def sonFavRate():
print "日期\t版本\t栏目名\t栏目进入数\t列表页菜谱点击数\t菜谱收藏数\t菜谱点击率\t菜谱收藏率\t栏目菜谱收藏率"
files = os.listdir("/home/zhangzhonghui/data/seqHit/")
for file in sorted(files):
if not file.startswith("20"):
continue
day = file[file.rfind("/") + 1:]
ts = {}
for line in open("/home/zhangzhonghui/data/seqHit/" + file):
if line.startswith("版本"):
continue
(v, stype, sret) = seqHit.readLine(line)
selfNum = 0
for st in sret.sonTypes:
if st in recipeInfo:
selfNum += DictUtil.sum(sret.sonTypes[st])
if selfNum > 0:
if recipeFavMethod in sret.gsonTypes:
favNum = DictUtil.sum(sret.gsonTypes[recipeFavMethod])
if stype not in ts:
ts[stype] = [0, 0, 0]
ts[stype][0] += sret.num
ts[stype][1] += selfNum
ts[stype][2] += favNum
print day + "\t" + v + "\t" + stype + favStr(
sret.num, selfNum, favNum)
#"\t%d\t%d\t%d\t%.4f\t%.4f\t%.4f"%(sret.num,selfNum,favNum,selfNum/(sret.num+1e-32),favNum/(selfNum+1e-32),favNum/(sret.num+1e-32))
for stype in ts:
(num, hnum, favNum) = ts[stype]
print day + "\tall_version\t" + stype + favStr(num, hnum, favNum)
def sonFavRate():
print "日期\t版本\t栏目名\t栏目进入数\t列表页菜谱点击数\t菜谱收藏数\t菜谱点击率\t菜谱收藏率\t栏目菜谱收藏率"
files=os.listdir("/home/zhangzhonghui/data/seqHit/")
for file in sorted(files):
if not file.startswith("20"):
continue
day=file[file.rfind("/")+1:]
ts={}
for line in open("/home/zhangzhonghui/data/seqHit/"+file):
if line.startswith("版本"):
continue
(v,stype,sret)=seqHit.readLine(line)
selfNum=0
for st in sret.sonTypes:
if st in recipeInfo:
selfNum+=DictUtil.sum(sret.sonTypes[st])
if selfNum > 0:
if recipeFavMethod in sret.gsonTypes:
favNum=DictUtil.sum(sret.gsonTypes[recipeFavMethod])
if stype not in ts:
ts[stype]=[0,0,0]
ts[stype][0]+=sret.num
ts[stype][1]+=selfNum
ts[stype][2]+=favNum
print day+"\t"+v+"\t"+stype+favStr(sret.num,selfNum,favNum)
#"\t%d\t%d\t%d\t%.4f\t%.4f\t%.4f"%(sret.num,selfNum,favNum,selfNum/(sret.num+1e-32),favNum/(selfNum+1e-32),favNum/(sret.num+1e-32))
for stype in ts:
(num,hnum,favNum)=ts[stype]
print day+"\tall_version\t"+stype+favStr(num,hnum,favNum)
def reportSonsFile(method, version, file, ts, dts, v2=100000):
if True:
for line in open(file):
if line.startswith("版本"):
continue
if True:
try:
(v, stype, sret) = seqHit.readLine(line)
except:
sys.stderr.write(line)
continue
tnum = sret.num + 1e-32
if tnum < 1.0:
continue
try:
v = int(v)
except:
v = 0
if v >= version and v <= v2 and stype.startswith(method):
if len(dts) <= 0 or dts[-1][0] != file:
dts.append((file, {}))
for type in sret.sonTypes:
rate = DictUtil.sum(sret.sonTypes[type])
type = hitItemName.getTypeName(type)
if type not in dts[-1][1]:
ts[type] = 1
dts[-1][1][type] = 0
dts[-1][1][type] += rate
'''
def ds(uuidFile, today):
regUserFile = "/home/zhangzhonghui/data/reg/regUser." + today
uuids = {}
for line in open(regUserFile):
cols = line.strip().split()
uuids[cols[0]] = [cols[1], ""]
for line in open(uuidFile):
cols = line.split()
if len(cols) < 4: continue
uuid = cols[0]
if uuid not in uuids: continue
day = cols[2]
if uuids[uuid][1] == "":
uuids[uuid][1] = day
else:
if uuids[uuid][1] > day:
uuids[uuid][1] = day
ds = {}
for uuid in uuids:
OK, lastDay = uuids[uuid]
if OK not in ds:
ds[OK] = {}
div = TimeUtil.daysDiv(lastDay, today)
if div not in ds[OK]:
ds[OK][div] = 1
else:
ds[OK][div] += 1
for OK in ds:
print OK, DictUtil.sum(ds[OK])
print ds[OK]
def reportSonsFile(method,version,file,ts,dts,v2=100000):
if True:
for line in open(file):
if line.startswith("版本"):
continue
if True:
try:
(v,stype,sret)=seqHit.readLine(line)
except:
sys.stderr.write(line)
continue
tnum=sret.num+1e-32
if tnum < 1.0:
continue
try:
v=int(v)
except:
v=0
if v >= version and v <= v2 and stype.startswith(method):
if len(dts) <= 0 or dts[-1][0] != file:
dts.append((file,{}))
for type in sret.sonTypes:
rate=DictUtil.sum(sret.sonTypes[type])
type=hitItemName.getTypeName(type)
if type not in dts[-1][1]:
ts[type]=1
dts[-1][1][type]=0
dts[-1][1][type]+=rate
'''
def ds(uuidFile,today):
regUserFile="/home/zhangzhonghui/data/reg/regUser."+today
uuids={}
for line in open(regUserFile):
cols=line.strip().split()
uuids[cols[0]]=[cols[1],""]
for line in open(uuidFile):
cols=line.split()
if len(cols) < 4:continue
uuid=cols[0]
if uuid not in uuids:continue
day=cols[2]
if uuids[uuid][1] == "":
uuids[uuid][1]=day
else:
if uuids[uuid][1] > day:
uuids[uuid][1]=day
ds={}
for uuid in uuids:
OK,lastDay=uuids[uuid]
if OK not in ds:
ds[OK]={}
div=TimeUtil.daysDiv(lastDay,today)
if div not in ds[OK]:
ds[OK][div]=1
else:
ds[OK][div]+=1
for OK in ds:
print OK,DictUtil.sum(ds[OK])
print ds[OK]
def count():
ts={}
tagFile="/home/zhangzhonghui/data/backup.1203/tagQuery.all.txt"
#tagFile="/home/zhangzhonghui/data/backup.1202/tag.txt"
for line in open(tagFile):
ts[line.strip()]=1
ts["v45_"+line.strip()]=1
dir=sys.argv[1]
files=os.listdir(dir)
files=sorted(files)
for file in files:
tck=clickCount.Click()
ack=clickCount.Click()
if not file.startswith("searchKeyword45.2014-"):
continue
for line in open(dir+"/"+file):
cols=line.strip().split("\t")
k=cols[0]
if k.startswith(searchKeyword45.CardFix):
continue
if k.endswith("##total##"):
(k,click)=clickCount.readClick(cols)
ack.merge(click)
continue
if k not in ts:
continue
try:
(k,click)=clickCount.readClick(cols)
except:
sys.stderr.write(line)
continue
tck.merge(click)
voidHit=0
if 0 in click.hitCount:
voidHit=click.hitCount[0]
hit=DictUtil.sum(click.hit)
if clickCount.SumMark in click.hit:
hit=click.hit[clickCount.SumMark]
if click.search==0:
click.search+=1e-12
#print k,click.search
#print "%s\t%s\t%d\t%d\t%d\t%f\t%f"%(file[-10:],k,click.search,hit,(click.search-voidHit),float(hit)/click.search,float(click.search-voidHit)/click.search)
#print click
hit=clickCount.dictSum(tck.hit)
voidHit=0
if 0 in tck.hitCount:
voidHit=tck.hitCount[0]
k="all"
if tck.search == 0:
tck.search+=1e-3
tn=float(hit)/tck.search
an=float(ack.hit[clickCount.SumMark])/ack.search
trate=float(tck.search-voidHit)/tck.search
arate=float(ack.search-ack.hitCount[0])/ack.search
#print "%s\t%f\t%f"%(file[-10:],float(tck.search)/ack.search,trate/arate)
print "%s\t%d\t%d\t%d\t%d\t%d\t%.3f\t%.3f\t%.3f\t%.3f\t%.3f"%(file[-10:],ack.search,ack.hit[clickCount.SumMark],tck.search,hit,(tck.search-voidHit),float(hit)/tck.search,trate,arate,tn/an,trate/arate)
def topnRateStr(d,n=500):
sd=sorted(d.items(),key=lambda e:e[1],reverse=True)
nn=0
sum=DictUtil.sum(d)
if sum == 0:
sum+=1e-32
s=""
for w,c in sd:
v=float(c)/sum
nn+=1
if nn > n:
break
s+="("+str(w)+",%d,%.4f"%(c,v)+")"
return "["+s+"]"
def topnRateStr(d, n=500):
sd = sorted(d.items(), key=lambda e: e[1], reverse=True)
nn = 0
sum = DictUtil.sum(d)
if sum == 0:
sum += 1e-32
s = ""
for w, c in sd:
v = float(c) / sum
nn += 1
if nn > n:
break
s += "(" + str(w) + ",%d,%.4f" % (c, v) + ")"
return "[" + s + "]"
def favRate():
files=os.listdir("/home/zhangzhonghui/data/seqHit/")
for file in sorted(files):
if not file.startswith("20"):
continue
for line in open("/home/zhangzhonghui/data/seqHit/"+file):
if line.startswith("版本"):
continue
(v,stype,sret)=seqHit.readLine(line)
if stype not in recipeInfo:
continue
selfNum=sret.num
if recipeFavMethod in sret.sonTypes:
favNum=DictUtil.sum(sret.sonTypes[recipeFavMethod])
print file+"\t"+v+"\t"+stype+"\t%d\t%d\t%.4f"%(selfNum,favNum,favNum/(selfNum+1e-32))
def favRate():
files = os.listdir("/home/zhangzhonghui/data/seqHit/")
for file in sorted(files):
if not file.startswith("20"):
continue
for line in open("/home/zhangzhonghui/data/seqHit/" + file):
if line.startswith("版本"):
continue
(v, stype, sret) = seqHit.readLine(line)
if stype not in recipeInfo:
continue
selfNum = sret.num
if recipeFavMethod in sret.sonTypes:
favNum = DictUtil.sum(sret.sonTypes[recipeFavMethod])
print file + "\t" + v + "\t" + stype + "\t%d\t%d\t%.4f" % (
selfNum, favNum, favNum / (selfNum + 1e-32))
def dictSum(d): if SumMark not in d: return DictUtil.sum(d) return d[SumMark]