def get_chunk_streams(self):
index = 0
while True:
data = b''
read_size = self.chunk_size
# Buffer until we either reach the end of the stream or get a whole chunk.
while True:
if self.blob_size:
read_size = min(self.chunk_size - len(data), self.blob_size - (index + len(data)))
temp = self.stream.read(read_size)
temp = _get_data_bytes_only('temp', temp)
data += temp
# We have read an empty string and so are at the end
# of the buffer or we have read a full chunk.
if temp == b'' or len(data) == self.chunk_size:
break
if len(data) == self.chunk_size:
if self.padder:
data = self.padder.update(data)
if self.encryptor:
data = self.encryptor.update(data)
yield index, data
else:
if self.padder:
data = self.padder.update(data) + self.padder.finalize()
if self.encryptor:
data = self.encryptor.update(data) + self.encryptor.finalize()
if len(data) > 0:
yield index, data
break
index += len(data)
def get_chunk_streams(self):
index = 0
while True:
data = b''
read_size = self.chunk_size
# Buffer until we either reach the end of the stream or get a whole chunk.
while True:
if self.blob_size:
read_size = min(self.chunk_size - len(data), self.blob_size - (index + len(data)))
temp = self.stream.read(read_size)
temp = _get_data_bytes_only('temp', temp)
data += temp
# We have read an empty string and so are at the end
# of the buffer or we have read a full chunk.
if temp == b'' or len(data) == self.chunk_size:
break
if len(data) == self.chunk_size:
if self.padder:
data = self.padder.update(data)
if self.encryptor:
data = self.encryptor.update(data)
yield index, BytesIO(data)
else:
if self.padder:
data = self.padder.update(data) + self.padder.finalize()
if self.encryptor:
data = self.encryptor.update(data) + self.encryptor.finalize()
if len(data) > 0:
yield index, BytesIO(data)
break
index += len(data)
def _put_blob(self, container_name, blob_name, blob, content_settings=None,
metadata=None, validate_content=False, lease_id=None, if_modified_since=None,
if_unmodified_since=None, if_match=None, if_none_match=None,
timeout=None):
'''
Creates a blob or updates an existing blob.
See create_blob_from_* for high level
functions that handle the creation and upload of large blobs with
automatic chunking and progress notifications.
:param str container_name:
Name of existing container.
:param str blob_name:
Name of blob to create or update.
:param bytes blob:
Content of blob as bytes (size < 64MB). For larger size, you
must call put_block and put_block_list to set content of blob.
:param ~azure.storage.blob.models.ContentSettings content_settings:
ContentSettings object used to set properties on the blob.
:param metadata:
Name-value pairs associated with the blob as metadata.
:param bool validate_content:
If true, calculates an MD5 hash of the blob content. The storage
service checks the hash of the content that has arrived
with the hash that was sent. This is primarily valuable for detecting
bitflips on the wire if using http instead of https as https (the default)
will already validate. Note that this MD5 hash is not stored with the
blob.
:param str lease_id:
Required if the blob has an active lease.
:param datetime if_modified_since:
A DateTime value. Azure expects the date value passed in to be UTC.
If timezone is included, any non-UTC datetimes will be converted to UTC.
If a date is passed in without timezone info, it is assumed to be UTC.
Specify this header to perform the operation only
if the resource has been modified since the specified time.
:param datetime if_unmodified_since:
A DateTime value. Azure expects the date value passed in to be UTC.
If timezone is included, any non-UTC datetimes will be converted to UTC.
If a date is passed in without timezone info, it is assumed to be UTC.
Specify this header to perform the operation only if
the resource has not been modified since the specified date/time.
:param str if_match:
An ETag value, or the wildcard character (*). Specify this header to perform
the operation only if the resource's ETag matches the value specified.
:param str if_none_match:
An ETag value, or the wildcard character (*). Specify this header
to perform the operation only if the resource's ETag does not match
the value specified. Specify the wildcard character (*) to perform
the operation only if the resource does not exist, and fail the
operation if it does exist.
:param int timeout:
The timeout parameter is expressed in seconds.
:return: ETag and last modified properties for the new Block Blob
:rtype: :class:`~azure.storage.blob.models.ResourceProperties`
'''
_validate_not_none('container_name', container_name)
_validate_not_none('blob_name', blob_name)
_validate_encryption_required(self.require_encryption, self.key_encryption_key)
request = HTTPRequest()
request.method = 'PUT'
request.host_locations = self._get_host_locations()
request.path = _get_path(container_name, blob_name)
request.query = {'timeout': _int_to_str(timeout)}
request.headers = {
'x-ms-blob-type': _to_str(self.blob_type),
'x-ms-lease-id': _to_str(lease_id),
'If-Modified-Since': _datetime_to_utc_string(if_modified_since),
'If-Unmodified-Since': _datetime_to_utc_string(if_unmodified_since),
'If-Match': _to_str(if_match),
'If-None-Match': _to_str(if_none_match)
}
_add_metadata_headers(metadata, request)
if content_settings is not None:
request.headers.update(content_settings._to_headers())
blob = _get_data_bytes_only('blob', blob)
if self.key_encryption_key:
encryption_data, blob = _encrypt_blob(blob, self.key_encryption_key)
request.headers['x-ms-meta-encryptiondata'] = encryption_data
request.body = blob
if validate_content:
computed_md5 = _get_content_md5(request.body)
request.headers['Content-MD5'] = _to_str(computed_md5)
return self._perform_request(request, _parse_base_properties)
def _put_blob(self, container_name, blob_name, blob, content_settings=None,
metadata=None, validate_content=False, lease_id=None, if_modified_since=None,
if_unmodified_since=None, if_match=None, if_none_match=None,
timeout=None):
'''
Creates a blob or updates an existing blob.
See create_blob_from_* for high level
functions that handle the creation and upload of large blobs with
automatic chunking and progress notifications.
:param str container_name:
Name of existing container.
:param str blob_name:
Name of blob to create or update.
:param bytes blob:
Content of blob as bytes (size < 64MB). For larger size, you
must call put_block and put_block_list to set content of blob.
:param ~azure.storage.blob.models.ContentSettings content_settings:
ContentSettings object used to set properties on the blob.
:param metadata:
Name-value pairs associated with the blob as metadata.
:param bool validate_content:
If true, calculates an MD5 hash of the blob content. The storage
service checks the hash of the content that has arrived
with the hash that was sent. This is primarily valuable for detecting
bitflips on the wire if using http instead of https as https (the default)
will already validate. Note that this MD5 hash is not stored with the
blob.
:param str lease_id:
Required if the blob has an active lease.
:param datetime if_modified_since:
A DateTime value. Azure expects the date value passed in to be UTC.
If timezone is included, any non-UTC datetimes will be converted to UTC.
If a date is passed in without timezone info, it is assumed to be UTC.
Specify this header to perform the operation only
if the resource has been modified since the specified time.
:param datetime if_unmodified_since:
A DateTime value. Azure expects the date value passed in to be UTC.
If timezone is included, any non-UTC datetimes will be converted to UTC.
If a date is passed in without timezone info, it is assumed to be UTC.
Specify this header to perform the operation only if
the resource has not been modified since the specified date/time.
:param str if_match:
An ETag value, or the wildcard character (*). Specify this header to perform
the operation only if the resource's ETag matches the value specified.
:param str if_none_match:
An ETag value, or the wildcard character (*). Specify this header
to perform the operation only if the resource's ETag does not match
the value specified. Specify the wildcard character (*) to perform
the operation only if the resource does not exist, and fail the
operation if it does exist.
:param int timeout:
The timeout parameter is expressed in seconds.
:return: ETag and last modified properties for the new Block Blob
:rtype: :class:`~azure.storage.blob.models.ResourceProperties`
'''
_validate_not_none('container_name', container_name)
_validate_not_none('blob_name', blob_name)
_validate_encryption_required(self.require_encryption, self.key_encryption_key)
request = HTTPRequest()
request.method = 'PUT'
request.host_locations = self._get_host_locations()
request.path = _get_path(container_name, blob_name)
request.query = {'timeout': _int_to_str(timeout)}
request.headers = {
'x-ms-blob-type': _to_str(self.blob_type),
'x-ms-lease-id': _to_str(lease_id),
'If-Modified-Since': _datetime_to_utc_string(if_modified_since),
'If-Unmodified-Since': _datetime_to_utc_string(if_unmodified_since),
'If-Match': _to_str(if_match),
'If-None-Match': _to_str(if_none_match)
}
_add_metadata_headers(metadata, request)
if content_settings is not None:
request.headers.update(content_settings._to_headers())
blob = _get_data_bytes_only('blob', blob)
if self.key_encryption_key:
encryption_data, blob = _encrypt_blob(blob, self.key_encryption_key)
request.headers['x-ms-meta-encryptiondata'] = encryption_data
request.body = blob
if validate_content:
computed_md5 = _get_content_md5(request.body)
request.headers['Content-MD5'] = _to_str(computed_md5)
return self._perform_request(request, _parse_base_properties)
def append_block(self,
container_name,
blob_name,
block,
validate_content=False,
maxsize_condition=None,
appendpos_condition=None,
lease_id=None,
if_modified_since=None,
if_unmodified_since=None,
if_match=None,
if_none_match=None,
timeout=None,
cpk=None):
'''
Commits a new block of data to the end of an existing append blob.
:param str container_name:
Name of existing container.
:param str blob_name:
Name of existing blob.
:param bytes block:
Content of the block in bytes.
:param bool validate_content:
If true, calculates an MD5 hash of the block content. The storage
service checks the hash of the content that has arrived
with the hash that was sent. This is primarily valuable for detecting
bitflips on the wire if using http instead of https as https (the default)
will already validate. Note that this MD5 hash is not stored with the
blob.
:param int maxsize_condition:
Optional conditional header. The max length in bytes permitted for
the append blob. If the Append Block operation would cause the blob
to exceed that limit or if the blob size is already greater than the
value specified in this header, the request will fail with
MaxBlobSizeConditionNotMet error (HTTP status code 412 - Precondition Failed).
:param int appendpos_condition:
Optional conditional header, used only for the Append Block operation.
A number indicating the byte offset to compare. Append Block will
succeed only if the append position is equal to this number. If it
is not, the request will fail with the
AppendPositionConditionNotMet error
(HTTP status code 412 - Precondition Failed).
:param str lease_id:
Required if the blob has an active lease.
:param datetime if_modified_since:
A DateTime value. Azure expects the date value passed in to be UTC.
If timezone is included, any non-UTC datetimes will be converted to UTC.
If a date is passed in without timezone info, it is assumed to be UTC.
Specify this header to perform the operation only
if the resource has been modified since the specified time.
:param datetime if_unmodified_since:
A DateTime value. Azure expects the date value passed in to be UTC.
If timezone is included, any non-UTC datetimes will be converted to UTC.
If a date is passed in without timezone info, it is assumed to be UTC.
Specify this header to perform the operation only if
the resource has not been modified since the specified date/time.
:param str if_match:
An ETag value, or the wildcard character (*). Specify this header to perform
the operation only if the resource's ETag matches the value specified.
:param str if_none_match:
An ETag value, or the wildcard character (*). Specify this header
to perform the operation only if the resource's ETag does not match
the value specified. Specify the wildcard character (*) to perform
the operation only if the resource does not exist, and fail the
operation if it does exist.
:param ~azure.storage.blob.models.CustomerProvidedEncryptionKey cpk:
Encrypts the data on the service-side with the given key.
Use of customer-provided keys must be done over HTTPS.
As the encryption key itself is provided in the request,
a secure connection must be established to transfer the key.
:param int timeout:
The timeout parameter is expressed in seconds.
:return:
ETag, last modified, append offset, and committed block count
properties for the updated Append Blob
:rtype: :class:`~azure.storage.blob.models.AppendBlockProperties`
'''
_validate_not_none('container_name', container_name)
_validate_not_none('blob_name', blob_name)
_validate_not_none('block', block)
_validate_encryption_unsupported(self.require_encryption,
self.key_encryption_key)
request = HTTPRequest()
request.method = 'PUT'
request.host_locations = self._get_host_locations()
request.path = _get_path(container_name, blob_name)
request.query = {
'comp': 'appendblock',
'timeout': _int_to_str(timeout),
}
request.headers = {
'x-ms-blob-condition-maxsize': _to_str(maxsize_condition),
'x-ms-blob-condition-appendpos': _to_str(appendpos_condition),
'x-ms-lease-id': _to_str(lease_id),
'If-Modified-Since': _datetime_to_utc_string(if_modified_since),
'If-Unmodified-Since':
_datetime_to_utc_string(if_unmodified_since),
'If-Match': _to_str(if_match),
'If-None-Match': _to_str(if_none_match)
}
_validate_and_add_cpk_headers(request,
encryption_key=cpk,
protocol=self.protocol)
request.body = _get_data_bytes_only('block', block)
if validate_content:
computed_md5 = _get_content_md5(request.body)
request.headers['Content-MD5'] = _to_str(computed_md5)
return self._perform_request(request, _parse_append_block)
def append_block(self, container_name, blob_name, block,
validate_content=False, maxsize_condition=None,
appendpos_condition=None,
lease_id=None, if_modified_since=None,
if_unmodified_since=None, if_match=None,
if_none_match=None, timeout=None):
'''
Commits a new block of data to the end of an existing append blob.
:param str container_name:
Name of existing container.
:param str blob_name:
Name of existing blob.
:param bytes block:
Content of the block in bytes.
:param bool validate_content:
If true, calculates an MD5 hash of the block content. The storage
service checks the hash of the content that has arrived
with the hash that was sent. This is primarily valuable for detecting
bitflips on the wire if using http instead of https as https (the default)
will already validate. Note that this MD5 hash is not stored with the
blob.
:param int maxsize_condition:
Optional conditional header. The max length in bytes permitted for
the append blob. If the Append Block operation would cause the blob
to exceed that limit or if the blob size is already greater than the
value specified in this header, the request will fail with
MaxBlobSizeConditionNotMet error (HTTP status code 412 - Precondition Failed).
:param int appendpos_condition:
Optional conditional header, used only for the Append Block operation.
A number indicating the byte offset to compare. Append Block will
succeed only if the append position is equal to this number. If it
is not, the request will fail with the
AppendPositionConditionNotMet error
(HTTP status code 412 - Precondition Failed).
:param str lease_id:
Required if the blob has an active lease.
:param datetime if_modified_since:
A DateTime value. Azure expects the date value passed in to be UTC.
If timezone is included, any non-UTC datetimes will be converted to UTC.
If a date is passed in without timezone info, it is assumed to be UTC.
Specify this header to perform the operation only
if the resource has been modified since the specified time.
:param datetime if_unmodified_since:
A DateTime value. Azure expects the date value passed in to be UTC.
If timezone is included, any non-UTC datetimes will be converted to UTC.
If a date is passed in without timezone info, it is assumed to be UTC.
Specify this header to perform the operation only if
the resource has not been modified since the specified date/time.
:param str if_match:
An ETag value, or the wildcard character (*). Specify this header to perform
the operation only if the resource's ETag matches the value specified.
:param str if_none_match:
An ETag value, or the wildcard character (*). Specify this header
to perform the operation only if the resource's ETag does not match
the value specified. Specify the wildcard character (*) to perform
the operation only if the resource does not exist, and fail the
operation if it does exist.
:param int timeout:
The timeout parameter is expressed in seconds.
:return:
ETag, last modified, append offset, and committed block count
properties for the updated Append Blob
:rtype: :class:`~azure.storage.blob.models.AppendBlockProperties`
'''
_validate_not_none('container_name', container_name)
_validate_not_none('blob_name', blob_name)
_validate_not_none('block', block)
_validate_encryption_unsupported(self.require_encryption, self.key_encryption_key)
request = HTTPRequest()
request.method = 'PUT'
request.host_locations = self._get_host_locations()
request.path = _get_path(container_name, blob_name)
request.query = {
'comp': 'appendblock',
'timeout': _int_to_str(timeout),
}
request.headers = {
'x-ms-blob-condition-maxsize': _to_str(maxsize_condition),
'x-ms-blob-condition-appendpos': _to_str(appendpos_condition),
'x-ms-lease-id': _to_str(lease_id),
'If-Modified-Since': _datetime_to_utc_string(if_modified_since),
'If-Unmodified-Since': _datetime_to_utc_string(if_unmodified_since),
'If-Match': _to_str(if_match),
'If-None-Match': _to_str(if_none_match)
}
request.body = _get_data_bytes_only('block', block)
if validate_content:
computed_md5 = _get_content_md5(request.body)
request.headers['Content-MD5'] = _to_str(computed_md5)
return self._perform_request(request, _parse_append_block)
