dict[Any, Any]:
"""Recursive parse jsonpath elements in dict
>>> import jsonpath_ng.ext as jsonpath
>>> jsonpath_parse_dict_items(
... {"foo": {"bar": parse("$.a.b")}, "qux": [parse("$.c"), parse("$.c")]},
... {"a":{"b":"baz"}, "c":"quux"}
... ) == {'foo': {'bar': 'baz'}, 'qux': ['quux', 'quux']}
True
:param jsonpath_dict: Dictionnary having values that need to be parsed
:type jsonpath_dict: dict
:param values_dict: Values dict used as args for parsing
:type values_dict: dict
:returns: Updated dict
:rtype: dict
"""
return dict_items_recursive_apply(jsonpath_dict, parse_jsonpath, **values_dict)
def update_nested_dict(
old_dict: Dict[Any, Any],
new_dict: Dict[Any, Any],
extend_list_values: bool = False,
allow_empty_values: bool = False,
allow_extend_duplicates: bool = True,
) -> Dict[Any, Any]:
"""Update recursively old_dict items with new_dict ones
>>> update_nested_dict(
... {"a": {"a.a": 1, "a.b": 2}, "b": 3},
... {"a": {"a.a": 10}}
... ) == {'a': {'a.a': 10, 'a.b': 2}, 'b': 3}
True
>>> update_nested_dict(
... {"a": {"a.a": [1, 2]}},
... {"a": {"a.a": [10, 2]}},
... extend_list_values=True,
... allow_extend_duplicates=True
... ) == {'a': {'a.a': [1, 2, 10, 2]}}
True
>>> update_nested_dict(
... {"a": {"a.a": [1, 2]}},
... {"a": {"a.a": [10, 2]}},
... extend_list_values=True,
... allow_extend_duplicates=False
... ) == {'a': {'a.a': [1, 2, 10]}}
True
>>> update_nested_dict(
... {"a": {"a.a": 1, "a.b": 2}, "b": 3},
... {"a": {"a.a": None}},
... ) == {'a': {'a.a': 1, 'a.b': 2}, 'b': 3}
True
>>> update_nested_dict(
... {"a": {"a.a": 1, "a.b": 2}, "b": 3},
... {"a": {"a.a": None}},
... allow_empty_values=True
... ) == {'a': {'a.a': None, 'a.b': 2}, 'b': 3}
True
:param old_dict: Dict to be updated
:type old_dict: dict
:param new_dict: Incomming dict
:type new_dict: dict
:param extend_list_values: (optional) Extend old_dict value if both old/new values are lists
:type extend_list_values: bool
:param allow_empty_values: (optional) Allow update with empty values
:type allow_empty_values: bool
:returns: Updated dict
:rtype: dict
"""
for k, v in new_dict.items():
if k in old_dict.keys():
if isinstance(v, dict) and isinstance(old_dict[k], dict):
old_dict[k] = update_nested_dict(
old_dict[k],
v,
extend_list_values=extend_list_values,
allow_empty_values=allow_empty_values,
allow_extend_duplicates=allow_extend_duplicates,
)
elif (
extend_list_values
and isinstance(old_dict[k], list)
and isinstance(v, list)
and (
# no common elements
not any([x for x in v if x in old_dict[k]])
# common elements
or any([x for x in v if x in old_dict[k]])
and allow_extend_duplicates
)
):
old_dict[k].extend(v)
elif (
extend_list_values
and isinstance(old_dict[k], list)
and isinstance(v, list)
# common elements
and any([x for x in v if x in old_dict[k]])
and not allow_extend_duplicates
):
old_dict[k].extend([x for x in v if x not in old_dict[k]])
elif (v and not allow_empty_values) or allow_empty_values:
old_dict[k] = v
else:
old_dict[k] = v
return old_dict
def items_recursive_apply(
input_obj: Union[Dict[Any, Any], List[Any]],
apply_method: Callable[..., Any],
**apply_method_parameters: Any,
) -> Union[Dict[Any, Any], List[Any]]:
"""Recursive apply method to items contained in input object (dict or list)
>>> items_recursive_apply(
... {"foo": {"bar":"baz"}, "qux": ["a","b"]},
... lambda k,v,x: v.upper()+x, **{"x":"!"}
... ) == {'foo': {'bar': 'BAZ!'}, 'qux': ['A!', 'B!']}
True
>>> items_recursive_apply(
... [{"foo": {"bar":"baz"}}, "qux"],
... lambda k,v,x: v.upper()+x,
... **{"x":"!"})
[{'foo': {'bar': 'BAZ!'}}, 'QUX!']
>>> items_recursive_apply(
... "foo",
... lambda k,v,x: v.upper()+x,
... **{"x":"!"})
'foo'
:param input_obj: Input object (dict or list)
:type input_obj: Union[dict,list]
:param apply_method: Method to be applied to dict elements
:type apply_method: :func:`apply_method`
:param apply_method_parameters: Optional parameters passed to the method
:type apply_method_parameters: dict
:returns: Updated object
:rtype: Union[dict, list]
"""
if isinstance(input_obj, dict):
return dict_items_recursive_apply(
input_obj, apply_method, **apply_method_parameters
)
elif isinstance(input_obj, list):
return list_items_recursive_apply(
input_obj, apply_method, **apply_method_parameters
)
else:
logger.warning("Could not use items_recursive_apply on %s" % type(input_obj))
return input_obj
def dict_items_recursive_apply(
config_dict: Dict[Any, Any],
apply_method: Callable[..., Any],
**apply_method_parameters: Any,
) -> Dict[Any, Any]:
"""Recursive apply method to dict elements
>>> dict_items_recursive_apply(
... {"foo": {"bar": "baz"}, "qux": ["a", "b"]},
... lambda k, v, x: v.upper() + x, **{"x": "!"}
... ) == {'foo': {'bar': 'BAZ!'}, 'qux': ['A!', 'B!']}
True
:param config_dict: Input nested dictionnary
:type config_dict: dict
:param apply_method: Method to be applied to dict elements
:type apply_method: :func:`apply_method`
:param apply_method_parameters: Optional parameters passed to the method
:type apply_method_parameters: dict
:returns: Updated dict
:rtype: dict
"""
result_dict: Dict[Any, Any] = deepcopy(config_dict)
for dict_k, dict_v in result_dict.items():
if isinstance(dict_v, dict):
result_dict[dict_k] = dict_items_recursive_apply(
dict_v, apply_method, **apply_method_parameters
)
elif any(isinstance(dict_v, t) for t in (list, tuple)):
result_dict[dict_k] = list_items_recursive_apply(
dict_v, apply_method, **apply_method_parameters
)
else:
result_dict[dict_k] = apply_method(
dict_k, dict_v, **apply_method_parameters
)
return result_dict
def list_items_recursive_apply(
config_list: List[Any],
apply_method: Callable[..., Any],
**apply_method_parameters: Any,
) -> List[Any]:
"""Recursive apply method to list elements
>>> list_items_recursive_apply(
... [{"foo": {"bar": "baz"}}, "qux"],
... lambda k, v, x: v.upper() + x,
... **{"x": "!"})
[{'foo': {'bar': 'BAZ!'}}, 'QUX!']
:param config_list: Input list containing nested lists/dicts
:type config_list: list
:param apply_method: Method to be applied to list elements
:type apply_method: :func:`apply_method`
:param apply_method_parameters: Optional parameters passed to the method
:type apply_method_parameters: dict
:returns: Updated list
:rtype: list
"""
result_list = deepcopy(config_list)
for list_idx, list_v in enumerate(result_list):
if isinstance(list_v, dict):
result_list[list_idx] = dict_items_recursive_apply(
list_v, apply_method, **apply_method_parameters
)
elif any(isinstance(list_v, t) for t in (list, tuple)):
result_list[list_idx] = list_items_recursive_apply(
list_v, apply_method, **apply_method_parameters
)
else:
result_list[list_idx] = apply_method(
list_idx, list_v, **apply_method_parameters
)
return result_list
def items_recursive_sort(
input_obj: Union[List[Any], Dict[Any, Any]]
) -> Union[List[Any], Dict[Any, Any]]:
"""Recursive sort dict items contained in input object (dict or list)
>>> items_recursive_sort(
... {"b": {"b": "c", "a": 0}, "a": ["b", {2: 0, 0: 1, 1: 2}]},
... ) == {"a": ["b", {0: 1, 1: 2, 2: 0}], "b": {"a": 0, "b": "c"}}
True
>>> items_recursive_sort(["b", {2: 0, 0: 1, 1:2}])
['b', {0: 1, 1: 2, 2: 0}]
>>> items_recursive_sort("foo")
'foo'
:param input_obj: Input object (dict or list)
:type input_obj: Union[dict,list]
:returns: Updated object
:rtype: Union[dict, list]
"""
if isinstance(input_obj, dict):
return dict_items_recursive_sort(input_obj)
elif isinstance(input_obj, list):
return list_items_recursive_sort(input_obj)
else:
logger.warning("Could not use items_recursive_sort on %s" % type(input_obj))
return input_obj
def dict_items_recursive_sort(config_dict: Dict[Any, Any]) -> Dict[Any, Any]:
"""Recursive sort dict elements
>>> dict_items_recursive_sort(
... {"b": {"b": "c", "a": 0}, "a": ["b", {2: 0, 0: 1, 1: 2}]},
... ) == {"a": ["b", {0: 1, 1: 2, 2: 0}], "b": {"a": 0, "b": "c"}}
True
:param config_dict: Input nested dictionnary
:type config_dict: dict
:returns: Updated dict
:rtype: dict
"""
result_dict: Dict[Any, Any] = deepcopy(config_dict)
for dict_k, dict_v in result_dict.items():
if isinstance(dict_v, dict):
result_dict[dict_k] = dict_items_recursive_sort(dict_v)
elif any(isinstance(dict_v, t) for t in (list, tuple)):
result_dict[dict_k] = list_items_recursive_sort(dict_v)
else:
result_dict[dict_k] = dict_v
return dict(sorted(result_dict.items()))
def list_items_recursive_sort(config_list: List[Any]) -> List[Any]:
"""Recursive sort dicts in list elements
>>> list_items_recursive_sort(["b", {2: 0, 0: 1, 1: 2}])
['b', {0: 1, 1: 2, 2: 0}]
:param config_list: Input list containing nested lists/dicts
:type config_list: list
:returns: Updated list
:rtype: list
"""
result_list: List[Any] = deepcopy(config_list)
for list_idx, list_v in enumerate(result_list):
if isinstance(list_v, dict):
result_list[list_idx] = dict_items_recursive_sort(list_v)
elif any(isinstance(list_v, t) for t in (list, tuple)):
result_list[list_idx] = list_items_recursive_sort(list_v)
else:
result_list[list_idx] = list_v
return result_list
def string_to_jsonpath(*args: Any, force: bool = False) -> Union[str, JSONPath]:
"""Get jsonpath for "$.foo.bar" like string
>>> string_to_jsonpath(None, "$.foo.bar")
Child(Child(Root(), Fields('foo')), Fields('bar'))
>>> string_to_jsonpath("$.foo.bar")
Child(Child(Root(), Fields('foo')), Fields('bar'))
>>> string_to_jsonpath('$.foo[0][*]')
Child(Child(Child(Root(), Fields('foo')), Index(index=0)), Slice(start=None,end=None,step=None))
>>> string_to_jsonpath("foo")
'foo'
>>> string_to_jsonpath("foo", force=True)
Fields('foo')
:param args: Last arg as input string value, to be converted
:type args: str
:param force: force conversion even if input string is not detected as a jsonpath
:type force: bool
:returns: Parsed value
:rtype: str or Child or Root
"""
path_str: str = args[-1]
if JSONPATH_MATCH.match(str(path_str)) or force:
try:
common_jsonpath = "$"
common_jsonpath_parsed = Root()
# combine with common jsonpath if possible
if WORKABLE_JSONPATH_MATCH.match(path_str):
path_suffix = path_str[len(common_jsonpath) + 1 :]
path_splits = path_suffix.split(".") if path_suffix else []
parsed_path = common_jsonpath_parsed
for path_split in path_splits:
path_split = path_split.strip("'").strip('"')
if "[" in path_split and ARRAY_FIELD_MATCH.match(path_split):
# handle nested array
indexed_path_and_indexes = path_split[:-1].split("[")
indexed_path = indexed_path_and_indexes[0]
parsed_path = Child(parsed_path, Fields(indexed_path))
for idx in range(len(indexed_path_and_indexes) - 1):
index: Union[int, str] = (
indexed_path_and_indexes[idx + 1][:-1]
if idx < len(indexed_path_and_indexes) - 2
else indexed_path_and_indexes[idx + 1]
)
# wildcard index
if index == "*":
parsed_path = Child(
parsed_path,
Slice(start=None, end=None, step=None),
)
continue
try:
index = int(index)
except ValueError:
# unsupported index
parsed_path = cached_parse(path_str)
break
# integer index
parsed_path = Child(
parsed_path,
Index(index=index),
)
elif "[" in path_split:
# unsupported array field
parsed_path = cached_parse(path_str)
break
else:
parsed_path = Child(parsed_path, Fields(path_split))
return parsed_path
else:
return cached_parse(path_str)
except Exception: # jsonpath_ng does not provide a proper exception
# If str_value does not contain a jsonpath, return it as is
return path_str
else:
return path_str
def format_string(key: str, str_to_format: Any, **format_variables: Any) -> Any:
"""Format "{foo}" like string
>>> format_string(None, "foo {bar}, {baz} ?", **{"bar": "qux", "baz": "quux"})
'foo qux, quux ?'
:param key: Input item key
:type key: str
:param str_to_format: Input item value, to be parsed
:type str_to_format: str
:returns: Parsed value
:rtype: str
"""
if not isinstance(str_to_format, str):
return str_to_format
# eodag mappings function usage, e.g. '{foo#to_bar}'
COMPLEX_QS_REGEX = re.compile(r"^(.+=)?([^=]*)({.+})+([^=&]*)$")
if COMPLEX_QS_REGEX.match(str_to_format) and "#" in str_to_format:
from eodag.api.product.metadata_mapping import format_metadata
result = format_metadata(str_to_format, **format_variables)
else:
# defaultdict usage will return "" for missing keys in format_args
try:
result = str_to_format.format_map(defaultdict(str, **format_variables))
except TypeError as e:
raise MisconfiguredError(
f"Unable to format str={str_to_format} using {str(format_variables)}: {str(e)}"
)
# try to convert string to python object
try:
return ast.literal_eval(result)
except (SyntaxError, ValueError):
return result
def parse_jsonpath(
key: str, jsonpath_obj: Union[str, jsonpath.Child], **values_dict: Dict[str, Any]
) -> Optional[str]:
"""Parse jsonpah in jsonpath_obj using values_dict
>>> import jsonpath_ng.ext as jsonpath
>>> parse_jsonpath(None, parse("$.foo.bar"), **{"foo": {"bar": "baz"}})
'baz'
:param key: Input item key
:type key: str
:param jsonpath_obj: Input item value, to be parsed
:type jsonpath_obj: str or jsonpath.Child
:param values_dict: Values used as args for parsing
:type values_dict: dict
:returns: Parsed value
:rtype: str
"""
if isinstance(jsonpath_obj, jsonpath.Child):
match = jsonpath_obj.find(values_dict)
return match[0].value if len(match) == 1 else None
else:
return jsonpath_obj
def nested_pairs2dict(pairs: Union[List[Any], Any]) -> Union[Any, Dict[Any, Any]]:
"""Create a dict using nested pairs
>>> nested_pairs2dict([["foo", [["bar", "baz"]]]])
{'foo': {'bar': 'baz'}}
:param pairs: Pairs of key / value
:type pairs: list or Any
:returns: Created dict
:rtype: dict or Any
"""
d = {}
try:
for k, v in pairs:
if isinstance(v, list):
v = nested_pairs2dict(v)
d[k] = v
except ValueError:
return pairs
return d
def get_geometry_from_various(
locations_config: List[Dict[str, Any]] = [], **query_args: Any
) -> BaseGeometry:
"""Creates a shapely geometry using given query kwargs arguments
:param locations_config: (optional) EODAG locations configuration
:type locations_config: list
:param query_args: Query kwargs arguments from core.search() method
:type query_args: dict
:returns: shapely Geometry found
:rtype: :class:`shapely.geometry.BaseGeometry`
:raises: :class:`ValueError`
"""
geom = None
if "geometry" in query_args:
geom_arg = query_args["geometry"]
bbox_keys = ["lonmin", "latmin", "lonmax", "latmax"]
if isinstance(geom_arg, dict) and all(k in geom_arg for k in bbox_keys):
# bbox dict
geom = Polygon(
(
(geom_arg["lonmin"], geom_arg["latmin"]),
(geom_arg["lonmin"], geom_arg["latmax"]),
(geom_arg["lonmax"], geom_arg["latmax"]),
(geom_arg["lonmax"], geom_arg["latmin"]),
)
)
elif isinstance(geom_arg, (list, tuple)) and len(geom_arg) >= 4:
# bbox list
geom = Polygon(
(
(geom_arg[0], geom_arg[1]),
(geom_arg[0], geom_arg[3]),
(geom_arg[2], geom_arg[3]),
(geom_arg[2], geom_arg[1]),
)
)
elif isinstance(geom_arg, str):
# WKT geometry
geom = shapely.wkt.loads(geom_arg)
elif isinstance(geom_arg, BaseGeometry):
geom = geom_arg
elif geom_arg is None:
pass
else:
raise TypeError("Unexpected geometry type: {}".format(type(geom_arg)))
# look for location name in locations configuration
locations_dict = {loc["name"]: loc for loc in locations_config}
# The location query kwargs can either be in query_args or in query_args["locations"],
# support for which were added in 2.0.0 and 2.1.0 respectively.
# The location query kwargs in query_args is supported for backward compatibility,
# the recommended usage is that they are in query_args["locations"]
locations = query_args.get("locations")
locations = locations if locations is not None else {}
# In query_args["locations"] we can check that the location_names are correct
locations = locations if locations is not None else {}
for location_name in locations:
if location_name not in locations_dict:
raise ValueError(
f"The location name {location_name} is wrong. "
f"It must be one of: {locations_dict.keys()}"
)
query_locations = {**query_args, **locations}
for arg in query_locations.keys():
if arg in locations_dict.keys():
found = False
pattern = query_locations[arg]
attr = locations_dict[arg]["attr"]
with shapefile.Reader(locations_dict[arg]["path"]) as shp:
for shaperec in shp.shapeRecords():
if re.search(pattern, shaperec.record[attr]):
found = True
new_geom = shape(shaperec.shape)
# get geoms union
geom = new_geom.union(geom) if geom else new_geom
if not found:
raise ValueError(
f"No match found for the search location '{arg}' "
f"with the pattern '{pattern}'."
)
return geom
class MockResponse:
"""Fake requests response"""
def __init__(self, json_data: Any, status_code: int) -> None:
self.json_data = json_data
self.status_code = status_code
self.content = json_data
def json(self) -> Any:
"""Return json data"""
return self.json_data
def raise_for_status(self) -> None:
"""raises an exception when the status is not ok"""
if self.status_code != 200:
raise HTTPError()
def md5sum(file_path: str) -> str:
"""Get file MD5 checksum
>>> import os
>>> md5sum(os.devnull)
'd41d8cd98f00b204e9800998ecf8427e'
:param file_path: input file path
:type file_path: str
:returns: MD5 checksum
:rtype: str
"""
hash_md5 = hashlib.md5()
with open(file_path, "rb") as f:
for chunk in iter(lambda: f.read(4096), b""):
hash_md5.update(chunk)
return hash_md5.hexdigest()
def obj_md5sum(data: Any) -> str:
"""Get MD5 checksum from JSON serializable object
>>> obj_md5sum(None)
'37a6259cc0c1dae299a7866489dff0bd'
:param data: JSON serializable input object
:type data: Any
:returns: MD5 checksum
:rtype: str
"""
return hashlib.md5(orjson.dumps(data, option=orjson.OPT_SORT_KEYS)).hexdigest()
@functools.lru_cache()
def cached_parse(str_to_parse: str) -> JSONPath:
"""Cached jsonpath_ng.ext.parse
>>> cached_parse.cache_clear()
>>> cached_parse("$.foo")
Child(Root(), Fields('foo'))
>>> cached_parse.cache_info()
CacheInfo(hits=0, misses=1, maxsize=128, currsize=1)
>>> cached_parse("$.foo")
Child(Root(), Fields('foo'))
>>> cached_parse.cache_info()
CacheInfo(hits=1, misses=1, maxsize=128, currsize=1)
>>> cached_parse("$.bar")
Child(Root(), Fields('bar'))
>>> cached_parse.cache_info()
CacheInfo(hits=1, misses=2, maxsize=128, currsize=2)
:param str_to_parse: string to parse as jsonpath
:type str_to_parse: str
:returns: parsed jsonpath
:rtype: :class:`jsonpath_ng.JSONPath`
"""
return parse(str_to_parse)
@functools.lru_cache()
def _mutable_cached_yaml_load(config_path: str) -> Any:
with open(os.path.abspath(os.path.realpath(config_path)), "r") as fh:
return yaml.load(fh, Loader=yaml.SafeLoader)
def cached_yaml_load(config_path: str) -> Dict[str, Any]:
"""Cached yaml.load
:param config_path: path to the yaml configuration file
:type config_path: str
:returns: loaded yaml configuration
:rtype: dict
"""
return copy_deepcopy(_mutable_cached_yaml_load(config_path))
@functools.lru_cache()
def _mutable_cached_yaml_load_all(config_path: str) -> List[Any]:
with open(config_path, "r") as fh:
return list(yaml.load_all(fh, Loader=yaml.Loader))
def cached_yaml_load_all(config_path: str) -> List[Any]:
"""Cached yaml.load_all
Load all configurations stored in the configuration file as separated yaml documents
:param config_path: path to the yaml configuration file
:type config_path: str
:returns: list of configurations
:rtype: list
"""
return copy_deepcopy(_mutable_cached_yaml_load_all(config_path))
def get_bucket_name_and_prefix(
url: str, bucket_path_level: Optional[int] = None
) -> Tuple[Optional[str], Optional[str]]:
"""Extract bucket name and prefix from URL
:param url: (optional) URL to use as product.location
:type url: str
:param bucket_path_level: (optional) bucket location index in path.split('/')
:type bucket_path_level: int
:returns: bucket_name and prefix as str
:rtype: tuple
"""
bucket, prefix = None, None
scheme, netloc, path, _, _, _ = urlparse(url)
subdomain = netloc.split(".")[0]
path = path.strip("/")
if scheme and bucket_path_level is None:
bucket = subdomain
prefix = path
elif not scheme and bucket_path_level is None:
prefix = path
elif bucket_path_level is not None:
parts = path.split("/")
bucket, prefix = parts[bucket_path_level], "/".join(
parts[(bucket_path_level + 1) :]
)
return bucket, prefix
def flatten_top_directories(
nested_dir_root: str, common_subdirs_path: Optional[str] = None
) -> None:
"""Flatten directory structure, removing common empty sub-directories
:param nested_dir_root: Absolute path of the directory structure to flatten
:type nested_dir_root: str
:param common_subdirs_path: (optional) Absolute path of the desired subdirectory to remove
:type common_subdirs_path: str
"""
if not common_subdirs_path:
subpaths_list = [p for p in Path(nested_dir_root).glob("**/*") if p.is_file()]
common_subdirs_path = os.path.commonpath(subpaths_list)
if Path(common_subdirs_path).is_file():
common_subdirs_path = os.path.dirname(common_subdirs_path)
if nested_dir_root != common_subdirs_path:
logger.debug(f"Flatten {common_subdirs_path} to {nested_dir_root}")
tmp_path = mkdtemp()
shutil.copytree(common_subdirs_path, tmp_path, dirs_exist_ok=True)
shutil.rmtree(nested_dir_root)
shutil.move(tmp_path, nested_dir_root)
def deepcopy(sth: Any) -> Any:
"""Customized and faster deepcopy inspired by https://stackoverflow.com/a/45858907
`_copy_list` and `_copy_dict` available for the moment
:param sth: Object to copy
:type sth: Any
:returns: Copied object
:rtype: Any
"""
_dispatcher: Dict[Type[Any], Callable[..., Any]] = {}
def _copy_list(
input_list: List[Any], dispatch: Dict[Type[Any], Callable[..., Any]]
):
ret = input_list.copy()
for idx, item in enumerate(ret):
cp = dispatch.get(type(item))
if cp is not None:
ret[idx] = cp(item, dispatch)
return ret
def _copy_dict(
input_dict: Dict[Any, Any], dispatch: Dict[Type[Any], Callable[..., Any]]
):
ret = input_dict.copy()
for key, value in ret.items():
cp = dispatch.get(type(value))
if cp is not None:
ret[key] = cp(value, dispatch)
return ret
_dispatcher[list] = _copy_list
_dispatcher[dict] = _copy_dict
cp = _dispatcher.get(type(sth))
if cp is None:
return sth
else:
return cp(sth, _dispatcher)
def parse_header(header: str) -> Message:
"""Parse HTTP header
>>> parse_header(
... 'Content-Disposition: form-data; name="field2"; filename="example.txt"'
... ).get_param("filename")
'example.txt'
:param header: header to parse
:type header: str
:returns: parsed header
:rtype: :class:`~email.message.Message`
"""
m = Message()
m["content-type"] = header
return m
def cast_scalar_value(value: Any, new_type: Any) -> Any:
"""Convert a scalar (not nested) value type to the given one
>>> cast_scalar_value('1', int)
1
>>> cast_scalar_value(1, str)
'1'
>>> cast_scalar_value('false', bool)
False
:param value: the scalar value to convert
:param new_type: the wanted type
:returns: scalar value converted to new_type
"""
if isinstance(value, str) and new_type is bool:
# Bool is a type with special meaning in Python, thus the special
# case
if value.capitalize() not in ("True", "False"):
raise ValueError(
"Only true or false strings (case insensitive) are "
"allowed for booleans"
)
# Get the real Python value of the boolean. e.g: value='tRuE'
# => eval(value.capitalize())=True.
# str.capitalize() transforms the first character of the string
# to a capital letter
return eval(value.capitalize())
return new_type(value)
