def fetch_check_data(data_file,
url,
hashes,
cache_directory='',
append='',
verbose=False):
"""
Get data file through a URL call and check its hash:
1. Check hash table for data file name.
2. Check hash subdirectory within cache directory for data file.
3. If data file not in cache, download, compute hash, and verify hash.
4. If hash correct, save file (+ append); otherwise, raise an error.
Parameters
----------
data_file : string
name of file (not the full path)
url : string
URL for data file
hashes : dictionary
file names and md5 hashes (if empty, simply download file from url)
cache_directory : string
cache directory (full path)
append : string
append to output file (ex: '.nii.gz')
verbose : bool
print statements?
Returns
-------
data_path : string
data file name (full path)
Examples
--------
>>> from mindboggle.mio.fetch_data import fetch_check_data
>>> from mindboggle.mio.fetch_data import cache_hashes
>>> # osf.io URL for OASIS-30_Atropos_template_to_MNI152_affine.txt
>>> data_file = 'OASIS-30_Atropos_template_to_MNI152_affine.txt'
>>> url = 'https://osf.io/ufydw/?action=download&version=1'
>>> hashes = cache_hashes()
>>> cache_directory = ''
>>> append = ''
>>> verbose = False
>>> data_path = fetch_check_data(data_file, url, hashes, cache_directory,
... append, verbose) # doctest: +SKIP
"""
import os
import shutil
from mindboggle.mio.fetch_data import fetch_data, fetch_hash
# ------------------------------------------------------------------------
# Set temporary cache directory if not specified:
# ------------------------------------------------------------------------
if not cache_directory:
cache_directory = os.path.join(os.environ['HOME'], 'hash_temp')
# ------------------------------------------------------------------------
# Check hash table for file name, and store corresponding hash:
# ------------------------------------------------------------------------
if hashes and data_file in list(hashes):
stored_hash = hashes[data_file]
# --------------------------------------------------------------------
# Create missing cache and hash directories:
# --------------------------------------------------------------------
if not os.path.exists(cache_directory):
if verbose:
print("Create missing cache directory: {0}".format(
cache_directory))
os.mkdir(cache_directory)
hash_dir = os.path.join(cache_directory, stored_hash)
if not os.path.exists(hash_dir):
if verbose:
print("Create missing hash directory: {0}".format(hash_dir))
os.mkdir(os.path.join(hash_dir))
# --------------------------------------------------------------------
# Check hash subdirectory for file:
# --------------------------------------------------------------------
data_path = os.path.join(hash_dir, data_file)
if os.path.exists(data_path):
if verbose:
print("File already exists and matches hash: {0}".format(url))
return data_path
# --------------------------------------------------------------------
# If file not in cache, download, compute hash, and verify:
# --------------------------------------------------------------------
else:
if verbose:
print("Retrieve file from URL: {0}".format(url))
# Download file as a temporary file:
temp_file = fetch_data(url)
# Compute the file's hash:
data_hash = fetch_hash(temp_file)
# If hash matches name of the hash directory, save file:
if os.path.join(cache_directory, data_hash) == hash_dir:
# Add append:
if append:
data_path += append
if verbose:
print("Copy file to cache: {0}".format(data_path))
shutil.copyfile(temp_file, data_path)
return data_path
else:
raise IOError("Retrieved hash does not match stored hash.")
else:
raise IOError("Data file '{0}' not in hash table.".format(data_file))
def fetch_check_data(data_file, url, hashes, cache_directory='', append='',
verbose=False):
"""
Get data file through a URL call and check its hash:
1. Check hash table for data file name.
2. Check hash subdirectory within cache directory for data file.
3. If data file not in cache, download, compute hash, and verify hash.
4. If hash correct, save file (+ append); otherwise, raise an error.
Parameters
----------
data_file : string
name of file (not the full path)
url : string
URL for data file
hashes : dictionary
file names and md5 hashes (if empty, simply download file from url)
cache_directory : string
cache directory (full path)
append : string
append to output file (ex: '.nii.gz')
verbose : bool
print statements?
Returns
-------
data_path : string
data file name (full path)
Examples
--------
>>> from mindboggle.mio.fetch_data import fetch_check_data
>>> from mindboggle.mio.fetch_data import cache_hashes
>>> # osf.io URL for OASIS-30_Atropos_template_to_MNI152_affine.txt
>>> data_file = 'OASIS-30_Atropos_template_to_MNI152_affine.txt'
>>> url = 'https://osf.io/ufydw/?action=download&version=1'
>>> hashes = cache_hashes()
>>> cache_directory = ''
>>> append = ''
>>> verbose = False
>>> data_path = fetch_check_data(data_file, url, hashes, cache_directory,
... append, verbose) # doctest: +SKIP
"""
import os
import shutil
from mindboggle.mio.fetch_data import fetch_data, fetch_hash
# ------------------------------------------------------------------------
# Set temporary cache directory if not specified:
# ------------------------------------------------------------------------
if not cache_directory:
cache_directory = os.path.join(os.environ['HOME'], 'hash_temp')
# ------------------------------------------------------------------------
# Check hash table for file name, and store corresponding hash:
# ------------------------------------------------------------------------
if hashes and data_file in list(hashes):
stored_hash = hashes[data_file]
# --------------------------------------------------------------------
# Create missing cache and hash directories:
# --------------------------------------------------------------------
if not os.path.exists(cache_directory):
if verbose:
print("Create missing cache directory: {0}".
format(cache_directory))
os.mkdir(cache_directory)
hash_dir = os.path.join(cache_directory, stored_hash)
if not os.path.exists(hash_dir):
if verbose:
print("Create missing hash directory: {0}".format(hash_dir))
os.mkdir(os.path.join(hash_dir))
# --------------------------------------------------------------------
# Check hash subdirectory for file:
# --------------------------------------------------------------------
data_path = os.path.join(hash_dir, data_file)
if os.path.exists(data_path):
if verbose:
print("File already exists and matches hash: {0}".format(url))
return data_path
# --------------------------------------------------------------------
# If file not in cache, download, compute hash, and verify:
# --------------------------------------------------------------------
else:
if verbose:
print("Retrieve file from URL: {0}".format(url))
# Download file as a temporary file:
temp_file = fetch_data(url)
# Compute the file's hash:
data_hash = fetch_hash(temp_file)
# If hash matches name of the hash directory, save file:
if os.path.join(cache_directory, data_hash) == hash_dir:
# Add append:
if append:
data_path += append
if verbose:
print("Copy file to cache: {0}".format(data_path))
shutil.copyfile(temp_file, data_path)
return data_path
else:
raise IOError("Retrieved hash does not match stored hash.")
else:
raise IOError("Data file '{0}' not in hash table.".
format(data_file))
def fetch_check_data(data_file, url='', hashes={}, cache_env='', cache='',
return_missing=False, lookup=True):
"""
Get data file through a URL call and check its hash.
Steps ::
If hashes provided:
1. Check hash table for data file.
2. Check hash subdirectory within cache directory for data file.
3. If data file not in cache, download file, compute hash,
and verify hash.
4. If hash correct, save file.
Otherwise, simply download file or return file path as a string.
Parameters
----------
data_file : string
data file name
url : string
URL for data file
hashes : dictionary
file names and md5 hashes (if empty, simply download file from url)
cache_env : string
environment variable name for cache path
cache : string
in case cache_env is not set, use as cache directory
return_missing : Boolean
if data_file not in hash, simply download data_file and return path
lookup : Boolean
Simply return data_file path
Returns
-------
data_path : string
data file name (full path)
Examples
--------
>>> import os
>>> from mindboggle.mio.fetch_data import hashes_url
>>> from mindboggle.mio.fetch_data import fetch_check_data
>>> hashes, url, cache_env, cache = hashes_url()
>>> data_file = hashes.keys()[0]
>>> fetch_check_data(data_file, url, hashes, cache_env, cache)
"""
import os
import sys
import shutil
from mindboggle.mio.fetch_data import fetch_data, fetch_hash
if lookup:
#---------------------------------------------------------------------
# If hashes provided, go through steps to check/download file:
#---------------------------------------------------------------------
if hashes:
if not cache_env:
cache_env = 'MINDBOGGLE_CACHE'
if not cache:
cache = os.path.join(os.environ['HOME'], 'hash_temp')
#-----------------------------------------------------------------
# Check hash table for file:
#-----------------------------------------------------------------
if data_file not in hashes.keys():
if return_missing:
data_path = data_file
print("Retrieved file not in hashes: {0}".
format(data_path))
return data_path
else:
sys.exit("Data file '{0}' not in hash table.".
format(data_file))
else:
stored_hash = hashes[data_file]
#-------------------------------------------------------------
# Create missing cache and hash directories:
#-------------------------------------------------------------
if cache_env in os.environ.keys():
cache = os.environ[cache_env]
if not os.path.exists(cache):
print("Create missing cache directory: {0}".format(cache))
os.mkdir(cache)
hash_dir = os.path.join(cache, stored_hash)
if not os.path.exists(hash_dir):
print("Create missing hash directory: {0}".
format(hash_dir))
os.mkdir(os.path.join(hash_dir))
#-------------------------------------------------------------
# Check hash subdirectory for file:
#-------------------------------------------------------------
data_path = os.path.join(hash_dir, data_file)
if os.path.exists(data_path):
return data_path
#-------------------------------------------------------------
# If file not in cache, download, compute hash, and verify:
#-------------------------------------------------------------
else:
print("Retrieve file from the Mindboggle website: {0}".
format(url+data_file))
# Download file as a temporary file:
temp_file = fetch_data(url+data_file)
# Compute the file's hash:
data_hash = fetch_hash(temp_file)
# If hash matches name of the hash directory, save file:
if os.path.join(cache, data_hash) == hash_dir:
print("Copy file to cache: {0}".format(data_path))
shutil.copyfile(temp_file, data_path)
return data_path
else:
print("Retrieved hash does not match stored hash.")
#---------------------------------------------------------------------
# If hashes not provided, simply download file:
#---------------------------------------------------------------------
elif url:
# Download file as a temporary file:
data_path = fetch_data(url+data_file)
print("Hashes not provided. Retrieved file: {0}".format(data_path))
return data_path
#---------------------------------------------------------------------
# If URL also not provided, simply return file path:
#---------------------------------------------------------------------
else:
data_path = data_file
print("Neither hashes nor URL provided. "
"Returning file path: {0}".format(data_path))
return data_path
#-------------------------------------------------------------------------
# Simply return file path:
#-------------------------------------------------------------------------
else:
data_path = data_file
print("Returning file path: {0}".format(data_path))
return data_path
def fetch_check_data(data_file,
url='',
hashes={},
cache_env='',
cache='',
return_missing=False,
lookup=True):
"""
Get data file through a URL call and check its hash.
Steps ::
If hashes provided:
1. Check hash table for data file.
2. Check hash subdirectory within cache directory for data file.
3. If data file not in cache, download file, compute hash,
and verify hash.
4. If hash correct, save file.
Otherwise, simply download file or return file path as a string.
Parameters
----------
data_file : string
data file name
url : string
URL for data file
hashes : dictionary
file names and md5 hashes (if empty, simply download file from url)
cache_env : string
environment variable name for cache path
cache : string
in case cache_env is not set, use as cache directory
return_missing : Boolean
if data_file not in hash, simply download data_file and return path
lookup : Boolean
Simply return data_file path
Returns
-------
data_path : string
data file name (full path)
Examples
--------
>>> import os
>>> from mindboggle.mio.fetch_data import hashes_url
>>> from mindboggle.mio.fetch_data import fetch_check_data
>>> hashes, url, cache_env, cache = hashes_url()
>>> data_file = hashes.keys()[0]
>>> fetch_check_data(data_file, url, hashes, cache_env, cache)
"""
import os
import sys
import shutil
from mindboggle.mio.fetch_data import fetch_data, fetch_hash
if lookup:
#---------------------------------------------------------------------
# If hashes provided, go through steps to check/download file:
#---------------------------------------------------------------------
if hashes:
if not cache_env:
cache_env = 'MINDBOGGLE_CACHE'
if not cache:
cache = os.path.join(os.environ['HOME'], 'hash_temp')
#-----------------------------------------------------------------
# Check hash table for file:
#-----------------------------------------------------------------
if data_file not in hashes.keys():
if return_missing:
data_path = data_file
print(
"Retrieved file not in hashes: {0}".format(data_path))
return data_path
else:
sys.exit(
"Data file '{0}' not in hash table.".format(data_file))
else:
stored_hash = hashes[data_file]
#-------------------------------------------------------------
# Create missing cache and hash directories:
#-------------------------------------------------------------
if cache_env in os.environ.keys():
cache = os.environ[cache_env]
if not os.path.exists(cache):
print("Create missing cache directory: {0}".format(cache))
os.mkdir(cache)
hash_dir = os.path.join(cache, stored_hash)
if not os.path.exists(hash_dir):
print(
"Create missing hash directory: {0}".format(hash_dir))
os.mkdir(os.path.join(hash_dir))
#-------------------------------------------------------------
# Check hash subdirectory for file:
#-------------------------------------------------------------
data_path = os.path.join(hash_dir, data_file)
if os.path.exists(data_path):
return data_path
#-------------------------------------------------------------
# If file not in cache, download, compute hash, and verify:
#-------------------------------------------------------------
else:
print("Retrieve file from the Mindboggle website: {0}".
format(url + data_file))
# Download file as a temporary file:
temp_file = fetch_data(url + data_file)
# Compute the file's hash:
data_hash = fetch_hash(temp_file)
# If hash matches name of the hash directory, save file:
if os.path.join(cache, data_hash) == hash_dir:
print("Copy file to cache: {0}".format(data_path))
shutil.copyfile(temp_file, data_path)
return data_path
else:
print("Retrieved hash does not match stored hash.")
#---------------------------------------------------------------------
# If hashes not provided, simply download file:
#---------------------------------------------------------------------
elif url:
# Download file as a temporary file:
data_path = fetch_data(url + data_file)
print("Hashes not provided. Retrieved file: {0}".format(data_path))
return data_path
#---------------------------------------------------------------------
# If URL also not provided, simply return file path:
#---------------------------------------------------------------------
else:
data_path = data_file
print("Neither hashes nor URL provided. "
"Returning file path: {0}".format(data_path))
return data_path
#-------------------------------------------------------------------------
# Simply return file path:
#-------------------------------------------------------------------------
else:
data_path = data_file
print("Returning file path: {0}".format(data_path))
return data_path
