# DNS缓存

def _getaddrinfo(*args, **kwargs):

if args in _dnscache:

# print('使用了dns缓存:%s' % _dnscache[args])

return _dnscache[args]

else:

_dnscache[args] = socket._getaddrinfo(*args, **kwargs)

# print('添加dns缓存:%s' % _dnscache[args])

return _dnscache[args]

if not hasattr(socket, '_getaddrinfo'):

socket._getaddrinfo = socket.getaddrinfo

# 浏览器请求头

header = {

'user-agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.103 Safari/537.36'}

# 登录网址

posturl = 'https://accounts.pixiv.net/api/login?lang=zh'

# 主页

baseurl = 'https://www.pixiv.net'

# 创建session

psession = requests.session()

downloadsession = requests.session()

downloadsession.keep_alive = False

# 创建一个保存作者id的集合

id_set = set()

userid = ''

def __init__(self):

return

# self.username = username

# self.password = password

# self.userid = self.login()

# self.filepath = 'Author/%s/' % self.userid

# 模拟登录

def login(self, username, password):

try:

loginkey = self.psession.get('https://accounts.pixiv.net/login', headers=self.header)

post_key = re.search('name="post_key".*?value="(.*?)"', loginkey.text).group(1)

postdata = {

'password': password,

'pixiv_id': username,

'post_key': post_key,

}

login_response = self.psession.post(self.posturl, headers=self.header, data=postdata)

userid = re.search('PHPSESSID=(.*?)_', str(self.psession.cookies))

if not userid:

return False

self.userid = userid.group(1)

# print(self.userid)

return True

except ConnectionError:

return self.login(username, password)

# 获取关注的所有画师id并存入id_set

def member(self, url='https://www.pixiv.net/bookmark.php?type=user'):

pixivresponse = self.psession.get(url, headers=self.header)

soup = BeautifulSoup(pixivresponse.text, 'lxml')

cla = soup.find(class_='members')

pages = soup.find(class_='_pager-complex')

members = cla.find_all('li')

for i in members:

id = re.search('href=".*?=(.*?)"', str(i)).group(1)

self.id_set.add(id)

# print(id)

if pages:

print(str(pages))

nextpage = re.search('<a class="button" href="(.*?)" rel="next">', str(pages))

if nextpage:

print(nextpage.group(1))

self.member(url='https://www.pixiv.net/bookmark.php' + nextpage.group(1).replace('&amp;', '&'))

return

# 根据画师id获取该画师的所有作品id

def member_id(self, id='4754550'):

ajax_url = 'https://www.pixiv.net/ajax/user/' + id + '/profile/all'

id_url = 'https://www.pixiv.net/member_illust.php?id=' + id

json_file = self.psession.get(ajax_url, headers=self.header)

# print(memberresponse.text)

id_dir = json_file.json()['body']['illusts']

for id in id_dir.keys():

self.save_id(id=id, filepath=2)

# 保存收藏id

def collection(self, id=False):

filepath = 'Author/收藏/%s/' % self.userid

bookmarks_url = 'https://www.pixiv.net/ajax/user/%s/illusts/bookmarks?tag=&offset=0&limit=9999&rest=show' % self.userid

if id:

filepath = 'Author/收藏/%s/' % id

bookmarks_url = 'https://www.pixiv.net/ajax/user/%s/illusts/bookmarks?tag=&offset=0&limit=9999&rest=show' % id

colresponse = self.psession.get(bookmarks_url, headers=self.header).json()

# print(colresponse)

for i in colresponse['body']['works']:

self.save_id(id=i['id'], filepath=filepath)

# 每日推荐

def discovery(self, num):

try:

headers = {

'Referer': 'https://www.pixiv.net/discovery',

'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.100 Safari/537.36'

}

filepath = 'Author/推荐/'

discovery_url = 'https://www.pixiv.net/rpc/recommender.php?type=illust&sample_illusts=auto&num_recommendations=1000&page=discovery&mode=all'

disresponse = self.psession.get(discovery_url, headers=headers).json()

# print(disresponse)

# print(len(disresponse['recommendations']))

for i in range(0, num):

id = disresponse['recommendations'][i]

self.save_id(id=id, filepath=filepath)

except ConnectionError:

return self.discovery(num=num)

# 下载排行

def rank(self, mode='daily', date=None, num=0):

jsonurl = 'https://www.pixiv.net/ranking.php?mode=%s&date=%s&format=json' % (mode, date)

filepath = 'Author/排行/%s/%s/' % (mode, date)

jsonres = self.psession.get(jsonurl, headers=self.header)

if jsonres.status_code == 200:

rankdate = jsonres.json()

# print(len(rankdate['contents']))

for i in range(0, num):

id = rankdate['contents'][i]['illust_id']

self.save_id(id=id, filepath=filepath)

# 根据作品id得到图片url和Referer

def save_id(self, id=None, filepath=None):

try:

_setdnscache()

id_url = 'https://www.pixiv.net/member_illust.php?mode=medium&illust_id=' + str(id)

more_url = 'https://www.pixiv.net/member_illust.php?mode=manga&illust_id=' + str(id)

id_ajax = 'https://www.pixiv.net/ajax/illust/' + str(id)

idresponse = self.psession.get(id_ajax, headers=self.header)

img = idresponse.json()['body']['urls']['original']

imgnum = idresponse.json()['body']['pageCount']

imgtype = idresponse.json()['body']['illustType']

if filepath == None:

filepath = 'Author/ids/'

elif filepath == 2:

title = idresponse.json()['body']['userName']

filepath = 'Author/members/%s/' % title

if not os.path.exists(filepath):

os.makedirs(filepath)

if imgnum > 1:

for i in range(0, imgnum):

newimg = img.replace('_p0', '_p' + str(i))

self.save_file(url=id_url, img=newimg, filepath=filepath)

else:

self.save_file(url=id_url, img=img, filepath=filepath)

if imgtype == 2:

self.save_gif(id=id, idpath=filepath)

except ConnectionError:

return self.save_id(id=id, filepath=filepath)

# 保存图片

def save_file(self, url, img, filepath):

_setdnscache()

save_headers = {

'Referer': url,

'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.100 Safari/537.36'

}

name = img.split('/')[-1]

if os.path.isfile(filepath + name):

return print(name + '已存在')

res = self.downloadsession.get(img, headers=save_headers)

if res.status_code == 200:

with open(filepath + name, 'wb') as f:

f.write(res.content)

f.close()

return print(name)

# 保存动图

def save_gif(self, id, idpath):

_setdnscache()

if os.path.isfile(idpath + str(id) + '.gif'):

return print(id + '.gif已存在')

href = 'https://www.pixiv.net/member_illust.php?mode=medium&illust_id=' + str(id)

headers = {

'Referer': href,

'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.100 Safari/537.36'

}

res = requests.get(href, headers=headers)

urls = re.findall('"(https://i\.pximg\..*?' + str(id) + '.*?)"', res.text)

url = urls[-1]

zurl = url.replace('c/128x128/img-master', 'img-zip-ugoira').replace('square1200.jpg', 'ugoira1920x1080.zip')

# zzurl = zurl.replace('1920x1080', '600x600')

dirpath = 'Iimgs/' + str(id) + '/'

if not os.path.exists(dirpath):

os.makedirs(dirpath)

response = requests.get(zurl, headers=headers)

if response.status_code == 200:

# print(zurl)

name = zurl.split('/')[-1]

with open('Iimgs/' + name, 'wb') as f:

f.write(response.content)

f.close()

zf = ZipFile('Iimgs/' + name, 'r')

for file in zf.namelist():

zf.extract(file, dirpath)

zf.close()

# self.imgname(dirpath)

frames = []

imglist = os.listdir(dirpath)

for img in imglist:

if os.path.exists(dirpath + img):

frames.append(Image.open(dirpath + img))

frames[0].save(idpath + str(id) + '.gif', save_all=True, append_images=frames[1:], duration=1 / 24,

format='GIF', loop=0)

# os.system('ffmpeg -r 24 -i ' + dirpath + '%d.jpg ' + idpath + str(id) + '.gif')

self.del_file(dirpath)

os.remove('Iimgs/' + name)

response.close()

return

# 清理解压内容

def del_file(self, path):

ls = os.listdir(path)

for i in ls:

d_path = os.path.join(path, i)

if os.path.isdir(d_path):

self.del_file(d_path)

else:

os.remove(d_path)

os.rmdir(path)

return

# 重命名解压文件

# def imgname(self, path):

# dirs = os.listdir(path)

# for img in dirs:

# img_path = os.path.join(path, img)

# while img.startswith('0') and len(img) >= 6:

# img = img[1:]

# newpath = os.path.join(path, img)

# os.rename(img_path, newpath)

# 下载关注的画师

def author(self):

self.member()

if '11' in self.id_set:

self.id_set.remove('11')

# print(self.id_set)

threads = []

for m_id in self.id_set:

t = Thread(target=self.member_id, args=(m_id,))

threads.append(t)

threadnum = len(threads)

for i in range(threadnum):

threads[i].start()

for i in range(threadnum):

threads[i].join()

def pagenum(self, url, p=1, filepath=None):

try:

pageurl = url + str(p)

wre = self.psession.get(pageurl, headers=self.header)

soup = BeautifulSoup(wre.text, 'lxml')

# print(wre.text)

data_items = re.search('data-items="(.*?)"', wre.text).group(1)

s = '{"body":' + data_items.replace('&quot;', '"') + '}'

# print(s)

# print(data_items.encode('latin-1').decode('unicode_escape'))

jsondata = json.loads(s, encoding='utf-8')

for i in jsondata['body']:

self.save_id(i["illustId"], filepath=filepath)

lis = soup.find(class_='page-list')

if not lis:

return 1

li = lis.find_all('li')[-1].string

print(li)

return int(li)

except ConnectionError:

return self.pagenum(url=url, p=p)

# 根据关键字下载

def word(self, name, num):

filepath = 'Author/关键字/%s/' % name

wordurl = "https://www.pixiv.net/search.php?word=%s %s&p=" % (name, num)

# print(wordurl)

p = 0

while True:

p = p + 1

page = self.pagenum(url=wordurl, p=p, filepath=filepath)

if p >= page:

def myfunc(self, *args, **kwargs):

if self.login_status:

return func(self, *args, **kwargs)

else:

return self.evar.set('请先登录')

def __init__(self, func, args=()):

super(MyThread, self).__init__()

self.func = func

self.args = args

def run(self):

self.result = self.func(*self.args)

def get_result(self):

try:

return self.result

except Exception:

a = User()

login_status = False

def __init__(self, master):

frame = tk.Canvas(master)

frame.pack()

# 登录框

ttk.Label(master, text='用户名', font=('Arial', 12)).place(x=50, y=50)

self.eu = ttk.Entry(master, show=None)

self.eu.place(x=150, y=50)

ttk.Label(master, text='密码', font=('Arial', 12)).place(x=50, y=100)

self.ep = ttk.Entry(master, show='.')

self.ep.place(x=150, y=100)

self.evar = tk.StringVar()

error = ttk.Label(master, textvariable=self.evar, font=('Arial', 12))

error.place(x=200, y=150)

ttk.Button(master, text='登录', command=self.user_pass).place(x=100, y=150)

# 个人下载框

ttk.Button(master, text='下载收藏的图片', command=self.mycollection).place(x=400, y=50)

ttk.Button(master, text='下载关注的画师图片', command=self.mymember).place(x=400, y=100)

self.disnum = ttk.Entry(master, show=None)

self.disnum.place(x=400, y=150, width=50)

ttk.Button(master, text='下载推荐的图片1-999张', command=self.discovery).place(x=400, y=180)

# 图片id下载框

ttk.Label(master, text='输入图片id', font=('Arial', 12)).place(x=0, y=200)

self.pid = ttk.Entry(master, show=None)

self.pid.place(x=100, y=200, width=100)

ttk.Button(master, text='下载', command=self.getid).place(x=220, y=200)

ttk.Label(master, text='无需登录', font=('Arial', 12)).place(x=320, y=200)

# 用户id下载框

ttk.Label(master, text='输入用户id', font=('Arial', 12)).place(x=0, y=250)

self.uid = ttk.Entry(master, show=None)

self.uid.place(x=100, y=250, width=100)

ttk.Button(master, text='下载用户作品', command=self.getuserwork).place(x=220, y=250)

ttk.Button(master, text='下载用户收藏', command=self.getusercol).place(x=320, y=250)

# 关键字下载框

ttk.Label(master, text='输入关键字', font=('Arial', 12)).place(x=0, y=300)

self.kword = ttk.Entry(master, show=None)

self.kword.place(x=100, y=300, width=100)

ttk.Button(master, text='下载收藏大于5000', command=self.getkey_5000).place(x=220, y=300)

ttk.Button(master, text='下载收藏大于10000', command=self.getkey_10000).place(x=350, y=300)

# 排行框

ttk.Label(master, text='年', font=('Arial', 12)).place(x=10, y=330)

ttk.Label(master, text='月', font=('Arial', 12)).place(x=70, y=330)

ttk.Label(master, text='日', font=('Arial', 12)).place(x=130, y=330)

ttk.Label(master, text='下载多少张(1-50)?', font=('Arial', 12)).place(x=200, y=330)

self.ranknum = ttk.Entry(master, show=None)

self.ranknum.place(x=200, y=360, width=50)

nowyear = int(time.strftime('%Y'))

self.year = ttk.Combobox(master, values=[i for i in range(2010, nowyear + 1)], state='readonly')

self.year.set(nowyear)

self.year.bind("<<ComboboxSelected>>", self.daynum)

self.year.place(x=10, y=360, width=50)

self.month = ttk.Combobox(master, values=[i for i in range(1, 13)], state='readonly')

self.month.bind("<<ComboboxSelected>>", self.daynum)

self.month.place(x=70, y=360, width=50)

self.dayvar = tk.IntVar()

self.day = ttk.Combobox(master, state='readonly')

ttk.Button(master, text='按月', command=lambda: self.get_rank(mode='monthly')).place(x=260, y=360, width=40)

ttk.Button(master, text='按周', command=lambda: self.get_rank(mode='weekly')).place(x=305, y=360, width=40)

ttk.Button(master, text='按日', command=lambda: self.get_rank(mode='daily')).place(x=350, y=360, width=40)

ttk.Label(master, text='建议下载完成后使用', font=('Arial', 10)).place(x=420, y=340)

ttk.Button(master, text='去除排行中重复的图', command=self.delfile).place(x=420, y=360, width=120)

# 设置动态的day下拉框

def daynum(self, *args):

yy = self.year.get()

mm = self.month.get()

# print(yy)

# print(mm)

if yy and mm:

self.dayvar.set(get_day(int(yy), int(mm)))

self.day['values'] = [i for i in range(1, self.dayvar.get() + 1)]

self.day.current(0)

self.day.place(x=130, y=360, width=50)

return None

def user_pass(self):

user = self.eu.get()

password = self.ep.get()

if user and password:

t = MyThread(self.a.login, args=(user, password))

t.start()

t.join()

message = t.get_result()

if message:

self.login_status = True

self.evar.set('用户%s登录成功' % self.a.userid)

else:

self.evar.set('账号或密码错误')

return

else:

self.evar.set('输入用户名和密码')

return False

@islogin

def mycollection(self):

t = Thread(target=self.a.collection)

t.start()

return self.evar.set('下载中')

@islogin

def mymember(self):

t = Thread(target=self.a.author)

t.start()

return self.evar.set('下载中')

def getid(self):

id = self.pid.get()

rid = re.match('^[1-9][0-9]{5,7}$', id)

if not rid:

return self.evar.set('请输入正确的id')

t = Thread(target=self.a.save_id, args=(id,))

t.start()

return self.evar.set('下载中')

@islogin

def getuserwork(self):

id = self.uid.get()

usid = re.match('^[1-9][0-9]{2,8}$', id)

if not usid:

return self.evar.set('请输入正确的id')

t = Thread(target=self.a.member_id, args=(id,))

t.start()

return self.evar.set('下载中')

@islogin

def getusercol(self):

id = self.uid.get()

usid = re.match('^[1-9][0-9]{2,8}$', id)

if not usid:

return self.evar.set('请输入正确的id')

t = Thread(target=self.a.collection, args=(id,))

t.start()

return self.evar.set('下载中')

@islogin

def getkey_5000(self):

kword = self.kword.get()

if not kword:

return self.evar.set('关键字不能为空')

t = Thread(target=self.a.word, args=(kword, 5000))

t.start()

return self.evar.set('下载中')

@islogin

def getkey_10000(self):

kword = self.kword.get()

if not kword:

return self.evar.set('关键字不能为空')

t = Thread(target=self.a.word, args=(kword, 10000))

t.start()

return self.evar.set('下载中')

@islogin

def discovery(self):

num = self.disnum.get()

renum = re.match('^[1-9][0-9]{0,2}$', num)

if not renum:

return self.evar.set('输入不规范')

num = int(num)

t = Thread(target=self.a.discovery, args=(num,))

t.start()

return self.evar.set('下载中')

def get_rank(self, mode):

year = self.year.get()

month = self.month.get()

day = self.day.get()

if year and month and day:

if len(month) < 2:

month = '0' + month

if len(day) < 2:

day = '0' + day

rankdate = year + month + day

print(rankdate)

nowtime = time.strftime("%Y%m%d")

print(nowtime)

if rankdate >= nowtime:

return self.evar.set('这个时间没有排行')

num = self.ranknum.get()

if not num:

return self.evar.set('数字不能为空')

numrule = re.match('(^[1-4][0-9]$)|(^50$)|(^[1-9]$)', num)

if not numrule:

return self.evar.set('输入1-50')

t = Thread(target=self.a.rank, args=(mode, rankdate, int(num)))

t.start()

return self.evar.set('下载中')

else:

return self.evar.set('请选择日期')

def delfile(self):

t = Thread(target=del_repeat, args=(1,))

t.start()

s = set()

ls = os.listdir(path)

for i in ls:

path2 = os.path.join(path, i)

if os.path.isdir(path2):

imgs = os.listdir(path2)

for img in imgs:

path3 = os.path.join(path2, img)

if img in s:

os.remove(path3)

else:

if m in [1, 3, 5, 7, 8, 10, 12]:

return 31

elif m in [4, 6, 9, 11]:

return 30

else:

if y % 4 == 0:

return 29

else:

root = tk.Tk()

root.title('下载')

root.geometry('550x400')

app = App(root)