def search(path):
for filename in os.listdir(path):
fp = os.path.join(path, filename)
if os.path.isfile(fp):
for i in types:
if filename.endswith('.'+i):
if filename.endswith('jpg') and filename[:1] in '1,2,3,4,5,6,7,8,9,0'.split(','):
if filename not in gysimg:gysimg.append(filename)
filename=T.sub(fp,basedir,'')[1:].replace('\\','/')
funcname='fr_'+T.varname(filename)
exec(sfr.format(filename,funcname))
print filename,funcname
break
elif os.path.isdir(fp):
search(fp)
def replace_raw(raw_text, regex=True):
for i, v in gdraw.items():
if len(i) == 2 and isinstance(i, tuple):
if not regex: continue
if i[0] in raw_text:
raw_text = T.regexReplace(raw_text, i[1], str(v))
else:
if py.istr(i):
raw_text = raw_text.replace(str(i), str(v))
else:
U.log(['unknow gdraw', i, v])
return raw_text
# print j.get('class')
if(str(j.get('class')).find(u'in-reply-to')>0):
rto=j.find(attrs={"class":"ng-binding"})
# rto=rto.get('href')
# rto=str(rto)
if(len(rto)<1):rto=None
continue
if(str(j.get('href')).find('http://www.zhihu.com/people/')!=-1):
href=j.get('href');href=str(href)
name=j.text.encode('gbk')
continue
# print name
# if href not in ls:ls.append(href)
href=T.sub(href,'/people/')
rto=T.sub(rto,'/people/')
href=href.replace('-','').replace('.','')
if(rto!=None):
rto=rto.replace('-','').replace('.','')
print '%s->%s'%(href,rto)
else:
print href
ls.append((href,rto))
print '}'
U.resetOut()
for j in ls :
print j
#coding=utf-8
gsuser='******'
gszhihu='https://www.zhihu.com/'
gszp=gszhihu+'people/'
gscmt0='''https://www.zhihu.com/node/AnswerCommentBoxV2?params=%7B%22answer_id%22%3A%22'''
gscmt1='%22%2C%22load_all%22%3Atrue%7D'
from bs4 import BeautifulSoup as bs
import os,sys,re,urllib2,chardet
from qgb import U,T
if(len(sys.argv)==2):
if(len(sys.argv[1])>0):
gsuser=sys.argv[1]
else:print 'Usage:',sys.argv[0],'zhihuUserUrl or id'
if(gsuser.find(gszp)>-1):
gsuser=T.sub(gsuser,gszp,'').strip()
print 'Processing '+gsuser+' ....'
# U.mkdir('zhihu')
# os.chdir('zhihu')
# U.mkdir(gsuser)
# os.chdir(gsuser)
i=0
while(True):
break
i+=1
fn='%s.html'%i
url=gszp+gsuser+('/answers?page='+str(i))
# U.pln(url)
# fh=open('%s.html'%i,'wb')
if(a[-4-5:]!='proxy.ini'): return False
return True
s127='''[listen]
ip = 127.0.0.1'''
s000='''[listen]
ip = 0.0.0.0'''
def rep(a):
f=open(a)
s=f.read()
f.close()
s=s.replace(s127,s000)
f=open(a,'w')
f.write(s)
f.close()
print s[:44]
sq=r'"'
if(len(sc)<11 or sc==None or sc.find('proxy.ini')==-1):
sc=s
# print sc
ls=[]
sc0=T.sub(sc,sq,sq)
if(len(sc0)>5):
for i in sc.split(sq):
print U.ct()
if(isp(i)):
print i,'\n','='*33
rep(i)
sp=bs(s,"html.parser" )
im=0
ls=[]
U.setOut('d:/test/dot/lby.dot')
print 'digraph G {'
for i in sp.find_all('li'):
im+=1;si='_'+str(im);si=''
aurl,burl=None,None
for j in i.find_all(attrs={"class":"comment-hd"}):
urls = re.findall(T.REURL, str(j),re.I)11
if(0<len(urls)<3):
aurl=urls[0]
try:
if(len(urls[1])<1):break
burl=urls[1]
except:pass
break
aurl=T.sub(aurl,'/people/')
burl=T.sub(burl,'/people/')
aurl=si+aurl.replace('-','').replace('.','')
if(burl!=None):
burl=burl.replace('-','').replace('.','')
print '%s->%s'%(aurl,burl)
# else:
# print aurl
# ls.append((aurl,burl))
print '}'
U.resetOut()