def parse(fcode):
url = "http://www.clcindex.com/category/%s/" % fcode
ip = {"http": random_ip()}
try:
res = requests.get(url, proxies=ip, timeout=3)
if res.status_code == 200:
print(url)
print(ip)
html = Selector(res.text, 'html')
clc_nos = html.xpath(
"//tr[@name='item-row']//td[2]/text()").extract()
for i, clc_no in enumerate(clc_nos):
clc_no = clc_no.replace("\t", "").replace("\n", "")
clc_name = html.xpath("//tr[@name='item-row']//td[3]//text()"
).extract()[i].replace("\t", "").replace(
"\n", "")
sql = "insert or replace into clc(fcode,info) values ('%s','%s')" % (
clc_no, clc_name)
curser = conn.cursor()
curser.execute(sql)
conn.commit()
utils.printf("%s 插入成功" % clc_no)
sql_up = "update clc set stat = 1 where fcode = '%s'" % fcode
curser = conn.cursor()
curser.execute(sql_up)
conn.commit()
else:
print("ip err")
except Exception as e:
print(e)
def startdown_list(self, message):
utils.printf('%s:开始下载列表页...' % self.provider)
if not self.list_path:
self.initpath()
self.refreshproxypool()
self.count = 0
conn = utils.init_db('mysql', 'apsjournal')
cur = conn.cursor()
current_year = time.strftime('%Y')
cur.execute(
"select url,stat from issue where stat=0 or year=%s or year=%s" %
(current_year, int(current_year) - 1))
rows = cur.fetchall()
self.totalcount = len(rows)
if self.totalcount == 0:
if len(os.listdir(self.list_path)) == 0:
utils.logerror('%s:没有新的issue不需要更新' % self.provider)
else:
self.sendwork('parse_list')
for url, _ in rows:
fdir = self.list_path + '/' + url.split('/')[-4]
if not os.path.exists(fdir):
os.makedirs(fdir)
fname = fdir + '/' + url.split('/')[-2] + '_' + url.split(
'/')[-1] + '.html'
self.sendwork('down_list', (url, fname))
def down_detail(self):
utils.printf("下载详情页开始...")
super().down_detail()
conn = utils.init_db('mysql', 'cqjtu_kingbook')
cur = conn.cursor()
while True:
cur.execute(
'select bookid,stat from book where stat=0 limit 10000')
rows = cur.fetchall()
conn.commit()
if len(rows) == 0:
break
for bookid, _ in rows:
print(bookid)
url = 'http://123.56.143.23/kingbookwaiwen/book/info.aspx?id={}'.format(
bookid)
dirname = '%s/%s' % (self.detail_path, bookid[:3])
if not os.path.exists(dirname):
os.makedirs(dirname)
filename = '%s/%s.html' % (dirname, bookid)
if os.path.exists(filename):
sql = 'update book set stat=1 where bookid="{}"'.format(
bookid)
cur.execute(sql)
conn.commit()
continue
resp = utils.get_html(url, proxies=self.proxy)
if not resp:
continue
with open(filename, mode='w', encoding='utf8') as f:
f.write(resp.content.decode())
sql = 'update book set stat=1 where bookid="{}"'.format(bookid)
cur.execute(sql)
conn.commit()
utils.printf("下载", bookid, "成功...")
def _summary_tvalues(self):
'''
Summary t-values
:return:
'''
print('')
utils.printf('=== Summary T-Values ===')
cols = [
'MIN', 'MEDIAN', 'MEAN', 'MAX', 'STD. DEV.', 'T-TEST', 'P(+)',
'P(-)'
]
df = pd.DataFrame(index=self.tvalues.keys(), columns=cols)
for k, v in self.tvalues.items():
tstats = self.model.tvalues[k]
df.loc[k, 'MIN'] = round(min(v), 4)
df.loc[k, 'MEDIAN'] = round(np.median(v), 4)
df.loc[k, 'MEAN'] = round(np.mean(v), 4)
df.loc[k, 'MAX'] = round(max(v), 4)
df.loc[k, 'STD. DEV.'] = round(np.std(v), 4)
df.loc[k, 'T-TEST'] = round(tstats, 4)
df.loc[k, 'P(+)'] = round(np.mean([int(c >= tstats) for c in v]),
4)
df.loc[k, 'P(-)'] = round(np.mean([int(c <= tstats) for c in v]),
4)
print(df)
def parse_index(self, message):
try:
utils.printf('%s:解析索引页开始...' % self.provider)
conn = utils.init_db('mysql', 'hepengineeringjournal', 4)
self.sqlList.clear()
cur = conn.cursor()
for filename, fullname in utils.file_list(self.index_path):
with open(fullname, encoding='utf8') as f:
text = f.read()
dic = json.loads(text, encoding='utf-8')
gch = filename.replace('.json', '')
dicitem = dic['resultValue']
issn = dicitem['issnNm']
cnno = dicitem['cnNm']
sql = 'update journal set issn="%s",cnno="%s" where journal_id="%s"' % (
issn, cnno, gch)
cur.execute(sql)
conn.commit()
cur.close()
conn.close()
utils.printf('%s:解析索引页完成...' % self.provider)
# self.sendwork('down_cover')
self.senddistributefinish('get_issuelist')
except:
exMsg = '* ' + traceback.format_exc()
print(exMsg)
utils.logerror(exMsg)
def startdown_detail(self, message):
if not self.detail_path:
self.initpath()
self.sqlList.clear()
self.refreshproxypool()
self.count = 0
conn = utils.init_db('mysql', 'hepengineeringjournal', 4)
cur = conn.cursor()
cur.execute(
'select article_id,journal_id from article where stat=0 and failcount<3'
)
rows = cur.fetchall()
self.totalcount = len(rows)
if self.totalcount == 0:
utils.printf('%s:下载详情页完成' % self.provider)
# self.sendwork('parse_detail_meta')
self.sendwork('parse_detail')
# self.sendwork('down_cover')
return
messagelist = []
for article_id, journal_id in rows:
fdir = '%s/%s' % (self.detail_path, journal_id)
if not os.path.exists(fdir):
os.makedirs(fdir)
messagelist.append((article_id, journal_id))
if len(messagelist) == 30:
blist = messagelist.copy()
self.sendwork('down_detail', blist)
# utils.printf('a'+len(messagelist))
# utils.printf(messagelist)
messagelist.clear()
if len(messagelist) > 0:
self.sendwork('down_detail', messagelist)
def run(self):
conn = utils.init_db('mysql', 'aipjournal')
cur = conn.cursor()
sql = "select url,stat from issue where stat=0 limit 1000;"
time_last = time.time()
cnt = 0
while True:
if url_queue.empty():
cur.execute(sql)
rows = cur.fetchall()
conn.commit()
if rows:
for row in rows:
url_queue.put(row)
elif sql_queue.empty():
break
time_now = time.time()
if (sql_queue.qsize() > 100) or (time_now - time_last > 60):
num = sql_queue.qsize()
while num > 0:
url, flag = sql_queue.get()
cur.execute(
"update issue set stat={} where url='{}'".format(
flag, url))
cnt += 1
num -= 1
conn.commit()
utils.printf('succssed:%d' % (cnt))
time_last = time.time()
time.sleep(1)
def down_cover():
conn = pymysql.connect(DBHOST, DBUSER, DBPWD, DB)
now_time = time.strftime('%Y%m%d')
dirpath = cover_path + '/' + now_time
if not os.path.exists(dirpath):
os.makedirs(dirpath)
sql_up = "update video set cover_stat = 1 where rawid = %s"
result = []
while True:
sql = "select rawid,cover_url from video where cover_stat=0 limit 1000"
cur = conn.cursor()
cur.execute(sql)
rows = cur.fetchall()
if len(rows) == 0:
break
else:
for rawid, cover_url in rows:
path = dirpath + '/%s.jpg' % rawid
res = utils.get_html(cover_url, proxies=proxy, timeout=50)
if res:
if os.path.exists(path):
result.append((rawid))
utils.printf("该", rawid, "存在...")
else:
if utils.Img2Jpg(res.content, path):
result.append((rawid))
utils.printf("下载", rawid, "成功...")
else:
print('%s -- down cover error' % rawid)
if utils.parse_results_to_sql(conn, sql_up, result, 100):
total = len(result)
result.clear()
print('更新 ', total, ' 个结果到数据库成功')
utils.parse_results_to_sql(conn, sql_up, result)
print('更新 ', len(result), ' 个结果到数据库成功')
def down_index(self, message):
try:
journalname = message[0]
year = message[1]
fname = self.index_path + '/' + journalname + '_' + str(
year) + '.html'
utils.printf('开始下载 %s' % fname)
if os.path.exists(fname):
self.senddistributefinish('process_index')
return
feature = 'issue-month-detail'
url = 'http://{}.sciencemag.org/content/by/year/{}'.format(
journalname, str(year))
resp = self.gethtml(url, feature)
if not resp:
self.sendwork('down_index', (journalname, year))
return
with open(fname, mode='w', encoding='utf8') as f:
f.write(resp.content.decode('utf8'))
utils.printf('下载 %s 成功' % fname)
self.senddistributefinish('process_index')
except:
exMsg = '* ' + traceback.format_exc()
print(exMsg)
utils.logerror(exMsg)
def handleMenuItem_(self, ns_item):
'''ObjC callback to handle and dispatch C{NSMenuItem}
clicks and shortcuts.
All clicks and shortcuts are dispatched to the I{action}
method of this I{NSDelegate}'s L{App} instance.
Unhandled clicks, shortcuts and dispatch errors are
silently ignored, unless L{App} C{raiser} keyword
argument was C{True}.
'''
item = ns2Item(ns_item)
act = item._action
for t, i in ((self.app, item), (self, ns_item)):
m = getattr(t, act, None)
if m and callable(m):
try:
m(i)
break
except Exception:
if _Globals.raiser:
printf('%s(%r): %r method %s ...',
_handleMenuItem_name, i, t, act)
raise
else:
if _Globals.raiser:
raise RuntimeError('%s(%r): %s' % ('unhandled', item, act))
def parse_detal():
for file, fullpath in utils.file_list(detailpath):
j_id = file.replace(".html", '')
with open(fullpath, encoding='utf8') as f:
text = f.read()
html = Selector(text, 'html')
title = html.xpath("//h3/text()").extract_first("")
title_en = html.xpath("//h4/text()").extract_first("").replace(
"'", "''")
div = html.xpath("//div[@class='perinfo']/text()").extract()
zbdw = dq = issn = cn = shijian = ""
for item in div:
if item.startswith("主办单位："):
zbdw = item.replace("主办单位：", "")
if item.startswith("地区："):
dq = item.replace("地区：", "")
if item.startswith("国际刊号："):
issn = item.replace("国际刊号：", "")
if item.startswith("国内刊号："):
cn = item.replace('国内刊号：', '')
if item.startswith("出版周期："):
shijian = item.replace("出版周期：", "")
# utils.printf(title,title_en,zbdw,dq,issn,cn,shijian)
sql = "update journal set 期刊名称_外文 = '%s' , 主办单位 = '%s' , 地区 = '%s' , 国际刊号 = '%s' , 国内刊号 = '%s' , 出版周期 = '%s' where 期刊id = '%s'" % (
title_en, zbdw, dq, issn, cn, shijian, j_id)
curser = db.cursor()
curser.execute(sql)
curser.commit()
utils.printf("更新%s信息成功" % title)
def startdown_list(self, message):
utils.printf('%s:开始下载列表页...' % self.provider)
if not self.list_path:
self.initpath()
self.refreshproxypool()
self.count = 0
conn = utils.init_db('mysql', 'science')
cur = conn.cursor()
cur.execute('select url,stat from issue where stat=0')
rows = cur.fetchall()
self.totalcount = len(rows)
if self.totalcount == 0:
if len(os.listdir(self.index_path)) == 0:
utils.logerror('%s:没有新的issue不需要更新' % self.provider)
utils.msg2weixin('%s:没有新的issue不需要更新' % self.provider)
else:
self.sendwork('parse_list')
for url, _ in rows:
fdir = self.list_path + '/' + url.split('.')[0]
if not os.path.exists(fdir):
os.makedirs(fdir)
fname = fdir + '/' + url.split('/')[-2] + '_' + url.split(
'/')[-1] + '.html'
url = 'http://' + url
self.sendwork('down_list', (url, fname))
def get_year2que():
for year in range(1949, 2021):
base_url = dic_journal['Pharmacological Reviews']
url = base_url + '/%s' % str(year)
message = (str(year), url)
year_que.put(message)
utils.printf("%s年url添加成功!~" % year)
def parse_detail(self, message):
conn = utils.init_db('mysql', 'hepjournal', 4)
cur = conn.cursor()
cur.execute(
'select journal_id,journal_name,issn,eissn,cnno from journal')
rows = cur.fetchall()
for journal_id, journal_name, issn, eissn, cnno in rows:
self.dic[journal_id] = (journal_name, issn, eissn, cnno)
cur.close()
conn.close()
self.predb3()
self.sqlList.clear()
stmt = """insert or ignore into modify_title_info_zt(lngid, rawid, creator, title, volume, issue, page, beginpage,
endpage, publisher, subject, date,creator_institution, date_created, source, identifier_pissn, identifier_eissn,
identifier_cnno, description, identifier_doi, language, country, provider, provider_url, provider_id, type, medium,
batch, gch)values(?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?)"""
count = 0
for filename, fullname in utils.file_list(self.detail_path):
onemessage = self.parse_detail_one(filename, fullname)
if onemessage:
self.sqlList.append(onemessage)
if utils.parse_results_to_sql(self.conn, stmt, self.sqlList, 50):
count += len(self.sqlList)
utils.printf('%s: 插入 %d 条数据到db3' % (self.provider, count))
self.sqlList.clear()
utils.parse_results_to_sql(self.conn, stmt, self.sqlList)
count += len(self.sqlList)
utils.printf('%s: 插入 %d 条数据到db3' % (self.provider, count))
self.conn.close()
self.conn = None
utils.msg2weixin('%s: 解析完成,成品文件为%s' %
(self.provider, self.template_file))
def down_cover(self, message):
HEADER = {
'User-Agent':
'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/68.0.3440.106 Safari/537.36',
}
bookid = message[0]
cover_url = message[1]
filename = self.cover_path + '/' + bookid + '.jpg'
if os.path.exists(filename):
self.senddistributefinish('process_cover', bookid)
return
try:
proxy = self.getproxy()
proxies = {'http': proxy, 'https': proxy}
resp = requests.get(cover_url,
headers=HEADER,
timeout=20,
proxies=proxies)
# resp = requests.get(cover_url, headers=HEADER, timeout=20)
except:
self.sendwork('down_cover', message)
return
if utils.Img2Jpg(resp.content, filename):
utils.printf('下载图片%s成功' % filename)
self.senddistributefinish('process_cover', bookid)
else:
self.sendwork('down_cover', message)
return
def startdown_list(self, message):
utils.printf('%s:开始下载列表页...' % self.provider)
if not self.list_path:
self.initpath()
self.refreshproxypool()
self.sqlList.clear()
self.count = 0
conn = utils.init_db('mysql', 'hepjournal', 4)
cur = conn.cursor()
cur.execute('select url,journal_id from issue where stat=0')
rows = cur.fetchall()
self.totalcount = len(rows)
if self.totalcount == 0:
if len(os.listdir(self.list_path)) == 0:
utils.logerror('%s:没有新的issue不需要更新' % self.provider)
else:
# self.sendwork('down_cover')
self.sendwork('parse_list')
for url, journal_id in rows:
fdir = self.list_path + '/' + journal_id
if not os.path.exists(fdir):
os.makedirs(fdir)
fname = fdir + '/' + journal_id + '_' + url.split(
'/')[-2] + '_' + url.split('/')[-1].replace('.shtml', '.html')
self.sendwork('down_list', (url, fname))
def down_detail():
utils.printf("下载详情页开始...")
now_time = datetime.datetime.now().strftime("%Y%m%d")
conn = pymysql.connect(DBHOST, DBUSER, DBPWD, DB)
cur = conn.cursor()
while True:
cur.execute('select bookid,stat from book where stat=0 limit 10000')
rows = cur.fetchall()
conn.commit()
if len(rows) == 0:
break
for bookid, _ in rows:
print(bookid)
url = 'http://10.5.23.18:8079/book/bookinfo.aspx?id={}'.format(bookid)
dir_path = detail_path + '/' + now_time
dirname = '%s/%s' % (dir_path,bookid[:3])
if not os.path.exists(dirname):
os.makedirs(dirname)
filename = '%s/%s.html' % (dirname,bookid)
if os.path.exists(filename):
sql = 'update book set stat=1 where bookid="{}"'.format(bookid)
cur.execute(sql)
conn.commit()
continue
resp = utils.get_html(url, proxies=proxy)
if not resp:
continue
with open(filename, mode='w', encoding='gb18030') as f:
f.write(resp.content.decode())
sql = 'update book set stat=1 where bookid="{}"'.format(bookid)
cur.execute(sql)
conn.commit()
utils.printf("下载", bookid, "成功...")
def parse_detail_meta(self, message):
conn = utils.init_db('mysql', 'aiaajournal', 2)
cur = conn.cursor()
cur.execute(
'select gch,journal_name,journal_name_en,pissn,eissn from journal')
rows = cur.fetchall()
for gch, journal_name, journal_name_en, pissn, eissn in rows:
self.dic[gch] = (journal_name, journal_name_en, pissn, eissn)
cur.close()
conn.close()
self.predb3('base_obj_meta_a_template_qk.db3',
'base_obj_meta_a_qk.aiaajournal')
self.sqlList.clear()
stmt = """insert into base_obj_meta_a (author,author_1st,organ,organ_1st,title,title_alt,keyword,pub_year,pub_date,
vol,num,journal_raw_id,journal_name,journal_name_alt,page_info,begin_page,end_page,subject,is_oa,down_cnt,lngid,
rawid,product,sub_db,
provider,sub_db_id,source_type,provider_url,country,language,batch,down_date,publisher,issn,eissn,abstract,
abstract_alt,doi,fund,ref_cnt,fulltext_type) values(?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,
?,?,?,?,?,?,?,?,?,?,?)"""
count = 0
for filename, fullname in utils.file_list(self.detail_path):
onemessage = self.parse_detail_one(filename, fullname, 'meta')
if onemessage:
self.sqlList.append(onemessage)
if utils.parse_results_to_sql(self.conn, stmt, self.sqlList, 50):
count += len(self.sqlList)
utils.printf('%s: 插入 %d 条数据到db3' % (self.provider, count))
self.sqlList.clear()
utils.parse_results_to_sql(self.conn, stmt, self.sqlList)
count += len(self.sqlList)
utils.printf('%s: 插入 %d 条数据到db3' % (self.provider, count))
self.conn.close()
self.conn = None
utils.msg2weixin('%s: 解析完成,成品文件为%s' %
(self.provider, self.template_file))
def parse_detail(self, message):
conn = utils.init_db('mysql', 'aiaabook', 2)
cur = conn.cursor()
cur.execute('select url,pub_year from book')
rows = cur.fetchall()
for url, pub_year in rows:
doi = '10.2514/' + url.split('/')[-1]
self.dic[doi] = (pub_year)
cur.close()
conn.close()
self.predb3()
self.sqlList.clear()
stmt = """insert or ignore into modify_title_info_zt(lngid, rawid, creator, title, identifier_pisbn,
identifier_eisbn, description, publisher,cover,title_series,
date,date_created, price, language, country, provider, provider_url, identifier_doi, provider_id,
type,medium, batch) values(?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?,?)
"""
count = 0
for filename, fullname in utils.file_list(self.detail_path):
onemessage = self.parse_detail_one(filename, fullname, 'zt')
# print(onemessage)
if onemessage:
self.sqlList.append(onemessage)
if utils.parse_results_to_sql(self.conn, stmt, self.sqlList, 50):
count += len(self.sqlList)
utils.printf('%s: 插入 %d 条数据到db3' % (self.provider, count))
self.sqlList.clear()
utils.parse_results_to_sql(self.conn, stmt, self.sqlList)
count += len(self.sqlList)
utils.printf('%s: 插入 %d 条数据到db3' % (self.provider, count))
self.conn.close()
self.conn = None
utils.msg2weixin('%s: 解析完成,成品文件为%s' %
(self.provider, self.template_file))
def generateFrequencyMatrix(self):
if self.sorted_data is None:
utils.printf('ERROR: Your (sorted) data should be loaded!')
return
if self.sorted_data.shape[1] != 2:
utils.printf(
'ERROR: Your (sorted) matrix should have 2 columns only (attribute, class)'
)
return
unique_attribute_values, indices = np.unique(
self.sorted_data[:, 0],
return_inverse=True) # first intervals: unique attribute values
unique_class_values = np.unique(
self.sorted_data[:, 1]) # classes (column index 1)
self.frequency_matrix = np.zeros(
(len(unique_attribute_values),
len(unique_class_values))) # init frequency_matrix
self.frequency_matrix_intervals = unique_attribute_values # init intervals (unique attribute values)
self.nclasses = len(unique_class_values) # number of classes
self.degrees_freedom = self.nclasses - 1 # degress of freedom (look at table)
# Generating first frequency values (contingency table), number of instances found in data: attribute-class
for row in np.unique(indices):
for col, clase in enumerate(unique_class_values):
self.frequency_matrix[row, col] += np.where(
self.sorted_data[np.where(
indices == row)][:, 1] == clase)[0].shape[0]
self.printInitialSummary()
def down_cover(self, message):
HEADER = {
'User-Agent':
'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/68.0.3440.106 Safari/537.36',
}
url = message[0]
cover_url = message[1]
filename = '%s/10.2514/%s.jpg' % (self.cover_path, url.split('/')[-1])
if os.path.exists(filename):
self.senddistributefinish('process_cover', url)
return
try:
proxy = self.getproxy()
proxies = {'http': proxy, 'https': proxy}
resp = requests.get('https://arc.aiaa.org' + cover_url,
headers=HEADER,
timeout=20,
proxies=proxies)
# resp = requests.get(cover_url, headers=HEADER, timeout=20)
except:
exMsg = '* ' + traceback.format_exc()
print(exMsg)
self.sendwork('down_cover', message)
return
if utils.Img2Jpg(resp.content, filename):
utils.printf('下载图片%s成功' % filename)
self.senddistributefinish('process_cover', url)
else:
self.sendwork('down_cover', message)
return
def downpage(url, proxy):
urlsplist = url.split('/')
index_dir = r'E:\lqx\AIP\issue' + '/' + urlsplist[-3]
index_file = index_dir + '/' + urlsplist[-2] + '_' + urlsplist[-1] + '.html'
if os.path.exists(index_file):
return True
url = url + '?size=all'
proxies = {
"http": "http://{}".format(proxy),
"https": "https://{}".format(proxy)
}
feature = 'Table of Contents'
try:
resp = utils.get_html(url, feature=feature, proxies=proxies)
except:
# exMsg = '* ' + traceback.format_exc()
# print(exMsg)
return -1
if not resp:
return -1
if resp.text.find(feature) < 0:
return -1
if resp.text.find('</html>') < 0:
return -1
if not os.path.exists(index_dir):
os.makedirs(index_dir)
with open(index_file, mode='w', encoding='utf8') as f:
f.write(resp.content.decode('utf8'))
utils.printf('下载', url, '成功...')
return True
def startdown_html(self, message):
infotype = message
if not self.html_path:
self.initpath()
self.refreshproxypool()
url = 'https://www.ydylcn.com/%s/index.shtml' % infotype
feature = 'class="list-link-1"'
fdir = '%s/%s' % (self.html_path, infotype)
if not os.path.exists(fdir):
os.makedirs(fdir)
fname = '%s/1.html' % fdir
utils.printf(fname)
if not os.path.exists(fname):
while True:
resp = self.gethtml(url, feature)
if resp:
break
text = resp.content.decode('utf8')
with open(fname, mode='w', encoding='utf8') as f:
f.write(text)
else:
with open(fname, encoding='utf8') as f:
text = f.read()
sel = Selector(text=text)
pagetotalnum = sel.xpath('//table//tr/td/text()')[1].re(
r'\s*/(.*)页')[0]
self.count = 0
self.totalcount = int(pagetotalnum) - 1
for page in range(2, int(pagetotalnum) + 1):
self.sendwork('down_html', (page, infotype))
def downcover():
url = 'https://aip.scitation.org'
cover_dir_fullpath = os.path.dirname(os.path.abspath(__file__)) + '/cover'
if not os.path.exists(cover_dir_fullpath):
os.makedirs(cover_dir_fullpath)
try:
resp = utils.get_html(url)
except:
# exMsg = '* ' + traceback.format_exc()
# print(exMsg)
return False
if not resp:
return False
if resp.text.find('</html>') < 0:
return False
soup = BeautifulSoup(resp.content.decode('utf8'), 'lxml')
divList = soup.select('div.publicationCoverImage')
# divpb = soup.select_one('div', data - widget - id='bfd39502-c303-4169-88ba-1d2b9bba85ab')
for divtag in divList:
coverurl = url + divtag.a.img['src']
covername = cover_dir_fullpath + '/' + divtag.a['href'].split(
'/')[-1].lower() + '.jpg'
if os.path.exists(covername):
continue
resp = utils.get_html(coverurl)
if utils.Img2Jpg(resp.content, covername):
utils.printf('下载', covername, '成功...')
time.sleep(3)
# apburl = 'https://aip.scitation.org/pb-assets/images/publications/apb/apl-bioeng-1483023557097.jpg'
# apbname = cover_dir_fullpath + '/' + 'apb.jpg'
# resp = utils.get_html(apburl)
# if utils.Img2Jpg(resp.content, apbname):
# utils.printf('下载', apbname, '成功...')
return True
def gethtml(self, url, feature=None, coverflag=False):
try:
resp = self.session.get(url,
headers=self.headers,
timeout=20,
proxies=self.proxies)
if not coverflag:
if resp.content.decode('utf-8').find('Just a moment...') > 0:
utils.printf('Just a moment...')
if not self.refreshflag:
self.refreshflag = True
self.refreshsession()
else:
while self.refreshflag:
time.sleep(1)
resp = self.session.get(url,
headers=self.headers,
timeout=20,
proxies=self.proxies)
if resp.status_code != 200:
print('code !=200')
return False
if resp.content.decode('utf-8').find('</html>') < 0:
print('not endwith </html>')
return False
if feature:
if resp.content.decode('utf-8').find(feature) < 0:
print('can not find feature')
utils.logerror(url)
return False
except:
return False
return resp
def parse_list():
conn = pymysql.connect(DBHOST, DBUSER, DBPWD, DB)
result = []
sql_in = "insert ignore into detail(provider_subject,title,url,add_time,look_time) values (%s,%s,%s,%s,%s)"
for _, filedir in utils.file_list(list_path):
# E:\work\美星外文\list\日文图书;随笔\2.html
utils.printf(filedir)
regex = r"E:\\work\\美星外文\\list\\(.*?)\\"
provider_subject = re.findall(regex, filedir)[0]
with open(filedir, mode='r', encoding='gb18030') as f:
text = f.read()
html = Selector(text, 'html')
list_urls = html.xpath(
"//tr[@class='tdbg_leftall']/td/strong/a/@href").extract()
for i, item in enumerate(list_urls):
title = html.xpath("//tr[@class='tdbg_leftall']/td/strong/a/text()"
).extract()[i].split(" ")[0]
url = "http://202.207.22.13:100/" + item
add_time = html.xpath(
"//tr[@class='tdbg_leftall']/td[3]/text()").extract()[i]
look_time = html.xpath(
"//tr[@class='tdbg_leftall']/td[4]/text()").extract()[i]
result.append((provider_subject, title, url, add_time, look_time))
utils.parse_results_to_sql(conn, sql_in, result)
print('插入', len(result), ' 个结果到数据库成功')
result.clear()
def startdown_list(self, message):
utils.printf('%s:开始下载列表页...' % self.provider)
if not self.list_path:
self.initpath()
self.sqlList.clear()
self.refreshproxypool()
self.count = 0
conn = utils.init_db('mysql', 'cambridgejournal')
cur = conn.cursor()
cur.execute('select url,stat from issue where stat=0')
rows = cur.fetchall()
self.totalcount = len(rows)
if self.totalcount == 0:
if len(os.listdir(self.list_path)) == 0:
utils.logerror('%s:没有新的issue不需要更新' % self.provider)
else:
# self.sendwork('down_cover')
self.sendwork('parse_list')
return
self.refreshsession()
for url, _ in rows:
fdir = self.list_path + '/' + url.split('/')[-3]
if not os.path.exists(fdir):
os.makedirs(fdir)
flast = url.split('/')[-1]
if flast.find('?pageNum=') > 0:
flast = flast.split('?')[0] + '_' + flast.split('=')[-1]
fname = fdir + '/' + flast + '.html'
self.sendwork('down_list', (url, fname))
def actTryTargetLogin(self, objBrowser, tryUsername, tryPassword, currentTry):
try:
# Fill Login field Information
objBrowser.select_form(nr = self.formLoginID)
objBrowser.form[self.formUsernameField] = tryUsername
objBrowser.form[self.formPasswordField] = tryPassword
# Print progress bar
utils.prints("%10s : %20s%12s%10s / %10s" %(tryUsername, tryPassword, '=' * 6, currentTry, self.sizePasslist))
# Send request
objBrowser.submit()
# Refresh page, useful for redirect after login
objBrowser.reload()
# If result has no login form -> Success **NEED IMPROVE**
# add login information to fndData, return True
if not actions.action_getFormInformation(objBrowser.forms()):
utils.printf("Found: %s:%s" %(tryUsername, tryPassword), "good")
self.credentials.append([tryUsername, tryPassword])
return True
return False
except mechanize.HTTPError as error:
utils.printf(error, "bad")
sys.exit(1)
def startdown_list(self, message):
utils.printf('%s:开始下载列表页...' % self.provider)
if not self.list_path:
self.initpath()
self.refreshproxypool()
self.sqlList.clear()
self.count = 0
conn = utils.init_db('mysql', 'aiaajournal', 2)
cur = conn.cursor()
cur.execute('select url,stat from issue where stat=0')
rows = cur.fetchall()
self.totalcount = len(rows)
if self.totalcount == 0:
if len(os.listdir(self.list_path)) == 0:
utils.logerror('%s:没有新的issue不需要更新' % self.provider)
else:
self.sendwork('parse_list')
for url, _ in rows:
urlsp = url.split('/')
base_name = '%s_%s.html' % (urlsp[-2], urlsp[-1])
fdir = '%s/%s' % (self.list_path, urlsp[-3])
fname = '%s/%s' % (fdir, base_name)
if not os.path.exists(fdir):
os.makedirs(fdir)
self.sendwork('down_list', (url, fname))
def process_list(self, message):
self.count = self.count + 1
sql = "update issue set stat=1 where url='{}'".format(message)
self.sqlList.append(sql)
if self.count % 40 == 1:
utils.printf('%s:下载成功 %s 页' % (self.provider, self.count))
conn = utils.init_db('mysql', 'aiaajournal', 2)
cur = conn.cursor()
for sql in self.sqlList:
cur.execute(sql)
conn.commit()
conn.close()
self.sqlList.clear()
if self.count % 100 == 0:
self.refreshproxypool()
if self.count == self.totalcount:
conn = utils.init_db('mysql', 'aiaajournal', 2)
cur = conn.cursor()
for sql in self.sqlList:
cur.execute(sql)
conn.commit()
conn.close()
self.sqlList.clear()
utils.printf('downloadlist finish')
self.sendwork('parse_list')
def _continue(self, smallest):
c1 = self._too_many_intervals()
c2 = self._more_merges(smallest)
utils.printf('- Too many intervals?: {}'.format(c1))
utils.printf('- Can we merge more?: {}'.format(c2))
return c1 or c2
def chi2(self):
if self.data is None:
utils.printf('ERROR: Your data matrix should be loaded!')
return
### Phase1: defining sigLevel values for every numeric attribute, and chimerge for every attribute-column
sigLevel0 = self._phase1()
### Phase2: merging attrinutes if needed (vertical-wise)
self._phase2(sigLevel0)
def loadFrequencyMatrix(self, frequency_matrix, unique_attribute_values):
'''
:param frequency_matrix: numpy array
:return: void
'''
if type(frequency_matrix) != np.array:
utils.printf('ERROR: data must be a numpy.array')
return
self.frequency_matrix = frequency_matrix
self.frequency_matrix_intervals = unique_attribute_values
self.nclasses = self.frequency_matrix.shape[1]
self.degrees_freedom = self.nclasses - 1
self.printInitialSummary()
def joinscandirlbname(lbnamemap,scandir_list):
out = {}
for (date,aid,lid,file_path) in scandir_list:
if not lbnamemap.has_key(lid):
utils.printf("Warning lid[%i] was not found in lbmap\n",lid)
continue
name = lbnamemap[lid][2]
if not out.has_key(aid):
out[aid] = []
dict_row = {"file_path" :file_path,
"aid":aid,
"lid":lid,
"lname":name,
"date":date}
out[aid].append(dict_row)
return out
def loadData(self, data, issorted=False):
'''
:param data: numpy matrix
:param issorted: boolean, if data is already sorted, no need to sort again (based on attribute_column)
:return:
'''
if type(data) != np.matrix and type(data) != np.array:
utils.printf('ERROR: data must be a numpy.matrix or numpy.array')
return
self.data = data # numpy.matrix (x,2). column index 0 refers to attributes column and index 1 classes
if not issorted:
self.sorted_data = np.array(np.sort(data.view('i8,i8'), order=['f0'], axis=0).view(np.float)) #always sorting column 0 (attribute column)
else:
self.sorted_data = np.array(data)
utils.printf('Sorted data: matrix {}x{}'.format(self.sorted_data.shape[0],self.sorted_data.shape[1]))
def loadData(self, data):
'''
:param data: numpy matrix
:return:
'''
if type(data) != np.matrix and type(data) != np.array:
utils.printf('ERROR: data must be a numpy.matrix or numpy.array')
return
self.data = np.array(data) # no need to sort at this point
self.nattributes = self.data.shape[1]-1 # last column refers to class label
self.nclasses = np.unique(self.data[:,self.nattributes]).shape[0]
self.degrees_freedom = self.nclasses - 1
self.chimerge_per_column = {colid:None for colid in range(self.nattributes)}
self.alpha_per_column = {colid:None for colid in range(self.nattributes)}
self.attribute_can_be_merged = {colid:True for colid in range(self.nattributes)}
utils.printf('Data: matrix {}x{} ({} numeric attributes)'.format(self.data.shape[0],self.data.shape[1], self.nattributes))
self._loadChiDistribution()
def generateFrequencyMatrix(self):
if self.sorted_data is None:
utils.printf('ERROR: Your (sorted) data should be loaded!')
return
if self.sorted_data.shape[1] != 2:
utils.printf('ERROR: Your (sorted) matrix should have 2 columns only (attribute, class)')
return
unique_attribute_values, indices = np.unique(self.sorted_data[:,0], return_inverse=True) # first intervals: unique attribute values
unique_class_values = np.unique(self.sorted_data[:,1]) # classes (column index 1)
self.frequency_matrix = np.zeros((len(unique_attribute_values), len(unique_class_values))) # init frequency_matrix
self.frequency_matrix_intervals = unique_attribute_values # init intervals (unique attribute values)
self.nclasses = len(unique_class_values) # number of classes
self.degrees_freedom = self.nclasses - 1 # degress of freedom (look at table)
# Generating first frequency values (contingency table), number of instances found in data: attribute-class
for row in np.unique(indices):
for col, clase in enumerate(unique_class_values):
self.frequency_matrix[row,col] += np.where(self.sorted_data[np.where(indices == row)][:,1] == clase)[0].shape[0]
self.printInitialSummary()
def _inConsistency(self):
#1. matrix with all attribute-columns (except class-column)
#2. find duplicates (register indexes)
#3. for every duplicated instance do:
# 3.1. calculate inconsistency_count = (n-ck) where n is the number of time such instance is duplicated and ck the largest number of duplicates of such instance among all classes
#4. incosistency rate sum all inconsistency_count and divide by the number of instances (total instances)
#
# IT SHOULD NOT BE OVER RAW DATA, BUT OVER THE MERGED DATA!!!
# To be fixed!
#
if self.data is None:
utils.printf('ERROR: Your data matrix should be loaded!')
return
# 1. matrix with only attribute values
# 2. identify duplicates
unique_values, unique_indexes = np.unique(self.data[:,:self.nattributes-1], return_inverse=True)
unique_counts = np.bincount(unique_indexes)
matching_instances = unique_values[unique_counts>1]
sum_inconsistencies = 0
total_instances = unique_indexes.shape[0]
# 3. calculating inconsistency_count for every instance
for matching_instance in matching_instances:
c = {}
for colid in range(self.nclasses):
c[colid] = (self.data[self.data[:,self.nattributes]==colid] == matching_instance).sum()
n = sum(c.values())
cmax = max(c.values())
inconsistency_count = n - cmax
sum_inconsistencies += inconsistency_count
# 4. inconsistency rate
inconsistency_rate = sum_inconsistencies / float(total_instances)
return inconsistency_rate
def main(args):
prog = args[0]
if len(args) < 3:
usage(prog)
sys.exit()
conf_file = []
conf_file_a = args[1]
conf_file_b = args[2]
printf("Enter username: "******"Enter password: "******"Scanning files %s and %s\n", conf_file_a, conf_file_b)
ldap_a = utils.LdapConnecter(user, passwd, conf_file=conf_file_a)
ldap_b = utils.LdapConnecter(user, passwd, conf_file=conf_file_b)
printf("binding to ldap from conf %s\n", conf_file_a)
ldap_a.bind()
printf("binding to ldap from conf %s\n", conf_file_b)
ldap_b.bind()
printf("\n")
ldap_groups = [v for (k,v) in ldap_a.conf["roles"].items()]
for ldap_group in ldap_groups:
printf("ldap group %s\n", ldap_group)
printf("----------------------------------------------\n")
printf("\n")
members_a = ldap_a.get_group_members_ssos(ldap_group)
members_b = ldap_b.get_group_members_ssos(ldap_group)
a_ssos = set(members_a.keys())
b_ssos = set(members_b.keys())
n_common = len(a_ssos & b_ssos)
printf("common member count: %d\n", n_common )
printf(" missing from %s:\n", conf_file_a)
missing_from_a = sorted(list(b_ssos - a_ssos))
if len(missing_from_a) > 0:
max_col_len = max([len(sso) for sso in missing_from_a])
for sso in missing_from_a:
lsso = "%s" %((sso + ":").ljust(max_col_len +2),)
printf(" %s%s\n", lsso, members_b[sso])
else:
printf(" None\n")
printf("\n")
printf(" missing from %s:\n", conf_file_b)
missing_from_b = sorted(list(a_ssos - b_ssos))
if len(missing_from_b) > 0:
max_col_len = max([len(sso) for sso in missing_from_b])
for sso in missing_from_a:
lsso = "%s" % ((sso + ":").ljustt(max_col_len +2),)
printf(" %s%s\n", lsso, members_a[sso])
else:
printf(" None\n")
printf("\n")
printf("\n")
ldap_a.unbind()
ldap_b.unbind()
def usage(prog):
printf("usage is %s <conf_file1> <conf_file2>\n", prog)
printf("\b")
printf("list the members of the specified group name\n")
printf("\n")
import sys
import os
def usage(prog):
printf("usage is %s <conf_file>\n", prog)
printf("get the user group and roles based on the conf file\n")
printf("You will be prompted for your SSO user and passwd\n")
printf("\n")
if __name__ == "__main__":
prog = sys.argv[0]
if len(sys.argv) < 2:
usage(prog)
sys.exit()
conf_file = sys.argv[1]
printf("Enter username: "******"Enter password: "******"User %s is a member of %d groups\n", user, len(groups))
printf("--------------------------------------------------------\n")
for (group_cn, group_dn) in groups:
printf("%s\n", group_dn)
printf("\n")
printf("User %s has %d roles\n", user, len(roles))
printf("--------------------------------------------------------\n")
for role in roles:
def usage(prog):
printf("usage is %s <conf_file>\n", prog)
printf("get the user group and roles based on the conf file\n")
printf("You will be prompted for your SSO user and passwd\n")
printf("\n")
def chimerge(self):
if self.frequency_matrix is None:
utils.printf('ERROR: Your frequency matrix should be loaded!')
return
chitest = {}
counter = 0
smallest = -1
while self._too_many_intervals():
###
# CHI2 TEST
###
chitest = {}
shape = self.frequency_matrix.shape
for r in range(shape[0] - 1):
interval = r,r+1
chi2 = self.chisqrtest(self.frequency_matrix[[interval],:][0])
if chi2 not in chitest:
chitest[chi2] = []
chitest[chi2].append( (interval) )
smallest = min(chitest.keys())
biggest = max(chitest.keys())
###
# SUMMARY
###
counter += 1
utils.printf('')
utils.printf('ROUND {}: {} intervals. Chi min:{}, Chi max:{}'.format(counter, self.frequency_matrix.shape[0], smallest, biggest))
utils.printf('CHI2 VALUES: {}'.format(chitest.keys()))
###
# MERGE
###
if self._more_merges(smallest):
utils.printf('MERGING INTERVALS: chi {} -> {}'.format(smallest, chitest[smallest]))
for (lower,upper) in list(reversed(chitest[smallest])): # reversed, to be able to remove rows on the fly
for col in range(shape[1]): # checking columns (to append values from row i+1 ---to be removed--- to row i)
self.frequency_matrix[lower,col] += self.frequency_matrix[upper,col] # appending frequencies to the remaining interval
self.frequency_matrix = np.delete(self.frequency_matrix, upper, 0) # removing interval (because we merged it in the previous step)
self.frequency_matrix_intervals = np.delete(self.frequency_matrix_intervals, upper, 0) # also removing the corresponding interval (real values)
utils.printf('NEW INTERVALS: ({}):{}'.format(len(self.frequency_matrix_intervals),self.frequency_matrix_intervals))
else:
break
self.chitestvalues = chitest
utils.printf('END (chi {} > {})\n'.format(smallest, self.threshold))
def printInitialSummary(self):
utils.printf('')
utils.printf('ROUND 0: Initial values:')
utils.printf('- Number of classes: {}'.format(self.nclasses))
utils.printf('- Degrees of Freedom: {} (deprecated)'.format(self.degrees_freedom))
utils.printf('- Threshold: {}'.format(self.threshold))
utils.printf('- Max number of intervals: {}'.format(self.max_number_intervals))
utils.printf('- Number of (unique) intervals: {}'.format(len(self.frequency_matrix_intervals)))
utils.printf('- Frequency matrix: {}x{} (sum {})'.format(self.frequency_matrix.shape[0], self.frequency_matrix.shape[1], self.frequency_matrix.sum()))
utils.printf('- Intervals: {}'.format(self.frequency_matrix_intervals))
def printInitialSummary(self):
utils.printf('')
utils.printf('ROUND 0: Initial values:')
utils.printf('- Number of attributes: {}'.format(self.nattributes))
utils.printf('- Number of classes: {}'.format(self.nclasses))
utils.printf('- Degrees of Freedom: {} (deprecated)'.format(self.degrees_freedom))
utils.printf('- alpha (initial value of sigLevel): {}'.format(self.alpha))
utils.printf('- delta (inConsistency level): {}'.format(self.delta))
def flush_row(self, row):
printf("flush-row: %d", row)
return
def put_shape(self, location, shape_vector):
printf("put-shape: location: [row: %d, col: %d], shape: %s",
location[0], location[1],
shape_vector)
return
def put_center_string(self, row, display_string):
printf("put-center-string: [row: %d, display-string: %s]", row, display_string)
return
def printFinalSummary(self):
utils.printf('FINAL SUMMARY')
utils.printf('{}{}'.format('Intervals: ',self.frequency_matrix_intervals))
utils.printf('{}{}'.format('Chi2: ',', '.join(['[{}-{}):{:5.1f}'.format(v[0][0],v[0][1],k) for k,v in utils.sortDictByValue(self.chitestvalues,False)])))
utils.printf('{} ({}x{})\n{}'.format('Interval-Class Frequencies',self.frequency_matrix.shape[0],self.frequency_matrix.shape[1],self.frequency_matrix))
lbRows = getLidAidNameFromLbs(con,cur)
lb2name = getLbid2name(lbRows)
cur.close()
con.close()
scan_list = scandir_filter(scandir(cache_dir),**kw)
cfFiles = joinscandirlbname(lb2name,scan_list)
return (scan_list,cfFiles)
def main(args):
kw = {}
for arg in args[1:]:
(k,v) = arg.split("=")
try:
kw[k.strip()] = int(v.strip())
except ValueError:
kw[k.strip()] = v.strip() #if its not an int treat it like a string
return getCfFiles(**kw)
if __name__ == "__main__":
(scan_list,cfFiles) = main(sys.argv)
for i in xrange(0,len(scan_list)):
utils.printf("scan_list[%i]=%s\n",i,scan_list[i])
for(aid,files) in cfFiles.items():
for file_info in files:
utils.printf("%s: %s\n",aid,file_info)
#example invocation
#./cfupload.py aid=452605 date_gte=2013060600
#./cfupload.py aid=452605
#/cfupload.py aid=682644 date_gte=2013052500 date_lte=2013052523
import os
def usage(prog):
printf("usage is %s <conf_file> <group_name>\n", prog)
printf("\b")
printf("list the members of the specified group name\n")
printf("\n")
if __name__ == "__main__":
prog = sys.argv[0]
if len(sys.argv) < 3:
usage(prog)
sys.exit()
conf_file = sys.argv[1]
group_name = sys.argv[2]
printf("Enter username: "******"Enter password: "******"group %s has %d members\n", group_name, n_groups);
printf("-----------------------------------------------------------\n")
llen = max([len(sso) for sso in ssos])
for sso in ssos:
lsso = "%s" % ((sso + ":").ljust(llen + 2),)
def put_string(self, location, display_string):
printf("put-string: location: [row: %d, col: %d], display-string: [fmt: %s, str: %s]",
location[0], location[1],
display_string[0], display_string[1])
return
def execute(frame, bc):
code = bc.code
pc = 0
while True:
# required hint indicating this is the top of the opcode dispatch
driver.jit_merge_point(pc=pc, code=code, bc=bc, frame=frame)
if pc >= len(code):
return W_Null()
c = ord(code[pc])
arg = ord(code[pc + 1])
pc += 2
if c == bytecode.LOAD_CONSTANT:
w_constant = bc.constants[arg]
frame.push(w_constant)
elif c == bytecode.LOAD_VAR:
frame.push(frame.vars[arg])
elif c == bytecode.LOAD_NULL:
frame.push(W_Null())
elif c == bytecode.LOAD_BOOLEAN:
frame.push(W_Boolean(bool(arg)))
elif c == bytecode.LOAD_PARAM:
frame.push_arg(frame.pop()) # push to the argument-stack
elif c == bytecode.DISCARD_TOP:
frame.pop()
elif c == bytecode.RETURN:
if frame.valuestack_pos > 0:
return frame.pop()
else:
return W_Null()
elif c == bytecode.BINARY_ADD:
right = frame.pop()
left = frame.pop()
w_res = left.add(right)
frame.push(w_res)
elif c == bytecode.BINARY_LT:
right = frame.pop()
left = frame.pop()
frame.push(left.lt(right))
elif c == bytecode.BINARY_GE:
right = frame.pop()
left = frame.pop()
frame.push(left.ge(right))
elif c == bytecode.BINARY_EQ:
right = frame.pop()
left = frame.pop()
frame.push(left.eq(right))
elif c == bytecode.BINARY_SUB:
right = frame.pop()
left = frame.pop()
frame.push(left.sub(right))
elif c == bytecode.BINARY_STRINGJOIN:
right = frame.pop()
left = frame.pop()
frame.push(left.append(right))
elif c == bytecode.JUMP_IF_FALSE:
if not frame.pop().is_true():
pc = arg
elif c == bytecode.JUMP_BACKWARD:
pc = arg
# required hint indicating this is the end of a loop
driver.can_enter_jit(pc=pc, code=code, bc=bc, frame=frame)
elif c == bytecode.CALL:
method = bc.functions[arg]
method.body.globals = [None]*bc.numvars # XXX
new_bc = method.body
new_frame = Frame(new_bc)
# reverse args index to preserve order
for i in range(len(method.params)):
index = len(method.params) - 1 - i
new_frame.vars[index] = frame.pop_arg()
res = execute(new_frame, new_bc)
frame.push(res)
elif c == bytecode.PRINT:
item = frame.pop()
printf(item.str())
elif c == bytecode.ASSIGN:
frame.vars[arg] = frame.pop()
else:
raise Exception("Unkown operation %s" % bytecode.bytecodes[c])
def _loadChiDistribution(self):
with open('data/chisquare_distribution.data','r') as f:
data = json.load(f)
self.chidistribution = {float(k):v for k,v in data.items()}
utils.printf('ChiSquare distribution table loaded. {} sigLevel and {} degrees of freedom.'.format(len(self.chidistribution.keys()),len(self.chidistribution.values()[0])-1))