Python 非标准图像分析API失败，urllib2错误：名称解析暂时失败
这是face++网站上发布的。一个简单的python脚本接受图像URL并返回图像中人脸的面部分析Python 非标准图像分析API失败，urllib2错误：名称解析暂时失败,python,image,api,urllib2,Python,Image,Api,Urllib2,这是face++网站上发布的。一个简单的python脚本接受图像URL并返回图像中人脸的面部分析 API_KEY = '<your API key here>' API_SECRET = '<your API secret here>' # Import system libraries and define helper functions import time from pprint import pformat def print_result(hint, re
API_KEY = '<your API key here>'
API_SECRET = '<your API secret here>'
# Import system libraries and define helper functions
import time
from pprint import pformat
def print_result(hint, result):
def encode(obj):
if type(obj) is unicode:
return obj.encode('utf-8')
if type(obj) is dict:
return {encode(k): encode(v) for (k, v) in obj.iteritems()}
if type(obj) is list:
return [encode(i) for i in obj]
return obj
print hint
result = encode(result)
print '\n'.join([' ' + i for i in pformat(result, width = 75).split('\n')])
# First import the API class from the SDK
# 首先，导入SDK中的API类
from facepp import API
api = API(API_KEY, API_SECRET)
# Here are the person names and their face images
# 人名及其脸部图片
IMAGE_DIR = 'http://cn.faceplusplus.com/static/resources/python_demo/'
PERSONS = [
('Jim Parsons', IMAGE_DIR + '1.jpg'),
('Leonardo DiCaprio', IMAGE_DIR + '2.jpg'),
('Andy Liu', IMAGE_DIR + '3.jpg')
]
TARGET_IMAGE = IMAGE_DIR + '4.jpg'
# Step 1: Detect faces in the 3 pictures and find out their positions and
# attributes
# 步骤1：检测出三张输入图片中的Face，找出图片中Face的位置及属性
FACES = {name: api.detection.detect(url = url)
for name, url in PERSONS}
for name, face in FACES.iteritems():
print_result(name, face)
Jim Parsons
{'face': [{'attribute': {'age': {'range': 7, 'value': 33},
'gender': {'confidence': 99.94,
'value': 'Male'},
'race': {'confidence': 99.6939,
'value': 'White'},
'smiling': {'value': 2.71368}},
'face_id': '830ddcd5b598d5feb545a613e3cb5e59',
'position': {'center': {'x': 42.307692, 'y': 23.5},
'eye_left': {'x': 37.637557, 'y': 20.637333},
'eye_right': {'x': 46.111991, 'y': 19.983833},
'height': 14.0,
'mouth_left': {'x': 39.380317, 'y': 27.826667},
'mouth_right': {'x': 46.24457, 'y': 27.269333},
'nose': {'x': 41.688009, 'y': 24.5085},
'width': 19.004525},
'tag': ''}],
'img_height': 1220,
'img_id': '518b8d3aca3a0b4d43183fba4939c9eb',
'img_width': 900,
'session_id': '7137fea60f964e8cb10aec9526204185',
'url': 'http://cn.faceplusplus.com/static/resources/python_demo/1.jpg'}
Andy Liu
{'face': [{'attribute': {'age': {'range': 7, 'value': 40},
'gender': {'confidence': 99.9976,
'value': 'Male'},
'race': {'confidence': 80.2355,
'value': 'Asian'},
'smiling': {'value': 21.9477}},
'face_id': 'c38a15080defaabebea4efa5e4bd466b',
'position': {'center': {'x': 58.363636, 'y': 38.106796},
'eye_left': {'x': 50.705455, 'y': 29.711408},
'eye_right': {'x': 66.452, 'y': 30.17233},
'height': 42.718447,
'mouth_left': {'x': 52.394, 'y': 53.006068},
'mouth_right': {'x': 65.139091,
'y': 52.509951},
'nose': {'x': 56.065636, 'y': 43.470874},
'width': 32.0},
'tag': ''}],
'img_height': 412,
'img_id': 'bf382efa07150a240aaf85e70cba4537',
'img_width': 550,
'session_id': '2bcfaa61ffc047ffa493e169a124bc05',
'url': 'http://cn.faceplusplus.com/static/resources/python_demo/3.jpg'}
Leonardo DiCaprio
{'face': [{'attribute': {'age': {'range': 5, 'value': 24},
'gender': {'confidence': 99.9998,
'value': 'Male'},
'race': {'confidence': 99.5828,
'value': 'White'},
'smiling': {'value': 6.56214}},
'face_id': '6cc1d6261835ef58a9325de9cc9ca4e8',
'position': {'center': {'x': 42.75, 'y': 49.916388},
'eye_left': {'x': 32.551, 'y': 44.078428},
'eye_right': {'x': 51.022667, 'y': 42.1801},
'height': 38.628763,
'mouth_left': {'x': 36.497333, 'y': 61.53495},
'mouth_right': {'x': 50.798, 'y': 61.055017},
'nose': {'x': 42.608167, 'y': 53.22592},
'width': 38.5},
'tag': ''}],
'img_height': 598,
'img_id': '399ed42afa5637b3e28b5ff7bec36787',
'img_width': 600,
'session_id': '93c4e7d442e84b1fa2ed42f5767e08fe',
'url': 'http://cn.faceplusplus.com/static/resources/python_demo/2.jpg'}
Traceback (most recent call last):
File "hello.py", line 70, in <module>
for name, url in PERSONS}
File "hello.py", line 70, in <dictcomp>
for name, url in PERSONS}
File "C:\Users\ssh105\facepp-python-sdk-master\facepp.py", line 237, in __call__
raise APIError(e.code, url, e.read())
facepp.APIError: code=432
url=http://api.faceplusplus.com/detection/detect?url=http%3A%2F%2Forigincache-frc.fbcdn.net%2F12545241_601559039998519_1096073656_n.jpg&api_secret=XXX
XXXXX&api_key=XXXXXXX
{
"error": "IMAGE_ERROR_FAILED_TO_DOWNLOAD: Temporary failure in name resolution",
"error_code": 1302
}
回溯（最近一次呼叫最后一次）：
文件“hello.py”，第70行，在
对于名称，url为PERSONS}
文件“hello.py”，第70行，在
对于名称，url为PERSONS}
文件“C:\Users\ssh105\facepp python sdk master\facepp.py”，第237行，在调用中__
raise APIRROR（e.code、url、e.read（））
facepp.APIRROR:代码=432
网址=http://api.faceplusplus.com/detection/detect?url=http%3A%2F%2Forigincache-frc.fbcdn.net%2F12545241_601559039998519_1096073656_n.jpg&api_secret=XXX
XXXXX&api_key=XXXXXXX
{
“错误”：“图像错误下载失败：名称解析暂时失败”，
“错误代码”：1302
}
# -*- coding: utf-8 -*-
# $File: facepp.py
# $Date: Thu May 16 14:59:36 2013 +0800
# $Author: jiakai@megvii.com
#
# This program is free software. It comes without any warranty, to
# the extent permitted by applicable law. You can redistribute it
# and/or modify it under the terms of the Do What The Fuck You Want
# To Public License, Version 2, as published by Sam Hocevar. See
# http://sam.zoy.org/wtfpl/COPYING (copied as below) for more details.
#
# DO WHAT THE F*** YOU WANT TO PUBLIC LICENSE
# Version 2, December 2004
#
# Copyright (C) 2004 Sam Hocevar <sam@hocevar.net>
#
# Everyone is permitted to copy and distribute verbatim or modified
# copies of this license document, and changing it is allowed as long
# as the name is changed.
#
# DO WHAT THE F*** YOU WANT TO PUBLIC LICENSE
# TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
#
# 0. You just DO WHAT THE F*** YOU WANT TO.
"""a simple facepp sdk
example:
api = API(key, secret)
api.detection.detect(img = File('/tmp/test.jpg'))"""
__all__ = ['File', 'APIError', 'API']
DEBUG_LEVEL = 1
import sys
import socket
import urllib
import urllib2
import json
import os
import os.path
import itertools
import mimetools
import mimetypes
import time
import tempfile
from collections import Iterable
from cStringIO import StringIO
class File(object):
"""an object representing a local file"""
path = None
content = None
def __init__(self, path):
self.path = path
self._get_content()
def _resize_cv2(self, ftmp):
try:
import cv2
except ImportError:
return False
img = cv2.imread(self.path)
assert img is not None and img.size != 0, 'Invalid image'
bigdim = max(img.shape[0], img.shape[1])
downscale = max(1., bigdim / 600.)
img = cv2.resize(img,
(int(img.shape[1] / downscale),
int(img.shape[0] / downscale)))
cv2.imwrite(ftmp, img)
return True
def _resize_PIL(self, ftmp):
try:
import PIL.Image
except ImportError:
return False
img = PIL.Image.open(self.path)
bigdim = max(img.size[0], img.size[1])
downscale = max(1., bigdim / 600.)
img = img.resize(
(int(img.size[0] / downscale), int(img.size[1] / downscale)))
img.save(ftmp)
return True
def _get_content(self):
"""read image content; resize the image if necessary"""
if os.path.getsize(self.path) > 2 * 1024 * 1024:
ftmp = tempfile.NamedTemporaryFile(
suffix = '.jpg', delete = False).name
try:
if not (self._resize_cv2(ftmp) or self._resize_PIL(ftmp)):
raise APIError(-1, None, 'image file size too large')
with open(ftmp, 'rb') as f:
self.content = f.read()
finally:
os.unlink(ftmp)
else:
with open(self.path, 'rb') as f:
self.content = f.read()
def get_filename(self):
return os.path.basename(self.path)
class APIError(Exception):
code = None
"""HTTP status code"""
url = None
"""request URL"""
body = None
"""server response body; or detailed error information"""
def __init__(self, code, url, body):
self.code = code
self.url = url
self.body = body
def __str__(self):
return 'code={s.code}\nurl={s.url}\n{s.body}'.format(s = self)
__repr__ = __str__
class API(object):
key = None
secret = None
server = 'http://api.faceplusplus.com/'
decode_result = True
timeout = None
max_retries = None
retry_delay = None
def __init__(self, key, secret, srv = None,
decode_result = True, timeout = 30, max_retries = 10,
retry_delay = 5):
""":param srv: The API server address
:param decode_result: whether to json_decode the result
:param timeout: HTTP request timeout in seconds
:param max_retries: maximal number of retries after catching URL error
or socket error
:param retry_delay: time to sleep before retrying"""
self.key = key
self.secret = secret
if srv:
self.server = srv
self.decode_result = decode_result
assert timeout >= 0 or timeout is None
assert max_retries >= 0
self.timeout = timeout
self.max_retries = max_retries
self.retry_delay = retry_delay
_setup_apiobj(self, self, [])
def wait_async(self, session_id, referesh_interval = 2):
"""wait for asynchronous operations to complete"""
while True:
rst = self.info.get_session(session_id = session_id)
if rst['status'] != u'INQUEUE':
return rst
_print_debug(rst)
time.sleep(referesh_interval)
def update_request(self, request):
"""overwrite this function to update the request before sending it to
server"""
pass
def _setup_apiobj(self, api, path):
if self is not api:
self._api = api
self._urlbase = api.server + '/'.join(path)
lvl = len(path)
done = set()
for i in _APIS:
if len(i) <= lvl:
continue
cur = i[lvl]
if i[:lvl] == path and cur not in done:
done.add(cur)
setattr(self, cur, _APIProxy(api, i[:lvl + 1]))
class _APIProxy(object):
_api = None
"""underlying :class:`API` object"""
_urlbase = None
def __init__(self, api, path):
_setup_apiobj(self, api, path)
def __call__(self, post = False, *args, **kargs):
if len(args):
raise TypeError('Only keyword arguments are allowed')
if type(post) is not bool:
raise TypeError('post argument can only be True or False')
form = _MultiPartForm()
add_form = False
for (k, v) in kargs.iteritems():
if isinstance(v, File):
add_form = True
form.add_file(k, v.get_filename(), v.content)
if post:
url = self._urlbase
for k, v in self._mkarg(kargs).iteritems():
form.add_field(k, v)
add_form = True
else:
url = self.geturl(**kargs)
request = urllib2.Request(url)
if add_form:
body = str(form)
request.add_header('Content-type', form.get_content_type())
request.add_header('Content-length', str(len(body)))
request.add_data(body)
self._api.update_request(request)
retry = self._api.max_retries
while True:
retry -= 1
try:
ret = urllib2.urlopen(request, timeout = self._api.timeout).read()
break
except urllib2.HTTPError as e:
raise APIError(e.code, url, e.read())
except (socket.error, urllib2.URLError) as e:
if retry < 0:
raise e
_print_debug('caught error: {}; retrying'.format(e))
time.sleep(self._api.retry_delay)
if self._api.decode_result:
try:
ret = json.loads(ret)
except:
raise APIError(-1, url, 'json decode error, value={0!r}'.format(ret))
return ret
def _mkarg(self, kargs):
"""change the argument list (encode value, add api key/secret)
:return: the new argument list"""
def enc(x):
if isinstance(x, unicode):
return x.encode('utf-8')
return str(x)
kargs = kargs.copy()
kargs['api_key'] = self._api.key
kargs['api_secret'] = self._api.secret
for (k, v) in kargs.items():
if isinstance(v, Iterable) and not isinstance(v, basestring):
kargs[k] = ','.join([enc(i) for i in v])
elif isinstance(v, File) or v is None:
del kargs[k]
else:
kargs[k] = enc(v)
return kargs
def geturl(self, **kargs):
"""return the request url"""
return self._urlbase + '?' + urllib.urlencode(self._mkarg(kargs))
def visit(self, browser = 'chromium', **kargs):
"""visit the url in browser"""
os.system('{0} "{1}"'.format(browser, self.geturl(**kargs)))
# ref: http://www.doughellmann.com/PyMOTW/urllib2/
class _MultiPartForm(object):
"""Accumulate the data to be used when posting a form."""
def __init__(self):
self.form_fields = []
self.files = []
self.boundary = mimetools.choose_boundary()
return
def get_content_type(self):
return 'multipart/form-data; boundary=%s' % self.boundary
def add_field(self, name, value):
"""Add a simple field to the form data."""
self.form_fields.append((name, value))
return
def add_file(self, fieldname, filename, content, mimetype = None):
"""Add a file to be uploaded."""
if mimetype is None:
mimetype = mimetypes.guess_type(filename)[0] or 'application/octet-stream'
self.files.append((fieldname, filename, mimetype, content))
return
def __str__(self):
"""Return a string representing the form data, including attached files."""
# Build a list of lists, each containing "lines" of the
# request. Each part is separated by a boundary string.
# Once the list is built, return a string where each
# line is separated by '\r\n'.
parts = []
part_boundary = '--' + self.boundary
# Add the form fields
parts.extend(
[ part_boundary,
'Content-Disposition: form-data; name="%s"' % name,
'',
value,
]
for name, value in self.form_fields
)
# Add the files to upload
parts.extend(
[ part_boundary,
'Content-Disposition: file; name="%s"; filename="%s"' % \
(field_name, filename),
'Content-Type: %s' % content_type,
'',
body,
]
for field_name, filename, content_type, body in self.files
)
# Flatten the list and add closing boundary marker,
# then return CR+LF separated data
flattened = list(itertools.chain(*parts))
flattened.append('--' + self.boundary + '--')
flattened.append('')
return '\r\n'.join(flattened)
def _print_debug(msg):
if DEBUG_LEVEL:
sys.stderr.write(str(msg) + '\n')
_APIS = [
'/detection/detect',
'/detection/landmark',
'/faceset/add_face',
'/faceset/create',
'/faceset/delete',
'/faceset/get_info',
'/faceset/remove_face',
'/faceset/set_info',
'/group/add_person',
'/group/create',
'/group/delete',
'/group/get_info',
'/group/remove_person',
'/group/set_info',
'/grouping/grouping',
'/info/get_app',
'/info/get_face',
'/info/get_faceset_list',
'/info/get_group_list',
'/info/get_image',
'/info/get_person_list',
'/info/get_quota',
'/info/get_session',
'/person/add_face',
'/person/create',
'/person/delete',
'/person/get_info',
'/person/remove_face',
'/person/set_info',
'/recognition/compare',
'/recognition/group_search',
'/recognition/identify',
'/recognition/recognize',
'/recognition/search',
'/recognition/test_train',
'/recognition/train',
'/recognition/verify',
'/train/group_search',
'/train/identify',
'/train/recognize',
'/train/search',
'/train/verify'
]
_APIS = [i.split('/')[1:] for i in _APIS]
#-*-编码：utf-8-*-
#$File:facepp.py
#$Date:Thu May 16 14:59:36 2013+0800
#$Author:jiakai@megvii.com
#
#这个程序是自由软件。对我来说，它没有任何保证
#适用法律允许的范围。你可以重新分配它
#和/或根据《做他妈的你想做的事》条款修改它
#由Sam Hocevar发布的公共许可证第2版。看见
# http://sam.zoy.org/wtfpl/COPYING （复制如下）了解更多详细信息。
#
#做你想公开许可的事情
#第2版，2004年12月
#
#版权所有（C）2004 Sam Hocevar
#
#每个人都被允许逐字复制和分发或修改
#此许可证文档的副本，并允许在
#随着名字的改变。
#
#做你想公开许可的事情
#复制、分发和修改的条款和条件
#
# 0. 你想干什么就干什么。
“”“一个简单的facepp sdk
例子：
api=api（密钥、机密）
api.detection.detect（img=File（'/tmp/test.jpg'））“”“
__全部\uuu8=['File'，'apirerror'，'API']
调试_级别=1
导入系统
导入套接字
导入URL库
导入urllib2
导入json
导入操作系统
导入操作系统路径
进口itertools
进口美托尔
导入模拟类型
导入时间
导入临时文件
从集合导入Iterable
从cStringIO导入StringIO
类文件（对象）：
“”“表示本地文件的对象”“”
路径=无
内容=无
定义初始化（自身，路径）：
self.path=path
self.\u获取内容（）
def _resize_cv2（自身，ftmp）：
尝试：
进口cv2
除恐怖外：
返回错误
img=cv2.imread（self.path）
断言img不是None且img.size！=0，“无效图像”
bigdim=max（img.shape[0]，img.shape[1]）
缩尺=最大值（1.，大尺寸/600）
img=cv2。调整大小（img，
（内部（图像形状[1]/缩小比例），
int（图像形状[0]/缩小比例）））
cv2.imwrite（ftmp，img）
返回真值
def _resize_PIL（自身，ftmp）：
尝试：
导入PIL.Image
除恐怖外：
返回错误
img=PIL.Image.open（self.path）
bigdim=max（img.size[0]，img.size[1]）
缩尺=最大值（1.，大尺寸/600）
img=img.resize(
（int（img.size[0]/缩小比例），int（img.size[1]/缩小比例）））
图像保存（ftmp）
返回真值
定义获取内容（自我）：
“”“读取图像内容；必要时调整图像大小”“”
如果os.path.getsize（self.path）>2*1024*1024：
ftmp=tempfile.NamedTemporaryFile(
后缀='.jpg'，delete=False）.name
尝试：
如果不是（自调整大小cv2（ftmp）或自调整大小PIL（ftmp））：
raise APIRROR（-1，无，“图像文件大小太大”）
将open（ftmp，'rb'）作为f：
self.content=f.read（）
最后：
操作系统取消链接（ftmp）
其他：
将open（self.path，'rb'）作为f：
self.content=f.read（）
def get_文件名（自身）：
返回os.path.basename（self.path）
类错误（异常）：
代码=无
“”“HTTP状态代码”“”
url=无
“请求URL”
正文=无
“”“服务器响应正文；或详细错误信息”“”
定义初始化（self、代码、url、正文）：
self.code=代码
self.url=url
self.body=身体
定义（自我）：
返回'code={s.code}\nurl={s.url}\n{s.body}'。格式（s=self）
__repr\uuuuu=\uuuuuu str__
类API（对象）：
键=无
秘密=无
服务器http://api.faceplusplus.com/'
解码结果=真
超时=无
最大重试次数=无
重试\u延迟=无
def u u init u;（self，key，secret，srv=None，
解码结果=True，超时=30，最大重试次数=10，
重试（延迟=5）：
“”：参数srv:API服务器地址
：param decode_result：是否对结果进行json_解码
：param timeout:HTTP请求超时（秒）
：param max_retries：捕获URL错误后的最大重试次数
或套接字错误
：param retry\u delay:重试前的睡眠时间“”
self.key=key
self.secret=秘密
如果srv：
self.server=srv
self.decode_result=解码_result
断言超时>=0或超时为无
断言最大重试次数>=0
self.timeout=超时
self.max\u retries=最大重试次数
self.retry\u delay=重试\u delay
_设置\u apiobj（self，self，[]））
def wait_ASSYNC（自身、会话id、引用间隔=