def showLastYears(word, years):
begdate, enddate = getTimeDomain(RFF.getDateList())
begdate = enddate - datetime.timedelta(days=years * 365)
spostdate = getPostDatebyTimeDomain(begdate, enddate,
RFF.getPostDataList())
#开始统计词频
feqlist = []
timeline = []
x = 0
xdate = begdate
print("begdate=", begdate, "enddate=", enddate)
while x <= years: #初始化频率数组
feqlist.append(0)
timeline.append(str(xdate.year) + "年")
print(str(xdate.year))
xdate += datetime.timedelta(days=365)
x += 1
#sposdate：[ [内容,作者,时间],[......],...... ]
for post in spostdate:
if post[0].find(word) > -1:
postdate = datetime.datetime.strptime(post[2], "%Y-%m-%d %H:%M")
satpos = postdate.year - begdate.year
print("satpos=", satpos, "\tpostdate=", postdate, "\tbegdate=",
begdate, "\tyear1=", postdate.year, "\tyear2=", begdate.year)
feqlist[satpos] += 1
#开始绘图
drawGraphic.linePlotGraphics(
'时间', '出现次数（帖子/回帖总数：' + str(len(spostdate)) + ')', timeline, feqlist,
'时间频率图(' + str(begdate.year) + "->" + str(enddate.year) + ")")
print('>>>>>图像加载完毕')
def showLastDays(word, days):
begdate, enddate = getTimeDomain(RFF.getDateList())
begdate = enddate - datetime.timedelta(days=days)
spostdate = getPostDatebyTimeDomain(begdate, enddate,
RFF.getPostDataList())
#开始统计词频
feqlist = []
timeline = []
x = 0
xdate = begdate
while x < days: #初始化频率数组
feqlist.append(0)
timeline.append(str(xdate.month) + "-" + str(xdate.day))
xdate += datetime.timedelta(days=1)
x += 1
#sposdate：[ [内容,作者,时间],[......],...... ]
for post in spostdate:
if post[0].find(word) > -1:
satpos = (datetime.datetime.strptime(post[2], "%Y-%m-%d %H:%M") -
begdate).days
feqlist[satpos - 1] += 1
#开始绘图
drawGraphic.linePlotGraphics(
'时间', '出现次数（帖子/回帖总数：' + str(len(spostdate)) + ')', timeline, feqlist,
'时间频率图(' + str(begdate.date()) + "->" + str(enddate.date()) + ")")
print('>>>>>图像加载完毕')
def showLastDays(authorname,days):
print("加载任务结果文件...")
buf = RFF.openResult()
datebuf = RFF.getDateList(buf)
begdate,enddate = getTimeDomain(datebuf)
del datebuf
print("计算时间区间...")
begdate = enddate - datetime.timedelta(days=days)
print("解析回帖数据...")
buf = RFF.getPostDataList(buf)
spostdate = []
if days > 0:
begdate = enddate - datetime.timedelta(days=days)
spostdate = getPostDatebyTimeDomain(begdate,enddate,buf)
else:
spostdate = getPostDatebyTimeDomain(begdate,enddate,buf)
del buf
print("开始统计.")
spostdate = getPostByAuthor(authorname,spostdate)
llen = len(spostdate)
#开始统计词频
feqlist = []
timeline = []
x = 0
xdate = begdate
if days > 30:
ommit_xlabel_per = days/30 #忽略x label的个数
ommit_xlabel_per-=1 #同上
while x<=days:
feqlist.append(0)
timeline.append(str(xdate.month)+"-"+str(xdate.day))
xdate += datetime.timedelta(days=1)
feqlist[x] = getCountByDate(xdate,spostdate)
x+=1
ppp = 0
while ppp < ommit_xlabel_per and x <= days:
feqlist.append(0)
timeline.append("")
xdate += datetime.timedelta(days=1)
feqlist[x] = getCountByDate(xdate,spostdate)
x+=1
ppp+=1
xdate -= datetime.timedelta(days=1)
timeline[len(timeline)-1] == str(xdate.date())
else:
while x < days: #初始化频率数组
feqlist.append(0)
timeline.append(str(xdate.month)+"-"+str(xdate.day))
xdate += datetime.timedelta(days=1)
feqlist[x] = getCountByDate(xdate,spostdate)
x+=1
#开始绘图
drawGraphic.linePlotGraphics('时间','出现次数（帖子/回帖总数：'+str(llen)+')',timeline,feqlist,"【"+ authorname +'】的活跃程度图('+ str(begdate.date()) + "->" + str(enddate.date()) +")")
print('>>>>>图像加载完毕')
def showLastDays(word, days):
print("加载任务结果文件...")
buf = RFF.openResult()
datebuf = RFF.getDateList(buf)
begdate, enddate = getTimeDomain(datebuf)
del datebuf
print("计算时间区间...")
begdate = enddate - datetime.timedelta(days=days)
print("解析回帖数据...")
buf = RFF.getPostDataList(buf)
spostdate = getPostDatebyTimeDomain(begdate, enddate, buf)
del buf
print("开始统计.")
#开始统计词频
feqlist = []
timeline = []
x = 0
xdate = begdate
if days > 30:
ommit_xlabel_per = days / 30 #忽略x label的个数
ommit_xlabel_per -= 1 #同上
while x <= days:
feqlist.append(0)
timeline.append(str(xdate.month) + "-" + str(xdate.day))
xdate += datetime.timedelta(days=1)
x += 1
ppp = 0
while ppp < ommit_xlabel_per and x <= days:
feqlist.append(0)
timeline.append("")
xdate += datetime.timedelta(days=1)
x += 1
ppp += 1
xdate -= datetime.timedelta(days=1)
timeline[len(timeline) - 1] == str(xdate.date())
else:
while x < days: #初始化频率数组
feqlist.append(0)
timeline.append(str(xdate.month) + "-" + str(xdate.day))
xdate += datetime.timedelta(days=1)
x += 1
#sposdate：[ [内容,作者,时间],[......],...... ]
for post in spostdate:
if post[0].find(word) > -1:
satpos = (datetime.datetime.strptime(post[2], "%Y-%m-%d %H:%M") -
begdate).days
feqlist[satpos - 1] += 1
#开始绘图
drawGraphic.linePlotGraphics(
'时间', '出现次数（帖子/回帖总数：' + str(len(spostdate)) + ')', timeline, feqlist,
"【" + word + '】的时间频率图(' + str(begdate.date()) + "->" +
str(enddate.date()) + ")")
print('>>>>>图像加载完毕')
def singleWordTF(word, datalist, scale=30):
#实现解析时间线，获取最小最大时间范围
begtime, endtime = getTimeDomain(RFF.getDateList())
print("对比日期范围：", begtime, "->", endtime)
c = endtime - begtime
blocks = int(c.days / scale)
feqlist = []
timeline = []
x = 0
#初始化频率数组
print('>>>>>开始处理.....')
xdate = begtime
while x <= blocks:
feqlist.append(0)
timeline.append(str(xdate.date()))
xdate += datetime.timedelta(days=scale)
x += 1
# [ [[帖子标题,作者,发帖时间] , [回帖列表：[回帖内容,作者,回帖时间],[回帖内容,作者,回帖时间],[[......]],.....]] ]
for post in datalist:
if post[0][0].find(word) > -1:
titledate = datetime.datetime.strptime(post[0][2],
"%Y-%m-%d %H:%M")
deltadate = titledate - begtime
feqpos = int(deltadate.days / scale)
feqlist[feqpos] += 1
replylist = post[1]
for reply in replylist:
if len(reply) < 3:
continue
if reply[0].find(word) > -1:
replydate = datetime.datetime.strptime(reply[2],
"%Y-%m-%d %H:%M")
deltadate = replydate - begtime
feqpos = int(deltadate.days / scale)
feqlist[feqpos] += 1
print('>>>>>处理完成，加载图像中.....')
print(str(feqlist))
print(str(timeline), str(feqlist))
#开始绘图
drawGraphic.linePlotGraphics(
'时间',
'出现次数（帖子/回帖总数：' + str(len(datalist * len(datalist[0][0][0]))) + ')',
timeline, feqlist, '时间频率图(' + str(begtime) + "->" + str(endtime) + ")")
print('>>>>>图像加载完毕')
def activeTimeAnaylize(authorname,days):
buf = RFF.openResult()
datebuf = RFF.getDateList(buf)
begdate,enddate = getTimeDomain(datebuf)
del datebuf
spostdate = []
buf = RFF.getPostDataList(buf)
if days > 0:
begdate = enddate - datetime.timedelta(days=days)
spostdate = getPostDatebyTimeDomain(begdate,enddate,buf)
else:
spostdate = getPostDatebyTimeDomain(begdate,enddate,buf)
del buf
spostdate = getPostByAuthor(authorname,spostdate) #[[内容,时间],[...],...]
tpostdata = sortandget(spostdate)
tpostdata = gatherbyDays(tpostdata) # [ [date,[ countlist ] ], ]
#for post in tpostdata:
# print(str(post))
#开始分析活跃时间段
#每天的情况都分析一次，然后叠加求均值
# [ [date,[ countlist ] ], ]
xvalue = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23]
FEQLIST = []
for post in tpostdata:
feqlist = [0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]
for time in post[1]:
hour = time.hour
feqlist[hour]+=1
FEQLIST.append(feqlist)
print(str(feqlist))
del tpostdata
#平均下
avgfeq = [0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]
hour = 0
for x in avgfeq:
sum = 0
for hoursum in FEQLIST:
sum+=hoursum[hour]
avgfeq[hour] = sum
hour+=1
print("after add up all :\n\n",str(avgfeq))
drawGraphic.linePlotGraphics('时间（小时）','发帖次数',xvalue,avgfeq,"【"+ authorname +'】的活跃时间段图(共 '+ str(len(FEQLIST)) +" 天数据)")
def showKeyWord(authorname,days):
buf = RFF.openResult()
datebuf = RFF.getDateList(buf)
begdate,enddate = getTimeDomain(datebuf)
del datebuf
spostdate = []
buf = RFF.getPostDataList(buf)
if days > 0:
begdate = enddate - datetime.timedelta(days=days)
spostdate = getPostDatebyTimeDomain(begdate,enddate,buf)
else:
spostdate = getPostDatebyTimeDomain(begdate,enddate,buf)
del buf
spostdate = getPostByAuthor(authorname,spostdate)
dp = ""
#开始统计关键词
#合并回帖
for post in spostdate:
dp += "。" + post[0]
del spostdate
kd = jieba.analyse.extract_tags(dp, topK=10,allowPOS=( 'n', 'v'))
print("\n\n贴吧ID：",authorname,":\n总计回帖长度（基于已有数据）:",len(dp),"\n关键词：\n")
feqlist = []
sumfeq = 0
for keyword in kd:
print(keyword,end="\t")
feqlist.append(0)
print("\n\n")
#显示条形图
#统计词频
ttt = 0
for keyword in kd:
feqlist[ttt] = dp.count(keyword)
sumfeq+=feqlist[ttt]
ttt+=1
print(str(feqlist))
drawGraphic.barHonGraphics("关键字","出现次数",kd,feqlist,"用户【"+authorname+"】的关键字")
