def check_object_creation(req):
"""
Check to ensure that everything is alright about an object to be created.
:param req: HTTP request object
:returns: HTTPLengthRequired -- missing content-length header and not
a chunked request
:returns: HTTPBadRequest -- missing or bad content-type header, or
bad metadata
:returns: HTTPNotImplemented -- unsupported transfer-encoding header value
"""
try:
req.message_length()
except ValueError as e:
return HTTPBadRequest(request=req,
content_type='text/plain',
body=str(e))
except AttributeError as e:
return HTTPNotImplemented(request=req,
content_type='text/plain',
body=str(e))
if req.content_length is None and \
req.headers.get('transfer-encoding') != 'chunked':
return HTTPLengthRequired(body='Missing Content-Length header.',
request=req,
content_type='text/plain')
if 'Content-Type' not in req.headers:
return HTTPBadRequest(request=req,
content_type='text/plain',
body='No content type')
return None
def check_object_creation(req, object_name):
"""
Check to ensure that everything is alright about an object to be created.
:param req: HTTP request object
:param object_name: name of object to be created
:returns: HTTPRequestEntityTooLarge -- the object is too large
:returns: HTTPLengthRequired -- missing content-length header and not
a chunked request
:returns: HTTPBadRequest -- missing or bad content-type header, or
bad metadata
:returns: HTTPNotImplemented -- unsupported transfer-encoding header value
"""
try:
ml = req.message_length()
except ValueError as e:
return HTTPBadRequest(request=req,
content_type='text/plain',
body=str(e))
except AttributeError as e:
return HTTPNotImplemented(request=req,
content_type='text/plain',
body=str(e))
if ml is not None and ml > MAX_FILE_SIZE:
return HTTPRequestEntityTooLarge(body='Your request is too large.',
request=req,
content_type='text/plain')
if req.content_length is None and \
req.headers.get('transfer-encoding') != 'chunked':
return HTTPLengthRequired(body='Missing Content-Length header.',
request=req,
content_type='text/plain')
if len(object_name) > MAX_OBJECT_NAME_LENGTH:
return HTTPBadRequest(body='Object name length of %d longer than %d' %
(len(object_name), MAX_OBJECT_NAME_LENGTH),
request=req,
content_type='text/plain')
if 'Content-Type' not in req.headers:
return HTTPBadRequest(request=req,
content_type='text/plain',
body='No content type')
try:
req = check_delete_headers(req)
except HTTPException as e:
return HTTPBadRequest(request=req,
body=e.body,
content_type='text/plain')
if not check_utf8(req.headers['Content-Type']):
return HTTPBadRequest(request=req,
body='Invalid Content-Type',
content_type='text/plain')
return check_metadata(req, 'object')
def REPLICATE(self, request):
"""
In Swift, this method handles REPLICATE requests for the Swift
Object Server. This is used by the object replicator to get hashes
for directories.
File Connector does not support this as it expects the underlying
Shared filesystem to take care of data durability
"""
return HTTPNotImplemented(request=request)
def REPLICATE(self, request):
"""
In Swift, this method handles REPLICATE requests for the Swift
Object Server. This is used by the object replicator to get hashes
for directories.
Gluster-Swift does not support this as it expects the underlying
GlusterFS to take care of replication
"""
return HTTPNotImplemented(request=request)
def REPLICATE(self, request):
"""
In Swift, this method handles REPLICATE requests for the Swift
Object Server. This is used by the object replicator to get hashes
for directories.
Swiftonfile does not support this as it expects the underlying
filesystem to take care of replication. Also, swiftonfile has no
notion of hashes for directories.
"""
return HTTPNotImplemented(request=request)
def __call__(self, env, start_response):
method = env['REQUEST_METHOD']
if method not in self.ALLOWED_METHODS:
raise HTTPNotImplemented()
path = env['PATH_INFO']
_, acc, cont, obj = split_path(env['PATH_INFO'],
0,
4,
rest_with_last=True)
if env.get('QUERY_STRING'):
path += '?' + env['QUERY_STRING']
if 'swift.authorize' in env:
resp = env['swift.authorize'](swob.Request(env))
if resp:
return resp(env, start_response)
req = swob.Request(env)
self.swift_sources.append(env.get('swift.source'))
self.txn_ids.append(env.get('swift.trans_id'))
try:
resp_class, raw_headers, body = self._find_response(method, path)
headers = HeaderKeyDict(raw_headers)
except KeyError:
if (env.get('QUERY_STRING')
and (method, env['PATH_INFO']) in self._responses):
resp_class, raw_headers, body = self._find_response(
method, env['PATH_INFO'])
headers = HeaderKeyDict(raw_headers)
elif method == 'HEAD' and ('GET', path) in self._responses:
resp_class, raw_headers, body = self._find_response(
'GET', path)
body = None
headers = HeaderKeyDict(raw_headers)
elif method == 'GET' and obj and path in self.uploaded:
resp_class = swob.HTTPOk
headers, body = self.uploaded[path]
else:
raise KeyError("Didn't find %r in allowed responses" %
((method, path), ))
# simulate object PUT
if method == 'PUT' and obj:
put_body = ''.join(iter(env['wsgi.input'].read, ''))
if 'swift.callback.update_footers' in env:
footers = HeaderKeyDict()
env['swift.callback.update_footers'](footers)
req.headers.update(footers)
etag = md5(put_body).hexdigest()
headers.setdefault('Etag', etag)
headers.setdefault('Content-Length', len(put_body))
# keep it for subsequent GET requests later
self.uploaded[path] = (dict(req.headers), put_body)
if "CONTENT_TYPE" in env:
self.uploaded[path][0]['Content-Type'] = env["CONTENT_TYPE"]
# simulate object POST
elif method == 'POST' and obj:
metadata, data = self.uploaded.get(path, ({}, None))
# select items to keep from existing...
new_metadata = dict((k, v) for k, v in metadata.items()
if (not is_user_meta('object', k)
and not is_object_transient_sysmeta(k)))
# apply from new
new_metadata.update(
dict((k, v) for k, v in req.headers.items() if (
is_user_meta('object', k) or is_object_transient_sysmeta(k)
or k.lower == 'content-type')))
self.uploaded[path] = new_metadata, data
# note: tests may assume this copy of req_headers is case insensitive
# so we deliberately use a HeaderKeyDict
self._calls.append(
FakeSwiftCall(method, path, HeaderKeyDict(req.headers)))
# Apply conditional etag overrides
conditional_etag = resolve_etag_is_at_header(req, headers)
# range requests ought to work, hence conditional_response=True
if isinstance(body, list):
resp = resp_class(req=req,
headers=headers,
app_iter=body,
conditional_response=req.method
in ('GET', 'HEAD'),
conditional_etag=conditional_etag)
else:
resp = resp_class(req=req,
headers=headers,
body=body,
conditional_response=req.method
in ('GET', 'HEAD'),
conditional_etag=conditional_etag)
wsgi_iter = resp(env, start_response)
self.mark_opened(path)
return LeakTrackingIter(wsgi_iter, self.mark_closed, path)
def PUT(self, req):
"""HTTP PUT request handler."""
if req.if_none_match is not None and '*' not in req.if_none_match:
# Sending an etag with if-none-match isn't currently supported
return HTTPBadRequest(request=req, content_type='text/plain',
body='If-None-Match only supports *')
container_info = self.container_info(
self.account_name, self.container_name, req)
policy_index = req.headers.get('X-Backend-Storage-Policy-Index',
container_info['storage_policy'])
obj_ring = self.app.get_object_ring(policy_index)
# pass the policy index to storage nodes via req header
req.headers['X-Backend-Storage-Policy-Index'] = policy_index
container_partition = container_info['partition']
containers = container_info['nodes']
req.acl = container_info['write_acl']
req.environ['swift_sync_key'] = container_info['sync_key']
object_versions = container_info['versions']
if 'swift.authorize' in req.environ:
aresp = req.environ['swift.authorize'](req)
if aresp:
return aresp
if not containers:
return HTTPNotFound(request=req)
try:
ml = req.message_length()
except ValueError as e:
return HTTPBadRequest(request=req, content_type='text/plain',
body=str(e))
except AttributeError as e:
return HTTPNotImplemented(request=req, content_type='text/plain',
body=str(e))
if ml is not None and ml > constraints.MAX_FILE_SIZE:
return HTTPRequestEntityTooLarge(request=req)
if 'x-delete-after' in req.headers:
try:
x_delete_after = int(req.headers['x-delete-after'])
except ValueError:
return HTTPBadRequest(request=req,
content_type='text/plain',
body='Non-integer X-Delete-After')
req.headers['x-delete-at'] = normalize_delete_at_timestamp(
time.time() + x_delete_after)
partition, nodes = obj_ring.get_nodes(
self.account_name, self.container_name, self.object_name)
# do a HEAD request for container sync and checking object versions
if 'x-timestamp' in req.headers or \
(object_versions and not
req.environ.get('swift_versioned_copy')):
# make sure proxy-server uses the right policy index
_headers = {'X-Backend-Storage-Policy-Index': policy_index,
'X-Newest': 'True'}
hreq = Request.blank(req.path_info, headers=_headers,
environ={'REQUEST_METHOD': 'HEAD'})
hresp = self.GETorHEAD_base(
hreq, _('Object'), obj_ring, partition,
hreq.swift_entity_path)
# Used by container sync feature
if 'x-timestamp' in req.headers:
try:
req_timestamp = Timestamp(req.headers['X-Timestamp'])
if hresp.environ and 'swift_x_timestamp' in hresp.environ and \
hresp.environ['swift_x_timestamp'] >= req_timestamp:
return HTTPAccepted(request=req)
except ValueError:
return HTTPBadRequest(
request=req, content_type='text/plain',
body='X-Timestamp should be a UNIX timestamp float value; '
'was %r' % req.headers['x-timestamp'])
req.headers['X-Timestamp'] = req_timestamp.internal
else:
req.headers['X-Timestamp'] = Timestamp(time.time()).internal
# Sometimes the 'content-type' header exists, but is set to None.
content_type_manually_set = True
detect_content_type = \
config_true_value(req.headers.get('x-detect-content-type'))
if detect_content_type or not req.headers.get('content-type'):
guessed_type, _junk = mimetypes.guess_type(req.path_info)
req.headers['Content-Type'] = guessed_type or \
'application/octet-stream'
if detect_content_type:
req.headers.pop('x-detect-content-type')
else:
content_type_manually_set = False
error_response = check_object_creation(req, self.object_name) or \
check_content_type(req)
if error_response:
return error_response
if object_versions and not req.environ.get('swift_versioned_copy'):
if hresp.status_int != HTTP_NOT_FOUND:
# This is a version manifest and needs to be handled
# differently. First copy the existing data to a new object,
# then write the data from this request to the version manifest
# object.
lcontainer = object_versions.split('/')[0]
prefix_len = '%03x' % len(self.object_name)
lprefix = prefix_len + self.object_name + '/'
ts_source = hresp.environ.get('swift_x_timestamp')
if ts_source is None:
ts_source = time.mktime(time.strptime(
hresp.headers['last-modified'],
'%a, %d %b %Y %H:%M:%S GMT'))
new_ts = Timestamp(ts_source).internal
vers_obj_name = lprefix + new_ts
copy_headers = {
'Destination': '%s/%s' % (lcontainer, vers_obj_name)}
copy_environ = {'REQUEST_METHOD': 'COPY',
'swift_versioned_copy': True
}
copy_req = Request.blank(req.path_info, headers=copy_headers,
environ=copy_environ)
copy_resp = self.COPY(copy_req)
if is_client_error(copy_resp.status_int):
# missing container or bad permissions
return HTTPPreconditionFailed(request=req)
elif not is_success(copy_resp.status_int):
# could not copy the data, bail
return HTTPServiceUnavailable(request=req)
reader = req.environ['wsgi.input'].read
data_source = iter(lambda: reader(self.app.client_chunk_size), '')
source_header = req.headers.get('X-Copy-From')
source_resp = None
if source_header:
if req.environ.get('swift.orig_req_method', req.method) != 'POST':
req.environ.setdefault('swift.log_info', []).append(
'x-copy-from:%s' % source_header)
src_container_name, src_obj_name = check_copy_from_header(req)
ver, acct, _rest = req.split_path(2, 3, True)
if isinstance(acct, unicode):
acct = acct.encode('utf-8')
source_header = '/%s/%s/%s/%s' % (ver, acct,
src_container_name, src_obj_name)
source_req = req.copy_get()
# make sure the source request uses it's container_info
source_req.headers.pop('X-Backend-Storage-Policy-Index', None)
source_req.path_info = source_header
source_req.headers['X-Newest'] = 'true'
orig_obj_name = self.object_name
orig_container_name = self.container_name
self.object_name = src_obj_name
self.container_name = src_container_name
sink_req = Request.blank(req.path_info,
environ=req.environ, headers=req.headers)
source_resp = self.GET(source_req)
# This gives middlewares a way to change the source; for example,
# this lets you COPY a SLO manifest and have the new object be the
# concatenation of the segments (like what a GET request gives
# the client), not a copy of the manifest file.
hook = req.environ.get(
'swift.copy_hook',
(lambda source_req, source_resp, sink_req: source_resp))
source_resp = hook(source_req, source_resp, sink_req)
if source_resp.status_int >= HTTP_MULTIPLE_CHOICES:
return source_resp
self.object_name = orig_obj_name
self.container_name = orig_container_name
data_source = iter(source_resp.app_iter)
sink_req.content_length = source_resp.content_length
if sink_req.content_length is None:
# This indicates a transfer-encoding: chunked source object,
# which currently only happens because there are more than
# CONTAINER_LISTING_LIMIT segments in a segmented object. In
# this case, we're going to refuse to do the server-side copy.
return HTTPRequestEntityTooLarge(request=req)
if sink_req.content_length > constraints.MAX_FILE_SIZE:
return HTTPRequestEntityTooLarge(request=req)
sink_req.etag = source_resp.etag
# we no longer need the X-Copy-From header
del sink_req.headers['X-Copy-From']
if not content_type_manually_set:
sink_req.headers['Content-Type'] = \
source_resp.headers['Content-Type']
if not config_true_value(
sink_req.headers.get('x-fresh-metadata', 'false')):
copy_headers_into(source_resp, sink_req)
copy_headers_into(req, sink_req)
# copy over x-static-large-object for POSTs and manifest copies
if 'X-Static-Large-Object' in source_resp.headers and \
req.params.get('multipart-manifest') == 'get':
sink_req.headers['X-Static-Large-Object'] = \
source_resp.headers['X-Static-Large-Object']
req = sink_req
if 'x-delete-at' in req.headers:
try:
x_delete_at = normalize_delete_at_timestamp(
int(req.headers['x-delete-at']))
if int(x_delete_at) < time.time():
return HTTPBadRequest(
body='X-Delete-At in past', request=req,
content_type='text/plain')
except ValueError:
return HTTPBadRequest(request=req, content_type='text/plain',
body='Non-integer X-Delete-At')
req.environ.setdefault('swift.log_info', []).append(
'x-delete-at:%s' % x_delete_at)
delete_at_container = normalize_delete_at_timestamp(
int(x_delete_at) /
self.app.expiring_objects_container_divisor *
self.app.expiring_objects_container_divisor)
delete_at_part, delete_at_nodes = \
self.app.container_ring.get_nodes(
self.app.expiring_objects_account, delete_at_container)
else:
delete_at_container = delete_at_part = delete_at_nodes = None
node_iter = GreenthreadSafeIterator(
self.iter_nodes_local_first(obj_ring, partition))
pile = GreenPile(len(nodes))
te = req.headers.get('transfer-encoding', '')
chunked = ('chunked' in te)
outgoing_headers = self._backend_requests(
req, len(nodes), container_partition, containers,
delete_at_container, delete_at_part, delete_at_nodes)
for nheaders in outgoing_headers:
# RFC2616:8.2.3 disallows 100-continue without a body
if (req.content_length > 0) or chunked:
nheaders['Expect'] = '100-continue'
pile.spawn(self._connect_put_node, node_iter, partition,
req.swift_entity_path, nheaders,
self.app.logger.thread_locals)
conns = [conn for conn in pile if conn]
min_conns = quorum_size(len(nodes))
if req.if_none_match is not None and '*' in req.if_none_match:
statuses = [conn.resp.status for conn in conns if conn.resp]
if HTTP_PRECONDITION_FAILED in statuses:
# If we find any copy of the file, it shouldn't be uploaded
self.app.logger.debug(
_('Object PUT returning 412, %(statuses)r'),
{'statuses': statuses})
return HTTPPreconditionFailed(request=req)
if len(conns) < min_conns:
self.app.logger.error(
_('Object PUT returning 503, %(conns)s/%(nodes)s '
'required connections'),
{'conns': len(conns), 'nodes': min_conns})
return HTTPServiceUnavailable(request=req)
bytes_transferred = 0
try:
with ContextPool(len(nodes)) as pool:
for conn in conns:
conn.failed = False
conn.queue = Queue(self.app.put_queue_depth)
pool.spawn(self._send_file, conn, req.path)
while True:
with ChunkReadTimeout(self.app.client_timeout):
try:
chunk = next(data_source)
except StopIteration:
if chunked:
for conn in conns:
conn.queue.put('0\r\n\r\n')
break
bytes_transferred += len(chunk)
if bytes_transferred > constraints.MAX_FILE_SIZE:
return HTTPRequestEntityTooLarge(request=req)
for conn in list(conns):
if not conn.failed:
conn.queue.put(
'%x\r\n%s\r\n' % (len(chunk), chunk)
if chunked else chunk)
else:
conns.remove(conn)
if len(conns) < min_conns:
self.app.logger.error(_(
'Object PUT exceptions during'
' send, %(conns)s/%(nodes)s required connections'),
{'conns': len(conns), 'nodes': min_conns})
return HTTPServiceUnavailable(request=req)
for conn in conns:
if conn.queue.unfinished_tasks:
conn.queue.join()
conns = [conn for conn in conns if not conn.failed]
except ChunkReadTimeout as err:
self.app.logger.warn(
_('ERROR Client read timeout (%ss)'), err.seconds)
self.app.logger.increment('client_timeouts')
return HTTPRequestTimeout(request=req)
except (Exception, Timeout):
self.app.logger.exception(
_('ERROR Exception causing client disconnect'))
return HTTPClientDisconnect(request=req)
if req.content_length and bytes_transferred < req.content_length:
req.client_disconnect = True
self.app.logger.warn(
_('Client disconnected without sending enough data'))
self.app.logger.increment('client_disconnects')
return HTTPClientDisconnect(request=req)
statuses, reasons, bodies, etags = self._get_put_responses(req, conns,
nodes)
if len(etags) > 1:
self.app.logger.error(
_('Object servers returned %s mismatched etags'), len(etags))
return HTTPServerError(request=req)
etag = etags.pop() if len(etags) else None
resp = self.best_response(req, statuses, reasons, bodies,
_('Object PUT'), etag=etag)
if source_header:
resp.headers['X-Copied-From'] = quote(
source_header.split('/', 3)[3])
if 'last-modified' in source_resp.headers:
resp.headers['X-Copied-From-Last-Modified'] = \
source_resp.headers['last-modified']
copy_headers_into(req, resp)
resp.last_modified = math.ceil(
float(Timestamp(req.headers['X-Timestamp'])))
return resp
def REPLICATION(self, request):
return HTTPNotImplemented(request=request)
def check_object_creation(req, object_name):
"""
Check to ensure that everything is alright about an object to be created.
:param req: HTTP request object
:param object_name: name of object to be created
:returns HTTPRequestEntityTooLarge: the object is too large
:returns HTTPLengthRequired: missing content-length header and not
a chunked request
:returns HTTPBadRequest: missing or bad content-type header, or
bad metadata
:returns HTTPNotImplemented: unsupported transfer-encoding header value
"""
try:
# 获取消息的长度，如果headers中没有包含消息长度，返回None，报错
ml = req.message_length()
except ValueError as e:
return HTTPBadRequest(request=req,
content_type='text/plain',
body=str(e))
except AttributeError as e:
return HTTPNotImplemented(request=req,
content_type='text/plain',
body=str(e))
# 如果请求消息的长度超过5G，报错
if ml is not None and ml > MAX_FILE_SIZE:
return HTTPRequestEntityTooLarge(body='Your request is too large.',
request=req,
content_type='text/plain')
if req.content_length is None and \
req.headers.get('transfer-encoding') != 'chunked':
return HTTPLengthRequired(body='Missing Content-Length header.',
request=req,
content_type='text/plain')
# 如果请求头中有拷贝源，但没有长度，报错
if 'X-Copy-From' in req.headers and req.content_length:
return HTTPBadRequest(body='Copy requests require a zero byte body',
request=req,
content_type='text/plain')
# 如果对象名称大于1024字节，报错
if len(object_name) > MAX_OBJECT_NAME_LENGTH:
return HTTPBadRequest(body='Object name length of %d longer than %d' %
(len(object_name), MAX_OBJECT_NAME_LENGTH),
request=req,
content_type='text/plain')
# 如果请求头中没有内容类型，报错
if 'Content-Type' not in req.headers:
return HTTPBadRequest(request=req,
content_type='text/plain',
body='No content type')
try:
# 检查请求头中关于对象删除的信息，失败报错
req = check_delete_headers(req)
except HTTPException as e:
return HTTPBadRequest(request=req,
body=e.body,
content_type='text/plain')
# 检查内容类型是否为utf-8编码，失败报错
if not check_utf8(req.headers['Content-Type']):
return HTTPBadRequest(request=req,
body='Invalid Content-Type',
content_type='text/plain')
# 检查请求中的用户自定义的元数据，失败报错
return check_metadata(req, 'object')
def __call__(self, env, start_response):
method = env['REQUEST_METHOD']
if method not in self.ALLOWED_METHODS:
raise HTTPNotImplemented()
path = env['PATH_INFO']
_, acc, cont, obj = split_path(env['PATH_INFO'],
0,
4,
rest_with_last=True)
if env.get('QUERY_STRING'):
path += '?' + env['QUERY_STRING']
if 'swift.authorize' in env:
resp = env['swift.authorize'](swob.Request(env))
if resp:
return resp(env, start_response)
req_headers = swob.Request(env).headers
self.swift_sources.append(env.get('swift.source'))
self.txn_ids.append(env.get('swift.trans_id'))
try:
resp_class, raw_headers, body = self._find_response(method, path)
headers = HeaderKeyDict(raw_headers)
except KeyError:
if (env.get('QUERY_STRING')
and (method, env['PATH_INFO']) in self._responses):
resp_class, raw_headers, body = self._find_response(
method, env['PATH_INFO'])
headers = HeaderKeyDict(raw_headers)
elif method == 'HEAD' and ('GET', path) in self._responses:
resp_class, raw_headers, body = self._find_response(
'GET', path)
body = None
headers = HeaderKeyDict(raw_headers)
elif method == 'GET' and obj and path in self.uploaded:
resp_class = swob.HTTPOk
headers, body = self.uploaded[path]
else:
raise KeyError("Didn't find %r in allowed responses" %
((method, path), ))
self._calls.append((method, path, req_headers))
# simulate object PUT
if method == 'PUT' and obj:
input = env['wsgi.input'].read()
etag = md5(input).hexdigest()
headers.setdefault('Etag', etag)
headers.setdefault('Content-Length', len(input))
# keep it for subsequent GET requests later
self.uploaded[path] = (deepcopy(headers), input)
if "CONTENT_TYPE" in env:
self.uploaded[path][0]['Content-Type'] = env["CONTENT_TYPE"]
# range requests ought to work, which require conditional_response=True
req = swob.Request(env)
resp = resp_class(req=req,
headers=headers,
body=body,
conditional_response=req.method in ('GET', 'HEAD'))
wsgi_iter = resp(env, start_response)
self.mark_opened(path)
return LeakTrackingIter(wsgi_iter, self, path)