def handle_uploaded_dataset_file(filename,
datatypes_registry,
ext='auto',
is_multi_byte=False):
is_valid, ext = handle_compressed_file(filename,
datatypes_registry,
ext=ext)
if not is_valid:
raise InappropriateDatasetContentError(
'The compressed uploaded file contains inappropriate content.')
if ext in AUTO_DETECT_EXTENSIONS:
ext = guess_ext(filename,
sniff_order=datatypes_registry.sniff_order,
is_multi_byte=is_multi_byte)
if check_binary(filename):
if not Binary.is_ext_unsniffable(
ext) and not datatypes_registry.get_datatype_by_extension(
ext).sniff(filename):
raise InappropriateDatasetContentError(
'The binary uploaded file contains inappropriate content.')
elif check_html(filename):
raise InappropriateDatasetContentError(
'The uploaded file contains inappropriate HTML content.')
return ext
def handle_uploaded_dataset_file( filename, datatypes_registry, ext = 'auto', is_multi_byte = False ):
is_valid, ext = handle_compressed_file( filename, datatypes_registry, ext = ext )
if not is_valid:
raise InappropriateDatasetContentError, 'The compressed uploaded file contains inappropriate content.'
if ext in AUTO_DETECT_EXTENSIONS:
ext = guess_ext( filename, sniff_order = datatypes_registry.sniff_order, is_multi_byte=is_multi_byte )
if check_binary( filename ):
if not Binary.is_ext_unsniffable(ext) and not datatypes_registry.get_datatype_by_extension( ext ).sniff( filename ):
raise InappropriateDatasetContentError, 'The binary uploaded file contains inappropriate content.'
elif check_html( filename ):
raise InappropriateDatasetContentError, 'The uploaded file contains inappropriate HTML content.'
return ext
def is_data_index_sample_file(file_path):
"""
Attempt to determine if a .sample file is appropriate for copying to ~/tool-data when
a tool shed repository is being installed into a Galaxy instance.
"""
# Currently most data index files are tabular, so check that first. We'll assume that
# if the file is tabular, it's ok to copy.
if is_column_based(file_path):
return True
# If the file is any of the following, don't copy it.
if checkers.check_html(file_path):
return False
if checkers.check_image(file_path):
return False
if checkers.check_binary(name=file_path):
return False
if checkers.is_bz2(file_path):
return False
if checkers.is_gzip(file_path):
return False
if checkers.check_zip(file_path):
return False
# Default to copying the file if none of the above are true.
return True
def is_data_index_sample_file(file_path):
"""
Attempt to determine if a .sample file is appropriate for copying to ~/tool-data when
a tool shed repository is being installed into a Galaxy instance.
"""
# Currently most data index files are tabular, so check that first. We'll assume that
# if the file is tabular, it's ok to copy.
if is_column_based(file_path):
return True
# If the file is any of the following, don't copy it.
if checkers.check_html(file_path):
return False
if checkers.check_image(file_path):
return False
if checkers.check_binary(name=file_path):
return False
if checkers.is_bz2(file_path):
return False
if checkers.is_gzip(file_path):
return False
if checkers.check_zip(file_path):
return False
# Default to copying the file if none of the above are true.
return True
z.close()
# Replace the zipped file with the decompressed file if it's safe to do so
if uncompressed is not None:
if dataset.type in ( 'server_dir', 'path_paste' ) or not in_place:
dataset.path = uncompressed
else:
shutil.move( uncompressed, dataset.path )
os.chmod(dataset.path, 0644)
dataset.name = uncompressed_name
data_type = 'zip'
if not data_type:
# TODO refactor this logic. check_binary isn't guaranteed to be
# correct since it only looks at whether the first 100 chars are
# printable or not. If someone specifies a known unsniffable
# binary datatype and check_binary fails, the file gets mangled.
if check_binary( dataset.path ) or Binary.is_ext_unsniffable(dataset.file_type):
# We have a binary dataset, but it is not Bam, Sff or Pdf
data_type = 'binary'
# binary_ok = False
parts = dataset.name.split( "." )
if len( parts ) > 1:
ext = parts[-1].strip().lower()
if not Binary.is_ext_unsniffable(ext):
file_err( 'The uploaded binary file contains inappropriate content', dataset, json_file )
return
elif Binary.is_ext_unsniffable(ext) and dataset.file_type != ext:
err_msg = "You must manually set the 'File Format' to '%s' when uploading %s files." % ( ext.capitalize(), ext )
file_err( err_msg, dataset, json_file )
return
if not data_type:
# We must have a text file
# Replace the zipped file with the decompressed file if it's safe to do so
if uncompressed is not None:
if dataset.type in ('server_dir',
'path_paste') or not in_place:
dataset.path = uncompressed
else:
shutil.move(uncompressed, dataset.path)
os.chmod(dataset.path, 0644)
dataset.name = uncompressed_name
data_type = 'zip'
if not data_type:
# TODO refactor this logic. check_binary isn't guaranteed to be
# correct since it only looks at whether the first 100 chars are
# printable or not. If someone specifies a known unsniffable
# binary datatype and check_binary fails, the file gets mangled.
if check_binary(dataset.path) or Binary.is_ext_unsniffable(
dataset.file_type):
# We have a binary dataset, but it is not Bam, Sff or Pdf
data_type = 'binary'
# binary_ok = False
parts = dataset.name.split(".")
if len(parts) > 1:
ext = parts[-1].strip().lower()
if not Binary.is_ext_unsniffable(ext):
file_err(
'The uploaded binary file contains inappropriate content',
dataset, json_file)
return
elif Binary.is_ext_unsniffable(
ext) and dataset.file_type != ext:
err_msg = "You must manually set the 'File Format' to '%s' when uploading %s files." % (
