#!/usr/bin/env python
u"""
utilities.py
Written by Tyler Sutterley (09/2024)
Download and management utilities for syncing time and auxiliary files
PYTHON DEPENDENCIES:
lxml: processing XML and HTML in Python
https://pypi.python.org/pypi/lxml
UPDATE HISTORY:
Updated 09/2024: add wrapper to importlib for optional dependencies
Updated 06/2022: add NASA Common Metadata Repository (CMR) queries
added function to build GES DISC subsetting API requests
Updated 04/2022: updated docstrings to numpy documentation format
Updated 10/2021: build python logging instance for handling verbose output
Updated 09/2021: added generic list from Apache http server
Updated 07/2021: add parser for converting file files to arguments
Updated 03/2021: added sha1 option for retrieving file hashes
Updated 01/2021: added username and password to ftp functions
added ftp connection check
Updated 12/2020: added file object keyword for downloads if verbose
add url split function for creating url location lists
Updated 11/2020: normalize source and destination paths in copy
make context an optional keyword argument in from_http
Updated 09/2020: copy from http and https to bytesIO object in chunks
Written 08/2020
"""
from __future__ import print_function, division
import sys
import os
import re
import io
import ssl
import json
import netrc
import ftplib
import shutil
import base64
import socket
import inspect
import hashlib
import logging
import importlib
import posixpath
import lxml.etree
import calendar, time
import dateutil.parser
if sys.version_info[0] == 2:
from cookielib import CookieJar
from urllib import urlencode
import urllib2
import urlparse
else:
from http.cookiejar import CookieJar
from urllib.parse import urlencode
import urllib.request as urllib2
import urllib.parse as urlparse
# PURPOSE: get absolute path within a package from a relative path
[docs]def get_data_path(relpath):
"""
Get the absolute path within a package from a relative path
Parameters
----------
relpath: str,
relative path
"""
# current file path
filename = inspect.getframeinfo(inspect.currentframe()).filename
filepath = os.path.dirname(os.path.abspath(filename))
if isinstance(relpath,list):
# use *splat operator to extract from list
return os.path.join(filepath,*relpath)
elif isinstance(relpath,str):
return os.path.join(filepath,relpath)
[docs]def import_dependency(
name: str,
extra: str = "",
raise_exception: bool = False
):
"""
Import an optional dependency
Adapted from ``pandas.compat._optional::import_optional_dependency``
Parameters
----------
name: str
Module name
extra: str, default ""
Additional text to include in the ``ImportError`` message
raise_exception: bool, default False
Raise an ``ImportError`` if the module is not found
Returns
-------
module: obj
Imported module
"""
# check if the module name is a string
msg = f"Invalid module name: '{name}'; must be a string"
assert isinstance(name, str), msg
# default error if module cannot be imported
err = f"Missing optional dependency '{name}'. {extra}"
module = type('module', (), {})
# try to import the module
try:
module = importlib.import_module(name)
except (ImportError, ModuleNotFoundError) as exc:
if raise_exception:
raise ImportError(err) from exc
else:
logging.debug(err)
# return the module
return module
# PURPOSE: get the hash value of a file
[docs]def get_hash(local, algorithm='MD5'):
"""
Get the hash value from a local file or ``BytesIO`` object
Parameters
----------
local: obj or str
BytesIO object or path to file
algorithm: str, default 'MD5'
hashing algorithm for checksum validation
- ``'MD5'``: Message Digest
- ``'sha1'``: Secure Hash Algorithm
"""
# check if open file object or if local file exists
if isinstance(local, io.IOBase):
if (algorithm == 'MD5'):
return hashlib.md5(local.getvalue()).hexdigest()
elif (algorithm == 'sha1'):
return hashlib.sha1(local.getvalue()).hexdigest()
elif os.access(os.path.expanduser(local),os.F_OK):
# generate checksum hash for local file
# open the local_file in binary read mode
with open(os.path.expanduser(local), 'rb') as local_buffer:
# generate checksum hash for a given type
if (algorithm == 'MD5'):
return hashlib.md5(local_buffer.read()).hexdigest()
elif (algorithm == 'sha1'):
return hashlib.sha1(local_buffer.read()).hexdigest()
else:
return ''
# PURPOSE: recursively split a url path
[docs]def url_split(s):
"""
Recursively split a url path into a list
Parameters
----------
s: str
url string
"""
head, tail = posixpath.split(s)
if head in ('http:','https:','ftp:','s3:'):
return s,
elif head in ('', posixpath.sep):
return tail,
return url_split(head) + (tail,)
# PURPOSE: convert file lines to arguments
def convert_arg_line_to_args(arg_line):
"""
Convert file lines to arguments
Parameters
----------
arg_line: str
line string containing a single argument and/or comments
"""
# remove commented lines and after argument comments
for arg in re.sub(r'\#(.*?)$',r'',arg_line).split():
if not arg.strip():
continue
yield arg
# PURPOSE: returns the Unix timestamp value for a formatted date string
[docs]def get_unix_time(time_string, format='%Y-%m-%d %H:%M:%S'):
"""
Get the Unix timestamp value for a formatted date string
Parameters
----------
time_string: str
formatted time string to parse
format: str, default '%Y-%m-%d %H:%M:%S'
format for input time string
"""
try:
parsed_time = time.strptime(time_string.rstrip(), format)
except (TypeError, ValueError):
pass
else:
return calendar.timegm(parsed_time)
# try parsing with dateutil
try:
parsed_time = dateutil.parser.parse(time_string.rstrip())
except (TypeError, ValueError):
return None
else:
return parsed_time.timestamp()
# PURPOSE: output a time string in isoformat
# PURPOSE: rounds a number to an even number less than or equal to original
[docs]def even(value):
"""
Rounds a number to an even number less than or equal to original
Parameters
----------
value: float
number to be rounded
"""
return 2*int(value//2)
# PURPOSE: rounds a number upward to its nearest integer
[docs]def ceil(value):
"""
Rounds a number upward to its nearest integer
Parameters
----------
value: float
number to be rounded upward
"""
return -int(-value//1)
# PURPOSE: make a copy of a file with all system information
[docs]def copy(source, destination, move=False, **kwargs):
"""
Copy or move a file with all system information
Parameters
----------
source: str
source file
destination: str
copied destination file
move: bool, default False
remove the source file
"""
source = os.path.abspath(os.path.expanduser(source))
destination = os.path.abspath(os.path.expanduser(destination))
# log source and destination
logging.info('{0} -->\n\t{1}'.format(source,destination))
shutil.copyfile(source, destination)
shutil.copystat(source, destination)
if move:
os.remove(source)
# PURPOSE: check ftp connection
[docs]def check_ftp_connection(HOST, username=None, password=None):
"""
Check internet connection with ftp host
Parameters
----------
HOST: str
remote ftp host
username: str or NoneType
ftp username
password: str or NoneType
ftp password
"""
# attempt to connect to ftp host
try:
f = ftplib.FTP(HOST)
f.login(username, password)
f.voidcmd("NOOP")
except IOError:
raise RuntimeError('Check internet connection')
except ftplib.error_perm:
raise RuntimeError('Check login credentials')
else:
return True
# PURPOSE: list a directory on a ftp host
[docs]def ftp_list(HOST, username=None, password=None, timeout=None,
basename=False, pattern=None, sort=False):
"""
List a directory on a ftp host
Parameters
----------
HOST: str or list
remote ftp host path split as list
username: str or NoneType
ftp username
password: str or NoneType
ftp password
timeout: int or NoneType, default None
timeout in seconds for blocking operations
basename: bool, default False
return the file or directory basename instead of the full path
pattern: str or NoneType, default None
regular expression pattern for reducing list
sort: bool, default False
sort output list
Returns
-------
output: list
items in a directory
mtimes: list
last modification times for items in the directory
"""
# verify inputs for remote ftp host
if isinstance(HOST, str):
HOST = url_split(HOST)
# try to connect to ftp host
try:
ftp = ftplib.FTP(HOST[0],timeout=timeout)
except (socket.gaierror,IOError):
raise RuntimeError('Unable to connect to {0}'.format(HOST[0]))
else:
ftp.login(username,password)
# list remote path
output = ftp.nlst(posixpath.join(*HOST[1:]))
# get last modified date of ftp files and convert into unix time
mtimes = [None]*len(output)
# iterate over each file in the list and get the modification time
for i,f in enumerate(output):
try:
# try sending modification time command
mdtm = ftp.sendcmd('MDTM {0}'.format(f))
except ftplib.error_perm:
# directories will return with an error
pass
else:
# convert the modification time into unix time
mtimes[i] = get_unix_time(mdtm[4:], format="%Y%m%d%H%M%S")
# reduce to basenames
if basename:
output = [posixpath.basename(i) for i in output]
# reduce using regular expression pattern
if pattern:
i = [i for i,f in enumerate(output) if re.search(pattern,f)]
# reduce list of listed items and last modified times
output = [output[indice] for indice in i]
mtimes = [mtimes[indice] for indice in i]
# sort the list
if sort:
i = [i for i,j in sorted(enumerate(output), key=lambda i: i[1])]
# sort list of listed items and last modified times
output = [output[indice] for indice in i]
mtimes = [mtimes[indice] for indice in i]
# close the ftp connection
ftp.close()
# return the list of items and last modified times
return (output, mtimes)
# PURPOSE: download a file from a ftp host
[docs]def from_ftp(HOST, username=None, password=None, timeout=None,
local=None, hash='', chunk=8192, verbose=False, fid=sys.stdout,
mode=0o775):
"""
Download a file from a ftp host
Parameters
----------
HOST: str or list
remote ftp host path
username: str or NoneType
ftp username
password: str or NoneType
ftp password
timeout: int or NoneType, default None
timeout in seconds for blocking operations
local: str or NoneType, default None
path to local file
hash: str, default ''
MD5 hash of local file
chunk: int, default 8192
chunk size for transfer encoding
verbose: bool, default False
print file transfer information
fid: obj, default sys.stdout
open file object to print if verbose
mode: oct, default 0o775
permissions mode of output local file
Returns
-------
remote_buffer: obj
BytesIO representation of file
"""
# create logger
loglevel = logging.INFO if verbose else logging.CRITICAL
logging.basicConfig(stream=fid, level=loglevel)
# verify inputs for remote ftp host
if isinstance(HOST, str):
HOST = url_split(HOST)
# try downloading from ftp
try:
# try to connect to ftp host
ftp = ftplib.FTP(HOST[0], timeout=timeout)
except (socket.gaierror,IOError):
raise RuntimeError('Unable to connect to {0}'.format(HOST[0]))
else:
ftp.login(username,password)
# remote path
ftp_remote_path = posixpath.join(*HOST[1:])
# copy remote file contents to bytesIO object
remote_buffer = io.BytesIO()
ftp.retrbinary('RETR {0}'.format(ftp_remote_path),
remote_buffer.write, blocksize=chunk)
remote_buffer.seek(0)
# save file basename with bytesIO object
remote_buffer.filename = HOST[-1]
# generate checksum hash for remote file
remote_hash = hashlib.md5(remote_buffer.getvalue()).hexdigest()
# get last modified date of remote file and convert into unix time
mdtm = ftp.sendcmd('MDTM {0}'.format(ftp_remote_path))
remote_mtime = get_unix_time(mdtm[4:], format="%Y%m%d%H%M%S")
# compare checksums
if local and (hash != remote_hash):
# convert to absolute path
local = os.path.abspath(local)
# create directory if non-existent
if not os.access(os.path.dirname(local), os.F_OK):
os.makedirs(os.path.dirname(local), mode)
# print file information
args = (posixpath.join(*HOST),local)
logging.info('{0} -->\n\t{1}'.format(*args))
# store bytes to file using chunked transfer encoding
remote_buffer.seek(0)
with open(os.path.expanduser(local), 'wb') as f:
shutil.copyfileobj(remote_buffer, f, chunk)
# change the permissions mode
os.chmod(local,mode)
# keep remote modification time of file and local access time
os.utime(local, (os.stat(local).st_atime, remote_mtime))
# close the ftp connection
ftp.close()
# return the bytesIO object
remote_buffer.seek(0)
return remote_buffer
def _create_default_ssl_context() -> ssl.SSLContext:
"""Creates the default SSL context
"""
context = ssl.SSLContext(ssl.PROTOCOL_TLS_CLIENT)
_set_ssl_context_options(context)
context.options |= ssl.OP_NO_COMPRESSION
return context
def _create_ssl_context_no_verify() -> ssl.SSLContext:
"""Creates an SSL context for unverified connections
"""
context = _create_default_ssl_context()
context.check_hostname = False
context.verify_mode = ssl.CERT_NONE
return context
def _set_ssl_context_options(context: ssl.SSLContext) -> None:
"""Sets the default options for the SSL context
"""
if sys.version_info >= (3, 10) or ssl.OPENSSL_VERSION_INFO >= (1, 1, 0, 7):
context.minimum_version = ssl.TLSVersion.TLSv1_2
else:
context.options |= ssl.OP_NO_SSLv2
context.options |= ssl.OP_NO_SSLv3
context.options |= ssl.OP_NO_TLSv1
context.options |= ssl.OP_NO_TLSv1_1
# default ssl context
_default_ssl_context = _create_ssl_context_no_verify()
# PURPOSE: check internet connection
def check_connection(HOST, context=_default_ssl_context):
"""
Check internet connection with http host
Parameters
----------
HOST: str
remote http host
context: obj, default ssl.SSLContext(ssl.PROTOCOL_TLS)
SSL context for ``urllib`` opener object
"""
# attempt to connect to http host
try:
urllib2.urlopen(HOST, timeout=20, context=context)
except urllib2.URLError:
raise RuntimeError('Check internet connection')
else:
return True
# PURPOSE: list a directory on an Apache http Server
[docs]def http_list(HOST, timeout=None, context=_default_ssl_context,
parser=lxml.etree.HTMLParser(), format='%Y-%m-%d %H:%M',
pattern='', sort=False):
"""
List a directory on an Apache http Server
Parameters
----------
HOST: str or list
remote http host path
timeout: int or NoneType, default None
timeout in seconds for blocking operations
context: obj, default ssl.SSLContext(ssl.PROTOCOL_TLS)
SSL context for ``urllib`` opener object
parser: obj, default lxml.etree.HTMLParser()
HTML parser for ``lxml``
format: str, default '%Y-%m-%d %H:%M'
format for input time string
pattern: str, default ''
regular expression pattern for reducing list
sort: bool, default False
sort output list
Returns
-------
colnames: list
column names in a directory
collastmod: list
last modification times for items in the directory
"""
# verify inputs for remote http host
if isinstance(HOST, str):
HOST = url_split(HOST)
# try listing from http
try:
# Create and submit request.
request = urllib2.Request(posixpath.join(*HOST))
response = urllib2.urlopen(request, timeout=timeout, context=context)
except (urllib2.HTTPError, urllib2.URLError):
raise Exception('List error from {0}'.format(posixpath.join(*HOST)))
else:
# read and parse request for files (column names and modified times)
tree = lxml.etree.parse(response, parser)
colnames = tree.xpath('//tr/td[not(@*)]//a/@href')
# get the Unix timestamp value for a modification time
collastmod = [get_unix_time(i,format=format)
for i in tree.xpath('//tr/td[@align="right"][1]/text()')]
# reduce using regular expression pattern
if pattern:
i = [i for i,f in enumerate(colnames) if re.search(pattern, f)]
# reduce list of column names and last modified times
colnames = [colnames[indice] for indice in i]
collastmod = [collastmod[indice] for indice in i]
# sort the list
if sort:
i = [i for i,j in sorted(enumerate(colnames), key=lambda i: i[1])]
# sort list of column names and last modified times
colnames = [colnames[indice] for indice in i]
collastmod = [collastmod[indice] for indice in i]
# return the list of column names and last modified times
return (colnames, collastmod)
# PURPOSE: download a file from a http host
[docs]def from_http(HOST, timeout=None, context=_default_ssl_context,
local=None, hash='', chunk=16384, verbose=False, fid=sys.stdout,
mode=0o775):
"""
Download a file from a http host
Parameters
----------
HOST: str or list
remote http host path split as list
timeout: int or NoneType, default None
timeout in seconds for blocking operations
context: obj, default ssl.SSLContext(ssl.PROTOCOL_TLS)
SSL context for ``urllib`` opener object
timeout: int or NoneType, default None
timeout in seconds for blocking operations
local: str or NoneType, default None
path to local file
hash: str, default ''
MD5 hash of local file
chunk: int, default 16384
chunk size for transfer encoding
verbose: bool, default False
print file transfer information
fid: obj, default sys.stdout
open file object to print if verbose
mode: oct, default 0o775
permissions mode of output local file
Returns
-------
remote_buffer: obj
BytesIO representation of file
"""
# create logger
loglevel = logging.INFO if verbose else logging.CRITICAL
logging.basicConfig(stream=fid, level=loglevel)
# verify inputs for remote http host
if isinstance(HOST, str):
HOST = url_split(HOST)
# try downloading from http
try:
# Create and submit request.
request = urllib2.Request(posixpath.join(*HOST))
response = urllib2.urlopen(request, timeout=timeout, context=context)
except:
raise Exception('Download error from {0}'.format(posixpath.join(*HOST)))
else:
# copy remote file contents to bytesIO object
remote_buffer = io.BytesIO()
shutil.copyfileobj(response, remote_buffer, chunk)
remote_buffer.seek(0)
# save file basename with bytesIO object
remote_buffer.filename = HOST[-1]
# generate checksum hash for remote file
remote_hash = hashlib.md5(remote_buffer.getvalue()).hexdigest()
# compare checksums
if local and (hash != remote_hash):
# convert to absolute path
local = os.path.abspath(local)
# create directory if non-existent
if not os.access(os.path.dirname(local), os.F_OK):
os.makedirs(os.path.dirname(local), mode)
# print file information
args = (posixpath.join(*HOST),local)
logging.info('{0} -->\n\t{1}'.format(*args))
# store bytes to file using chunked transfer encoding
remote_buffer.seek(0)
with open(os.path.expanduser(local), 'wb') as f:
shutil.copyfileobj(remote_buffer, f, chunk)
# change the permissions mode
os.chmod(local,mode)
# return the bytesIO object
remote_buffer.seek(0)
return remote_buffer
# PURPOSE: "login" to NASA Earthdata with supplied credentials
[docs]def build_opener(username, password, context=_default_ssl_context,
password_manager=False, get_ca_certs=False, redirect=False,
authorization_header=True, urs='https://urs.earthdata.nasa.gov'):
"""
Build ``urllib`` opener for NASA Earthdata with supplied credentials
Parameters
----------
username: str or NoneType, default None
NASA Earthdata username
password: str or NoneType, default None
NASA Earthdata password
context: obj, default ssl.SSLContext(ssl.PROTOCOL_TLS)
SSL context for ``urllib`` opener object
password_manager: bool, default False
Create password manager context using default realm
get_ca_certs: bool, default False
Get list of loaded “certification authority” certificates
redirect: bool, default False
Create redirect handler object
authorization_header: bool, default True
Add base64 encoded authorization header to opener
urs: str, default 'https://urs.earthdata.nasa.gov'
Earthdata login URS 3 host
"""
# https://docs.python.org/3/howto/urllib2.html#id5
handler = []
# create a password manager
if password_manager:
password_mgr = urllib2.HTTPPasswordMgrWithDefaultRealm()
# Add the username and password for NASA Earthdata Login system
password_mgr.add_password(None, urs, username, password)
handler.append(urllib2.HTTPBasicAuthHandler(password_mgr))
# Create cookie jar for storing cookies. This is used to store and return
# the session cookie given to use by the data server (otherwise will just
# keep sending us back to Earthdata Login to authenticate).
cookie_jar = CookieJar()
handler.append(urllib2.HTTPCookieProcessor(cookie_jar))
# SSL context handler
if get_ca_certs:
context.get_ca_certs()
handler.append(urllib2.HTTPSHandler(context=context))
# redirect handler
if redirect:
handler.append(urllib2.HTTPRedirectHandler())
# create "opener" (OpenerDirector instance)
opener = urllib2.build_opener(*handler)
# Encode username/password for request authorization headers
# add Authorization header to opener
if authorization_header:
b64 = base64.b64encode('{0}:{1}'.format(username, password).encode())
opener.addheaders = [("Authorization","Basic {0}".format(b64.decode()))]
# Now all calls to urllib2.urlopen use our opener.
urllib2.install_opener(opener)
# All calls to urllib2.urlopen will now use handler
# Make sure not to include the protocol in with the URL, or
# HTTPPasswordMgrWithDefaultRealm will be confused.
return opener
# PURPOSE: list a directory on NASA GES DISC https server
[docs]def gesdisc_list(HOST, username=None, password=None, build=False,
timeout=None, urs='urs.earthdata.nasa.gov',
parser=lxml.etree.HTMLParser(), format='%Y-%m-%d %H:%M',
pattern='', sort=False):
"""
List a directory on NASA GES DISC servers
Parameters
----------
HOST: str or list
remote https host
username: str or NoneType, default None
NASA Earthdata username
password: str or NoneType, default None
NASA Earthdata password
build: bool, default True
Build opener with NASA Earthdata credentials
timeout: int or NoneType, default None
timeout in seconds for blocking operations
urs: str, default 'urs.earthdata.nasa.gov'
Earthdata login URS 3 host
parser: obj, default lxml.etree.HTMLParser()
HTML parser for ``lxml``
format: str, default '%Y-%m-%d %H:%M'
format for input time string
pattern: str, default ''
regular expression pattern for reducing list
sort: bool, default False
sort output list
Returns
-------
colnames: list
column names in a directory
collastmod: list
last modification times for items in the directory
"""
# use netrc credentials
if build and not (username or password):
username,_,password = netrc.netrc().authenticators(urs)
# build urllib2 opener with credentials
if build:
build_opener(username, password, password_manager=True,
authorization_header=False)
# verify inputs for remote https host
if isinstance(HOST, str):
HOST = url_split(HOST)
# try listing from https
try:
# Create and submit request.
request=urllib2.Request(posixpath.join(*HOST))
response=urllib2.urlopen(request, timeout=timeout)
except (urllib2.HTTPError, urllib2.URLError):
raise Exception('List error from {0}'.format(posixpath.join(*HOST)))
else:
# read and parse request for files (column names and modified times)
tree = lxml.etree.parse(response,parser)
colnames = tree.xpath('//tr/td[not(@*)]//a/@href')
# get the Unix timestamp value for a modification time
lastmod = [get_unix_time(i,format=format)
for i in tree.xpath('//tr/td[@align="right"][1]/text()')]
# reduce using regular expression pattern
if pattern:
i = [i for i,f in enumerate(colnames) if re.search(pattern,f)]
# reduce list of column names and last modified times
colnames = [colnames[indice] for indice in i]
lastmod = [lastmod[indice] for indice in i]
# sort the list
if sort:
i = [i for i,j in sorted(enumerate(colnames), key=lambda i: i[1])]
# sort list of column names and last modified times
colnames = [colnames[indice] for indice in i]
lastmod = [lastmod[indice] for indice in i]
# return the list of column names and last modified times
return (colnames,lastmod)
# PURPOSE: filter the CMR json response for desired data files
[docs]def cmr_filter_json(search_results, endpoint="data",
request_type="application/x-netcdf"):
"""
Filter the CMR json response for desired data files
Parameters
----------
search_results: dict
json response from CMR query
endpoint: str, default 'data'
url endpoint type
- ``'data'``: NASA Earthdata https archive
- ``'opendap'``: NASA Earthdata OPeNDAP archive
- ``'s3'``: NASA Earthdata Cumulus AWS S3 bucket
request_type: str, default 'application/x-netcdf'
data type for reducing CMR query
Returns
-------
granule_names: list
Model granule names
granule_urls: list
Model granule urls
granule_mtimes: list
Model granule modification times
"""
# output list of granule ids, urls and modified times
granule_names = []
granule_urls = []
granule_mtimes = []
# check that there are urls for request
if ('feed' not in search_results) or ('entry' not in search_results['feed']):
return (granule_names,granule_urls)
# descriptor links for each endpoint
rel = {}
rel['data'] = "http://esipfed.org/ns/fedsearch/1.1/data#"
rel['opendap'] = "http://esipfed.org/ns/fedsearch/1.1/service#"
rel['s3'] = "http://esipfed.org/ns/fedsearch/1.1/s3#"
# iterate over references and get cmr location
for entry in search_results['feed']['entry']:
granule_names.append(entry['producer_granule_id'])
granule_mtimes.append(get_unix_time(entry['updated'],
format='%Y-%m-%dT%H:%M:%S.%f%z'))
for link in entry['links']:
# skip inherited granules
if ('inherited' in link.keys()):
continue
# append if selected endpoint
if (link['rel'] == rel[endpoint]):
granule_urls.append(link['href'])
break
# alternatively append if selected data type
if ('type' not in link.keys()):
continue
if (link['type'] == request_type):
granule_urls.append(link['href'])
break
# return the list of urls, granule ids and modified times
return (granule_names, granule_urls, granule_mtimes)
# PURPOSE: cmr queries for GRACE/GRACE-FO products
[docs]def cmr(short_name, version=None, start_date=None, end_date=None,
provider='GES_DISC', endpoint='data', request_type='application/x-netcdf',
verbose=False, fid=sys.stdout):
"""
Query the NASA Common Metadata Repository (CMR) for model data
Parameters
----------
short_name: str
Model shortname in the CMR system
version: str or NoneType, default None
Model version
start_date: str or NoneType, default None
starting date for CMR product query
end_date: str or NoneType, default None
ending date for CMR product query
provider: str, default 'GES_DISC'
CMR data provider
- ``'GES_DISC'``: GESDISC
- ``'GESDISCCLD'``: GESDISC Cumulus
- ``'PODAAC'``: PO.DAAC Drive
- ``'POCLOUD'``: PO.DAAC Cumulus
endpoint: str, default 'data'
url endpoint type
- ``'data'``: NASA Earthdata https archive
- ``'opendap'``: NASA Earthdata OPeNDAP archive
- ``'s3'``: NASA Earthdata Cumulus AWS S3 bucket
request_type: str, default 'application/x-netcdf'
data type for reducing CMR query
verbose: bool, default False
print CMR query information
fid: obj, default sys.stdout
open file object to print if verbose
Returns
-------
granule_names: list
Model granule names
granule_urls: list
Model granule urls
granule_mtimes: list
Model granule modification times
"""
# create logger
loglevel = logging.INFO if verbose else logging.CRITICAL
logging.basicConfig(stream=fid, level=loglevel)
# build urllib2 opener with SSL context
# https://docs.python.org/3/howto/urllib2.html#id5
handler = []
# Create cookie jar for storing cookies
cookie_jar = CookieJar()
handler.append(urllib2.HTTPCookieProcessor(cookie_jar))
handler.append(urllib2.HTTPSHandler(context=ssl.SSLContext()))
# create "opener" (OpenerDirector instance)
opener = urllib2.build_opener(*handler)
# build CMR query
cmr_format = 'json'
cmr_page_size = 2000
CMR_HOST = ['https://cmr.earthdata.nasa.gov','search',
'granules.{0}'.format(cmr_format)]
# build list of CMR query parameters
CMR_KEYS = []
CMR_KEYS.append('?provider={0}'.format(provider))
CMR_KEYS.append('&sort_key[]=start_date')
CMR_KEYS.append('&sort_key[]=producer_granule_id')
CMR_KEYS.append('&scroll=true')
CMR_KEYS.append('&page_size={0}'.format(cmr_page_size))
# dictionary of product shortnames and version
CMR_KEYS.append('&short_name={0}'.format(short_name))
if version:
CMR_KEYS.append('&version={0}'.format(version))
# append keys for start and end time
# verify that start and end times are in ISO format
start_date = isoformat(start_date) if start_date else ''
end_date = isoformat(end_date) if end_date else ''
CMR_KEYS.append('&temporal={0},{1}'.format(start_date, end_date))
# full CMR query url
cmr_query_url = "".join([posixpath.join(*CMR_HOST),*CMR_KEYS])
logging.info('CMR request={0}'.format(cmr_query_url))
# output list of granule names and urls
granule_names = []
granule_urls = []
granule_mtimes = []
cmr_scroll_id = None
while True:
req = urllib2.Request(cmr_query_url)
if cmr_scroll_id:
req.add_header('cmr-scroll-id', cmr_scroll_id)
response = opener.open(req)
# get scroll id for next iteration
if not cmr_scroll_id:
headers = {k.lower():v for k,v in dict(response.info()).items()}
cmr_scroll_id = headers['cmr-scroll-id']
# read the CMR search as JSON
search_page = json.loads(response.read().decode('utf8'))
ids,urls,mtimes = cmr_filter_json(search_page,
endpoint=endpoint, request_type=request_type)
if not urls:
break
# extend lists
granule_names.extend(ids)
granule_urls.extend(urls)
granule_mtimes.extend(mtimes)
# return the list of granule ids, urls and modification times
return (granule_names, granule_urls, granule_mtimes)
# PURPOSE: build requests for the GES DISC subsetting API
[docs]def build_request(short_name, dataset_version, url, variables=[],
format='bmM0Lw', service='L34RS_MERRA2', version='1.02',
bbox=[-90,-180,90,180], **kwargs):
"""
Build requests for the GES DISC subsetting API
Parameters
----------
short_name: str
Model shortname in the CMR system
url: str
url for granule returned by the CMR system
variables: list, default []
Variables for product to subset
format: str, default 'bmM0Lw'
Coded output format for GES DISC subsetting API
service: str, default 'L34RS_MERRA2'
GES DISC subsetting API service
version: str, default '1.02'
GES DISC subsetting API service version
bbox: list, default [-90,-180,90,180]
Bounding box to spatially subset
**kwargs: dict, default {}
Additional parameters for GES DISC subsetting API
Returns
-------
request_url: str
Formatted url for GES DISC subsetting API
"""
# split CMR supplied url for granule
HOST,*args = url_split(url)
api_host = posixpath.join(HOST,'daac-bin','OTF','HTTP_services.cgi?')
# create parameters to be encoded
kwargs['FILENAME'] = posixpath.join(posixpath.sep, *args)
kwargs['FORMAT'] = format
kwargs['SERVICE'] = service
kwargs['VERSION'] = version
kwargs['BBOX'] = ','.join(map(str, bbox))
kwargs['SHORTNAME'] = short_name
kwargs['DATASET_VERSION'] = dataset_version
kwargs['VARIABLES'] = ','.join(variables)
# return the formatted request url
request_url = api_host + urlencode(kwargs)
return request_url