def initializeUserDict(tag, count = 5):
usr_dict = {}
for most_popular in get_popular(tag = tag)[0 : count]:
for person in get_urlposts(most_popular['href']):
usr = person['user']
usr_dict[usr] = {}
return usr_dict
def build_tag_matrix(tags):
"""
Given a list of bookmarks returns a matrix with
tags as columns and urls as rows.
Returns: tags, urls, matrix
"""
bookmarks = []
for tag in tags:
marks = get_popular(tag=tag)
for bookmark in marks:
url = bookmark["url"]
tags = bookmark["tags"]
bookmarks.append({"url": url, "tags": tags})
tag_list = set(tag for tag in bookmark["tags"] for bookmark in bookmarks)
url_list = []
matrix = []
for bookmark in bookmarks:
row = []
for tag in tag_list:
row.append(1 if tag in bookmark["tags"] else 0)
url_list.append(bookmark["url"])
matrix.append(row)
return tag_list, url_list, matrix
def build_tag_matrix(tags):
"""
Given a list of bookmarks returns a matrix with
tags as columns and urls as rows.
Returns: tags, urls, matrix
"""
bookmarks = []
for tag in tags:
marks = get_popular(tag=tag)
for bookmark in marks:
url = bookmark["url"]
tags = bookmark["tags"]
bookmarks.append({"url": url, "tags": tags})
tag_list = set(tag for tag in bookmark["tags"] for bookmark in bookmarks)
url_list = []
matrix = []
for bookmark in bookmarks:
row = []
for tag in tag_list:
row.append(1 if tag in bookmark["tags"] else 0)
url_list.append(bookmark["url"])
matrix.append(row)
return tag_list, url_list, matrix
def initialize(tag,count=5):
user_dict = {}
for p1 in get_popular(tag=tag)[0:count]:
for p2 in get_urlposts(p1['url']):
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag, count=5):
user_dict = {}
for p1 in get_popular(tag=tag)[0:count]:
for p2 in get_userposts(p1['href']):
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag, count=5):
user_dict = {}
for p1 in get_popular(tag=tag)[0:count]:
for p2 in get_urlposts(p1["url"]):
user = p2["user"]
user_dict[user] = {}
return user_dict
def initializeUserDict(tag,count=5):
user_dict = {}
all_items = {}
# get the top count' popular posts
for p1 in get_popular(tag=tag)[0:count]:
# find all users who posted this
for p2 in get_urlposts(p1['href']):
user = p2['user']
user_dict.setdefault(user,{})
# get user posts
for i in range(3):
try:
posts = get_userposts(user)
break
except:
print "Failed user "+user+", retrying"
time.sleep(4)
for post in posts:
url = post['href']
user_dict[user][url] = 1.0
all_items[url] = 1
#fill in missing items with 0
[ratings.setdefault(item,0.0) for item in all_items for ratings in user_dict.values()]
return user_dict
def initializeUserDict(tag ,count = 5):
user_dict = {}
#get the top count popular posts
for p1 in get_popular(tag = tag)[0:count]:
for p2 in get_urlposts(p1['url']):
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag, count=5):
user_dict={}
# 获取前count个最受欢迎的张
for p1 in get_popular(tag = tag)[0:count]:
for p2 in get_urlposts(p1['href']):
user=p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag, count=5):
user_dict = {}
for p1 in get_popular(tag=tag)[0:count]:
for p2 in get_urlposts(p1['url']):
user = p2['user']
if len(user) > 0:
user_dict[user] = {}
return user_dict
def initialze_user_dict(tag, count=5):
user_dict = {}
# 获取前count个最受欢迎的链接张贴记录
for p1 in get_popular(tag=tag)[0:count]:
for p2 in get_urlposts(p1['url']): # 此处与书中不同，现在返回字典的'href' 键已经改为 'url'
user = p2['user']
user_dict[user] = {}
return user_dict
def initlizeUserDict(tag, count = 5):
user_dict = {}
#获取所有用户都提交过的链接
for p1 in get_popular(tag=tag)[0:count]:
for p2 in get_urlposts(p1['href']):
user = p2['user']
user_dict[user]={}
return user_dict
def initialize_user_dict(tag,count=5):
user_dict={}
# Get the top "count" popular posts
for p1 in pydelicious.get_popular(tag=tag)[0:count]:
# Find all users who posted the same post url
for p2 in pydelicious.get_urlposts(p1['url']):
user=p2['user']
user_dict[user]={}
return user_dict
def initlizeUserDict(tag, count=5):
user_dict = {}
#获取所有用户都提交过的链接
for p1 in get_popular(tag=tag)[0:count]:
for p2 in get_urlposts(p1['href']):
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag,count=5):
user_dict={}
#get the top "count" popular posts
for p1 in get_popular(tag=tag)[0:count]: #from api
#find all users who posted this
for p2 in get_urlposts(p1['href']):
user=p2['user']
user_dict[user]={} #looking at a key:value pair?
return user_dict
def initialize_user_dict(tag, count=5):
user_dict = {}
# Get the top count popular post
for p1 in get_popular(tag=tag)[0:count]:
# Find all users who posted this
for p2 in get_urlposts(p1['url']):
user=p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag, count=5):
"""docstring for initializeUserDict"""
user_dict = {}
for p1 in get_popular(tag=tag)[0:count]:
for p2 in get_urlposts(p1['url']):
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag, count=5):
user_dict = {}
# отримати рахунок найпопулярніших лінків
for p1 in get_popular(tag=tag)[0:count]:
# знайти всіх користувачів, які зберегли цей лінк
for p2 in get_urlposts(p1['href']):
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag,count=5):
user_dict={}
#获取前count个最受欢迎的链接张贴记录
for p1 in get_popular(tag=tag)[0:count]:
#查看所有张贴该链接的用户
for p2 in get_urlposts(p1['herf']):
user=p2['user']
user_dict[user]={}
return user_dict
def initializeUserDict(tag,count=5):
user_dict={}
# get the top count' popular posts
for p1 in get_popular(tag=tag)[0:count]:
# find all users who posted this
for p2 in get_urlposts(p1['url']):
user=p2['user']
user_dict[user]={}
return user_dict
def initializeUserDict(tag,count=2):
user_dict={}
# get the top count' popular posts
for p1 in get_popular(tag=tag)[0:count]:
# find all users who posted this
for p2 in get_urlposts(p1['href']):
user=p2['user']
user_dict[user]={}
return user_dict
def initializeUserDict(tag, count=5):
user_dict = {}
# get top "count" of popular posts:
for p1 in get_popular(tag=tag)[0:count]:
# Find the users who poseted them:
for p2 in get_urlposts(p1['href']):
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag,count=5):
user_dict={}
# count개의 인기 게시물을 얻음
for p1 in get_popular(tag=tag)[0:count]:
# 이 게시물을 올린 모든 사용자를 얻음
for p2 in get_urlposts(p1['href']):
user=p2['user']
user_dict[user]={}
return user_dict
def initializeUserDict(tag, count=5):
user_dict = {}
# 获取某tag下前count个最受欢迎的url post记录
for p1 in get_popular(tag=tag)[0:count]:
# 获取所有post该url的用户
for p2 in get_urlposts(p1['href']):
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag, count=5):
user_dict={}
for p1 in get_popular(tag=tag)[0:count]:
# get_urlposts()는 주어진 url에 대한 모든 게시글을 리턴
for p2 in get_urlposts(p1['url']):
# 리턴된 게시글들의 '유저이름'을 가져옴. '유저이름'의 value는 일단 null로 둠
user=p2['user']
user_dict[user]={}
return user_dict
def initializeUserDict(tag,count = 5):
user_dict = {}
#获取前count个最受欢迎的链接张贴记录
for p1 in get_popular(tag=tag)[0:count]:
#查找所有张贴记录的用户
for p2 in get_urlposts(p1['url']):
user = p2['user']
user_dict[user] = {} #得到一个包含用户的空字典
return user_dict
def initializeUserDict(tag, count = 5):
user_dict = {}
# read in top "count" popular posts
for p1 in get_popular(tag = tag)[0:count]:
#find all users who posted this
for p2 in get_urlposts(p1['url']):
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag, count=5):
user_dict = {}
# count개의 인기 게시물을 얻음
for p1 in get_popular(tag=tag)[0:count]:
# 이 게시물을 올린 모든 사용자를 얻음
for p2 in get_urlposts(p1['href']):
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag, count=5):
user_dict={}
# отримати рахунок найпопулярніших лінків
for p1 in get_popular(tag=tag)[0:count]:
# знайти всіх користувачів, які зберегли цей лінк
for p2 in get_urlposts(p1['href']):
user=p2['user']
user_dict[user]={}
return user_dict
def initializeUserDict(tag, count=5):
user_dict = {}
#获取当前count个最受欢迎的链接张贴记录
for p1 in get_popular(tag=tag)[0:count]:
#查找所有张贴该链接的用户
for p2 in get_urlposts(p1['href']):
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag, count=30): # 引数に tag と count をとる、count の指定がなければ 5
user_dict = {}
# popular な投稿を 0 から count 番目まで取得
for p1 in get_popular(tag=tag)[0:count]:
# このリンクを投稿したすべてのユーザを取得
for p2 in get_urlposts(p1['href']):
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag, count=5):
"""Grabs the user ids of active del.icio.us users."""
user_dict = {}
# get top `count` popular posts...
for p1 in get_popular(tag=tag)[0:count]:
# ...and add all users who posted them
for p2 in get_urlposts(p1['href']):
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag, count=5):
user_dict = {}
for p1 in get_popular(tag=tag)[0:count]:
#find all users that posted this
for p2 in get_urlposts(p1['href']):
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag, count=5):
user_dict={}
#get count most popular links
for p1 in get_popular(tag=tag)[0:count]:
#find all users saved this link
for p2 in get_urlposts(p1['url']):
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag, count=5):
"""Grabs the user ids of active del.icio.us users."""
user_dict = {}
# get top `count` popular posts...
for p1 in get_popular(tag=tag)[0:count]:
# ...and add all users who posted them
for p2 in get_urlposts(p1['href']):
user = p2['user']
user_dict[user] = {}
return user_dict
def _init_user_dict(self, count=5):
user_dict = {}
for p1 in pydelicious.get_popular(tag=self._tag)[0:count]:
for p2 in pydelicious.get_urlposts(p1['url']):
user = p2['user']
user_dict[user] = {}
if __debug__: print(user_dict)
self._usr_dict = user_dict
def initialize_user_dict(tag, count=5):
"""
Returns: length n dict of users: empty dict pairs.
"""
user_dict = {}
for post in get_popular(tag=tag)[:count]:
for url in get_urlposts(post['url']):
user = url['user']
user_dict[user] = {}
return user_dict
def _init_user_dict(self, count=5):
user_dict = {}
for p1 in pydelicious.get_popular(tag=self._tag)[0:count]:
for p2 in pydelicious.get_urlposts(p1['url']):
user = p2['user']
user_dict[user] = {}
if __debug__: print(user_dict)
self._usr_dict = user_dict
def initializeUserDict(tag, count=5):
user_dict={}
# popularな投稿をcount番目まで取得
for p1 in get_popular(tag=tag)[0:count]:
# このリンクを投稿したすべてのユーザを取得
for p2 in get_urlposts(p1['href']):
user = p2['user']
user_dict[user]={}
return user_dict
def initialiseUserDict(tag,count=5):
user_dict={}
for p1 in pydelicious.get_popular(tag=tag):
print p1
for p2 in pydelicious.get_urlposts(p1['url'])[0:count]:
print p2
user = p2['user']
if len(user):
user_dict[user]={}
return user_dict
def initializeUserDict(tag, count = 5):
'''Get a list of users who recently posted a popular link.'''
user_dict = {}
# get the top count popular posts
for p1 in get_popular(tag = tag)[0:count]:
# find all users who posted this
for p2 in get_urlposts(p1['href']):
user = p2['user']
user_dict.setdefault(user, {})
return user_dict
def initUserDict(tag, count = 5):
user_dict = {}
# get the top count popular posts
for p1 in pydelicious.get_popular(tag=tag)[0:count]:
# find all users who posted this
# print pydelicious.get_urlposts(p1['url'])
for p2 in pydelicious.get_urlposts(p1['url']):
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag,count=5):
user_dict={}
# get popular blog posts
for p1 in get_popular(tag=tag)[0:count]:
for p2 in get_urlposts(p1['href']):
user=p2['user']
if user:
user_dict[user]={}
return user_dict
def initUserDict(tag, n):
user_dict={}
#Fåtop count' populära posts
for p1 in get_popular(tag)[0:n]:
#Hitta alla användare som postade den
for p2 in get_urlposts(p1['href']):
user=p2['user']
user_dict[user]={}
return user_dict
def initializeUserDict(tag, count=5):
suer_dict = {}
a = get_popular(tag=tag)[0:count]
print a
for p1 in a:
# Find all user who posted it
for p2 in get_urlposts(p1['href']):
user = p2['user']
user_dict[user] = {}
return user_dict
def initialize_object_dict(self, tag, count=5):
"""
Initialize object (user) dict with users from
the popular posts for tag
"""
for popular in get_popular(tag=tag)[0:count]:
for posts in get_urlposts(popular['url']):
object = posts['user']
if self.verbose: print object
self.object_dict[object] = dict()
def initializeUserDict(tag, count=5):
user_dict = {}
# get the top count' popular posts
for p1 in get_popular(tag=tag)[0:count]:
for p2 in get_urlposts(p1['url']):
user = p2['user']
#print(p1['url'] + ' ' + user)
user_dict[user] = {}
return user_dict
def initializeUserDictByTag(tag, count=5):
user_dict = {}
# 获取前count个最受欢迎的链接张贴记录
for p1 in pydelicious.get_popular(tag=tag)[0:count] :
# 查找所有张贴该链接的用户
for p2 in pydelicious.get_urlposts(p1['url']) :
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag, count=5):
user_dict = {}
for p1 in get_popular(tag=tag)[0:count]:
print(p1)
if (not p1["href"]):
break
for p2 in get_urlposts(p1["href"]):
user = p2['user']
user_dict[user] = {}
return user_dict
def initializeUserDict(tag, count=5):
user_dict = {}
try:
# get the top count popular posts
for p1 in get_popular(tag=tag)[0:count]:
# Find all users who posted this
for p2 in get_urlposts(p1['href']):
user = p2['user']
user_dict[user] = {}
#
except Exception:
print('function')
return user_dict
def initializeUserDict(tag='programming', count=1):
user_dict = {}
# get the top count' popular posts
for p1 in get_popular(tag=tag)[0:count]:
# find all users who posted this
print "working on count", count
pst = get_urlposts(p1['url'])
print "pst len", len(pst)
for p2 in pst:
user = p2['user']
user_dict[user] = {}
print "user=" + user
return user_dict
def initializeUserDict(tag, count=5):
user_dict = {}
#获取前count个最受欢迎的张贴记录
for p1 in get_popular(tag=tag)[0:count]:
print "Got a post !"
#查找所有张贴该记录的用户
#原来的API已经改了，不用href，而是url
for i in range(3):
try:
for p2 in get_urlposts(p1['url']):
user = p2['user']
user_dict[user] = {}
break
except:
print "Failed....retrying"
time.sleep(4)
return user_dict
def build_tag_dict(tags):
all_items = {}
tags_dict = {}
# Find links posted by all get_user
for tag in tags:
bookmarks = get_popular(tag=tag)
for bookmark in bookmarks:
url = bookmark["url"]
for tag in bookmark["tags"]:
tags_dict.setdefault(tag, {})
tags_dict[tag][url] = 1.0
all_items[url] = 1
# Fill in missing items with zero.
for ratings in tags_dict.values():
for item in all_items:
if item not in ratings:
ratings[item] = 0.0
return tags_dict
import pydelicious
from urllib import urlencode
from urllib2 import urlopen
# import get_userposts,get_popular,get_urlposts
pydelicious.get_popular(tag="programming")
# pydelicious.get_userposts('dorsia')
# def initializeUserDict(tag,count=5):
# user_dict={}
# for p1 in get_popular(tag=tag)[0:count]:
# for p2 in get_urlposts(p1['href']):
# user = p2['user']
# user_dict[user] = {}
# return user_dict
def cls():
if os.name == 'posix':
os.system('clear')
else:
os.system('cls')
cls()
# como o módulo pydelicious pode não estar instalado coloquei este try/except
try:
import pydelicious
except ImportError:
print(" ")
print(" você tem que instalar primeiro a biblioteca pydelicious")
print("")
print(" sudo aptitude install python-setuptools")
print(
" sudo easy_install http://pydelicious.googlecode.com/files/pydelicious-0.6.zip"
)
print(" ")
sys.exit(1)
if len(sys.argv) != 2:
print('Uso: %s tag' % sys.argv[0])
sys.exit(1)
populares = pydelicious.get_popular(tag=str(sys.argv[1]))
# print("\n".join([ x.get('url') for x in populares] ))
print(populares)
def initUserDict(tag, count=5):
user_dict = {}
for post in get_popular(tag=tag)[0:count]:
for post2 in get_urlposts(post['url']):
user_dict[post2['user']] = {}
return user_dict
time.sleep(4)
for post in posts:
url = post['url']
user_dict[user][url] = 1.0
all_items[url] = 1
#用0填充缺失的项（这一步必要吗？？）
for ratings in user_dict.values():
for item in all_items:
if item not in ratings:
ratings[item] = 0.0
#全局测试代码
if 0:
pl = get_popular('programming')
url = pl[0]['url']
for item in pl:
url = item['url']
print url
for item in pl:
url = item['url']
print url
print get_urlposts(url)
print get_urlposts('http://www.holub.com/goodies/uml/')
#print url
# print get_urlposts(url)
#全局测试代码二
if 0:
delusers = initializeUserDict('programming')
from pydelicious import get_popular, get_urlposts, get_userposts print(get_popular(tag="programing"))
#!/usr/bin/env python # -*- coding: utf-8 -*- # @Time : 2017/2/26 16:19 # @Author : Shuqi.qin # @Site : # @File : test.py # @Software: PyCharm import recommendations as rc res1 = rc.sim_distance(rc.critics, "Lisa Rose", "Gene Seymour") res2 = rc.sim_pearson(rc.critics, "Lisa Rose", "Gene Seymour") res3 = rc.topMatches(rc.critics, "Toby", n=3) res4 = rc.getRecommendationsBaseOnPersonSimilarity(rc.critics, "Toby") print 'over' import pydelicious print pydelicious.get_popular(tag='programming')
