Python错误：内置函数或方法对象没有属性'StringIO

Question

我只是想下载一张图片，然后把它上传到亚马逊的S3上。但现在不太顺利。

'builtin_function_or_method' object has no attribute 'StringIO'
Traceback (most recent call last):
  File "flickrDump.py", line 16, in <module>
    imgpath = s3.upload_thumbnail(thumbnail_name=tools.randomString(10), thumbnail_data=tdata,bucket="fabletest")
  File "../lib/s3.py", line 52, in upload_thumbnail
    k.set_contents_from_string(thumbnail_data)
  File "/usr/lib/pymodules/python2.6/boto/s3/key.py", line 539, in set_contents_from_string
    self.set_contents_from_file(fp, headers, replace, cb, num_cb, policy)
  File "/usr/lib/pymodules/python2.6/boto/s3/key.py", line 455, in set_contents_from_file
    self.send_file(fp, headers, cb, num_cb)
  File "/usr/lib/pymodules/python2.6/boto/s3/key.py", line 366, in send_file
    return self.bucket.connection.make_request('PUT', self.bucket.name,
AttributeError: 'str' object has no attribute 'connection'

我用来下载和上传的代码是这样的：

tdata = tools.download("http://farm5.static.flickr.com/4148/5124630813_c11b05e6da_z.jpg")
imgpath = s3.upload_thumbnail(thumbnail_name=tools.randomString(10), thumbnail_data=tdata,bucket="fabletest")
print imgpath

我使用的库是s3库。我是在某个地方下载的，所以应该是标准的。

from boto.s3.connection import S3Connection
from boto.s3.key import Key
from boto.s3.bucket import Bucket
import datetime

ACCESSKEY = 'MYKEY'
SECRETKEY = 'MYSECRET'


def get_bucket_path(bucket,filename,https=False):
    path = None
    if isinstance(bucket, Bucket):
        path = bucket.name
    else:
        path = bucket

    if https:
        return "https://s3.amazonaws.com/%s/%s" % (path, filename)
    else:
        return "http://s3.amazonaws.com/%s/%s" % (path, filename)

def _aws_keys():
    return ACCESSKEY, SECRETKEY

def _conn():
    key,secret = _aws_keys()
    return S3Connection(key,secret)

def cache_bucket(conn = _conn()):
    bucket = conn.create_bucket('mimvicache')    bucket.make_public()
    return bucket

class AwsException(Exception):
    def __init__(self,value):
        self.errorval = value
    def __str__(self):
        return repr(self.errorval)

def upload_thumbnail(thumbnail_name,thumbnail_data=None,thumbnail_path=None,bucket=cache_bucket
(),conn=_conn(),notes=None,image_id=None):
    k = Key(bucket)
    k.key = thumbnail_name

    if notes is not None:
        k.set_metadata("notes",notes)

    if image_id is not None:
        k.set_metadata("image_id",image_id)


    if thumbnail_data is not None:
        k.set_contents_from_string(thumbnail_data)
    elif thumbnail_path is not None:
        k.set_contents_from_filename(thumbnail_path)
    else:
        raise AwsException("No file name")

    k.set_acl('public-read')

    return get_bucket_path(bucket.name,k.key)

有没有人能帮我把这张图片上传到S3上？

Answer 1

在你的代码中：

return self.bucket.connection.make_request('PUT', self.bucket.name,......
AttributeError: 'str' object has no attribute 'connection'

这意味着 self.bucket 被当作字符串来处理，而你显然不能在字符串上调用 "connection" 这个方法。

所以为了进一步分析，看看 upload_thumbnail 这个函数，它期望接收 bucket=cache_bucket() 作为参数。也就是说，它需要一个 bucket 对象。

def upload_thumbnail(thumbnail_name,thumbnail_data=None,thumbnail_path=None,bucket=cache_bucket
(),conn=_conn(),notes=None,image_id=None)

而你在代码中传入的是字符串！！ -> (bucket="fabletest")

imgpath = s3.upload_thumbnail(thumbnail_name=tools.randomString(10), thumbnail_data=tdata,bucket="fabletest")

你的代码应该像这样。你可能需要对其进行一些清理。但关键是要把 bucket 和 connection 对象传递给 upload_thumbnail 函数。

import S3
connection = S3.AWSAuthConnection('your access key', 'your secret key')
buck = connection.create_bucket('mybucketname')
tdata = tools.download("http://farm5.static.flickr.com/4148/5124630813_c11b05e6da_z.jpg")
imgpath = s3.upload_thumbnail(thumbnail_name=tools.randomString(10), thumbnail_data=tdata,bucket=buck, conn=connection)
print imgpath