Python 處理數(shù)據(jù)的實例詳解
最近用python(3.2的版本)寫了根據(jù)特定規(guī)則,處理數(shù)據(jù)的一個小程序,用到了一些python常用的基礎(chǔ)知識,在此總結(jié)一下:
1,python讀文件
2,python寫文件
3,python的流程控制
4,python的for循環(huán)
5,python的集合,或字符串里判斷是否存在某個元素
6,python的邏輯或,邏輯與
7,python的正則過濾
8,python的字符串忽略空格,和以某個字符串開頭和按某個字符拆分成list
python的打開文件的模式:
關(guān)于open 模式:
w 以寫方式打開,
a 以追加模式打開 (從 EOF 開始, 必要時創(chuàng)建新文件)
r+ 以讀寫模式打開
w+ 以讀寫模式打開 (參見 w )
a+ 以讀寫模式打開 (參見 a )
rb 以二進制讀模式打開
wb 以二進制寫模式打開 (參見 w )
ab 以二進制追加模式打開 (參見 a )
rb+ 以二進制讀寫模式打開 (參見 r+ )
wb+ 以二進制讀寫模式打開 (參見 w+ )
ab+ 以二進制讀寫模式打開 (參見 a+ )
處理代碼如下:
def showtxt(path,outpathname,detailpath): greenpath=r"C://Users//qindongliang//Desktop//tnstxt//green.txt"; redpath=r"C://Users//qindongliang//Desktop//tnstxt//red.txt"; redset=listtxt(redpath) greenset=listtxt(greenpath) print("紅色詞數(shù)量: ",len(redset)) print("綠色詞數(shù)量: ",len(greenset)) #符合1條件的內(nèi)容寫入 f1=open(r"C:/Users/qindongliang/Desktop/tnstxt/result//"+detailpath+"http://1.txt",encoding="UTF-8",mode="a+") #符合2條件的內(nèi)容寫入 f2=open(r"C:/Users/qindongliang/Desktop/tnstxt/result//"+detailpath+"http://2.txt",encoding="UTF-8",mode="a+") #符合3條件的內(nèi)容寫入 f3=open(r"C:/Users/qindongliang/Desktop/tnstxt/result//"+detailpath+"http://3.txt",encoding="UTF-8",mode="a+") #符合4條件的內(nèi)容寫入 f4=open(r"C:/Users/qindongliang/Desktop/tnstxt/result//"+detailpath+"http://4.txt",encoding="UTF-8",mode="a+") delcount=1; f=open(path,encoding="UTF-8",mode="r+") fnew=open(outpathname,encoding="UTF-8",mode="a+") flog=open(outpathname+".log",encoding="UTF-8",mode="a+") #count=1; for line in f: list=line.strip().split("/t") line=line.strip() catalogid=list[0] score=list[1] keyword=clear(list[4].strip()) if keyword in redset: if catalogid.startswith("018022") or catalogid.startswith("018035") or catalogid.startswith("014023003") : f1.write(line+"/n")#符合1條件寫入 fnew.write(line+"/n")#符合1條件寫入 else: flog.write(line+" 不符合條件1 "+"/n") delcount=delcount+1 if keyword in greenset: if not (catalogid.startswith("018022") or catalogid.startswith("018035") or catalogid.startswith("014023003")) : fnew.write(line+"/n") else: f2.write(line+"/n") flog.write(line+" 不符合條件2"+"/n") delcount=delcount+1 flist=formatStrList(keyword) if "sexy" in flist or "sex" in flist: if catalogid.startswith("018022") or catalogid.startswith("018035") or catalogid.startswith("014023003") : f3.write(line+"/n") fnew.write(line+"/n") else: flog.write(line+" 不符合條件3"+"/n") delcount=delcount+1 #if (keyword.find("underwear")!=-1) & keyword.find("sexy")==-1 & keyword.find("sex")==-1: if "underwear" in flist and "sexy" not in flist and "sex" not in flist: if catalogid.startswith("014032") : f4.write(line+"/n") fnew.write(line+"/n") else: flog.write(line+" 不符合條件4"+"/n") delcount=delcount+1 #print(list[0]," ",list[1]," ",list[4]) #print() flog.write("刪除總數(shù)目: "+str(delcount)) f.close() f1.close() f2.close() f3.close() f4.close() fnew.close() flog.close() import re def clear(str): str=re.sub("[/"/"/'/'+]","",str) return str def formatStrList(keyword): list=keyword.split(" ") for item in list: item.strip(); return list def listtxt(path): f=open(path,encoding="UTF-8") s=set() for line in f: s.add(line.strip()) f.close() return s path1=r"C://Users//qindongliang//Desktop//tnstxt//highfrequency.txt" pathout1=r"C://Users//qindongliang//Desktop//tnstxt//detail//a_highfrequency.txt" detail1path="highfrequency" path2=r"C://Users//qindongliang//Desktop//tnstxt//highfrequency_d1.txt" pathout2=r"C://Users//qindongliang//Desktop//tnstxt//detail//b_highfrequency_d1.txt" detail2path="highfrequency_d1" #showtxt(path1,pathout1,detail1path) showtxt(path2,pathout2,detail2path)
新聞熱點
疑難解答