def getCanonicalURI(self, bucket=None, object=None):
URI = ''
if self.path_style and bucket is not None and bucket != '':
URI += '/' + bucket
if object is not None:
URI += '/' + object
if not URI:
URI = '/'
return common_util.encode_object_key(URI)
def __make_canonicalstring(self,
method,
bucket_name,
key,
path_args,
headers,
expires=None):
str_list = []
str_list.append(method + '\n')
# 添加所有相关的头部字段（Content-MD5, Content-Type, Date和以x-amz开头的），并排序
interesting_headers = {} # 使用字典表示
content_list = ['content-type', 'content-md5', 'date']
if isinstance(headers, dict):
for hash_key in headers.keys():
lk = hash_key.lower() # headers的key值的小写
# 忽略不相关的HTTP头字段
if lk in content_list or lk.startswith(
common_util.AMAZON_HEADER_PREFIX):
s = headers.get(hash_key) # 获得headers中的值列表
interesting_headers[lk] = ''.join(s)
keylist = interesting_headers.keys()
# 如果有amz的时间标记就无需加入原有的date标记
if common_util.ALTERNATIVE_DATE_HEADER in keylist:
interesting_headers.setdefault('date', '')
# 如果过期时间不为空，则将过期时间填入date字段中
if expires:
interesting_headers['date'] = expires
# 这些字段必须要加入，故即使没有设置也要加入
if not 'content-type' in keylist:
interesting_headers['content-type'] = ''
if not 'content-md5' in keylist:
interesting_headers['content-md5'] = ''
# 取出字典中的key并进行排序
keylist = sorted(interesting_headers.keys())
# 最后加入所有相关的HTTP头部字段 (例如: 所有以x-amz-开头的)
for k in keylist:
header_key = str(k)
if header_key.startswith(common_util.AMAZON_HEADER_PREFIX):
str_list.append(header_key + ':' +
interesting_headers[header_key])
else:
str_list.append(interesting_headers[header_key])
str_list.append('\n')
URI = ''
# 使用存储空间名和对象名构建路径
if bucket_name is not None and bucket_name != '':
URI += '/'
URI += bucket_name
if not self.path_style:
URI += '/'
# 对象名不为空，则添加对象名到待签名的字符串中
if key is not None:
# 再加入一个反斜杠
if not URI.endswith('/'):
URI += '/'
URI += common_util.encode_object_key(key)
str_list.append(URI) if URI else str_list.append('/')
# 最后检查路径参数里是否有ACL，有则加入
if path_args:
e1 = '?'
e2 = '&'
for path_key, path_value in path_args.items():
flag = True
if path_key not in common_util.ALLOWED_RESOURCE_PARAMTER_NAMES:
flag = False
if flag:
path_key = common_util.toString(path_key)
if path_value is None:
e1 += path_key + '&'
continue
e2 += path_key + '=' + common_util.toString(
path_value) + '&'
e = (e1 + e2).replace('&&', '&').replace('?&', '?')[:-1]
str_list.append(e)
return ''.join(item for item in str_list) # 返回待签名的字符串
def __make_canonicalstring(self,
method,
bucket_name,
key,
path_args,
headers,
expires=None):
str_list = []
str_list.append(method + '\n')
interesting_headers = {}
content_list = ['content-type', 'content-md5', 'date']
if isinstance(headers, dict):
for hash_key in headers.keys():
lk = hash_key.lower()
if lk in content_list or lk.startswith(
common_util.AMAZON_HEADER_PREFIX):
s = headers.get(hash_key)
interesting_headers[lk] = ''.join(s)
keylist = interesting_headers.keys()
if common_util.ALTERNATIVE_DATE_HEADER in keylist:
interesting_headers.setdefault('date', '')
if expires:
interesting_headers['date'] = expires
if 'content-type' not in keylist:
interesting_headers['content-type'] = ''
if 'content-md5' not in keylist:
interesting_headers['content-md5'] = ''
keylist = sorted(interesting_headers.keys())
for k in keylist:
header_key = str(k)
if header_key.startswith(common_util.AMAZON_HEADER_PREFIX):
str_list.append(header_key + ':' +
interesting_headers[header_key])
else:
str_list.append(interesting_headers[header_key])
str_list.append('\n')
URI = ''
if bucket_name is not None and bucket_name != '':
URI += '/'
URI += bucket_name
if not self.path_style:
URI += '/'
if key is not None:
if not URI.endswith('/'):
URI += '/'
URI += common_util.encode_object_key(key)
if URI:
str_list.append(URI)
else:
str_list.append('/')
if path_args:
e1 = '?'
e2 = '&'
canonMap = {}
for _key, value in path_args.items():
canonMap[_key] = value
cannoList = sorted(canonMap.items(), key=lambda d: d[0])
for val in cannoList:
path_key, path_value = val[0], val[1]
flag = True
if path_key.lower(
) not in common_util.ALLOWED_RESOURCE_PARAMTER_NAMES:
flag = False
if flag:
path_key = common_util.encode_item(
common_util.toString(path_key), ' ,:?&%')
if path_value is None:
e1 += path_key + '&'
continue
e2 += path_key + '=' + common_util.toString(
path_value) + '&'
e = (e1 + e2).replace('&&', '&').replace('?&', '?')[:-1]
str_list.append(e)
return ''.join(str_list)