python之初體驗(一)

背景：又要作狗血的數據遷移、數據清洗，每次面對此類需求，內心都會把pm祖宗老老少小都問候個遍。解決辦法，要麼用java寫一大堆支撐代碼，要麼在vm上寫蹩腳的shell，都很不爽。最近嘗試了下python解決此類問題，有點小爽，心中安喜 : "BB,我不再怕pm這些腦殘需求了"。

環境準備: Ubuntu 13.04。

         #建議至少安裝Python2.7/3.2版本，畢竟同Python 2.X/3.x仍是有區別的


         sudo apt-get install python2.7 python2.7-dev

         #安裝libssl和libevent編譯環境
               sudo apt-get install build-essential libssl-dev libevent-dev libjpeg-dev libxml2-dev libxslt-dev

               #安裝mysqldb

               sudo easy_install mysql-python

               #測試

               whereis python | python -V

Python開始: 有了上面的環境準備，就能夠書寫pthon了。建立python文件,touch firstPython.py。文件名是firstPython，擴展名是py。編輯此文件，相似php、java，python也有本身的函數庫。

                 

                 主方法，python文件被執行的入口，

                

                 讀取主方法傳入的參數，sys.argv返回的是一個參數數組，sys.argv[index]。

                

                 定義成員方法，在主方法中調用執行。

                

                 profile是成員方法的參數，因爲python是弱語言類型，因此變量不須要聲明類型，這點有別於強語言類型c++、java。

                 因爲個人需求背景是數據遷移，因此在python中有效的訪問數據庫很重要。

                

                 看到這個數據庫握手鍊接、fetch數據，是否是感到很方便、簡潔，和php同樣的類庫風格。

                 python數據結構，內置類型

                 一、list：列表（動態數組， c++標準庫的vector，能夠在一個列表中包含不一樣類型的元素）

                

                 列表下標從0開始，－1是最後一個元素。取list的元素數量:len(list)。

                 建立連續的list

                 L.append(var)  #追加元素

                 L.insert(index,var)

                 L.pop(var)   #返回最後一個元素，並從list中刪除

                 L.remove(var)   #刪除第一次出現的該元素 

                 L.count(var)    #該元素在列表中出現的個數 

                 L.index(var)    #該元素的位置,無則拋異常  

                 L.extend(list)  #追加list，即合併list到L上 

                 L.sort()        #排序 

                 L.reverse()     #倒序

                 二、dictionary 字典(c++標準庫的map)

                

                 每個元素是一個pair鍵值對，key是Integer或String類型，value是任意類型。

                 dictionary的方法: 
                 D.get(key, 0)       #同dict[key]，多了個沒有則返回缺省值，0。[]沒有則拋異常  
                 D.has_key(key)      #有該鍵返回TRUE，不然FALSE  
                 D.keys()            #返回字典鍵的列表  
                 D.values()          #以列表的形式返回字典中的值，返回值的列表中可包含重複元素  
                 D.items()           #將全部的字典項以列表方式返回，這些列表中的每一項都來自於(鍵,值),可是項在返回時並無特殊的順序           
 
                 D.update(dict2)     #增長合併字典  
                 D.popitem()         #獲得一個pair，並從字典中刪除它。已空則拋異常  
                 D.clear()           #清空字典，同del dict  
                 D.copy()            #拷貝字典  
                 D.cmp(dict1,dict2)  #比較字典，(優先級爲元素個數、鍵大小、鍵值大小)  第一個大返回1，小返回-1，同樣返回0 
              
                 dictionary的複製  
                 dict1 = dict        #別名  
                 dict2=dict.copy()   #克隆，即另外一個拷貝。

 

示例代碼：

import MySQLdbimport timeimport datetimeimport sys,oshms_connections = {}transfer_connections = {}totalResult = []def queryFromHms():    print "query from hms beginning..."    db=MySQLdb.connect(host=hms_connections.get('host'),user=hms_connections.get('user'),passwd=hms_connections.get('passwd'),db=hms_connections.get('db'),port=hms_connections.get('port'))    try:        cursor = db.cursor()        resultPerDay = {}        sql = "select a.user_id,a.hotel_id,a.parent_group_id  from hotel_sub_account a inner join lm_transfer_hotel b on a.hotel_id = b.hotel_id and b.QTA_STATUS=1"        print sql        cursor.execute(sql)        results = cursor.fetchall()        for row in results:           print row           user_id  = row[0]           hotel_id =  row[1]           parent_group_id = row[2]           totalResult.append({"user_id":user_id, "hotel_id" : hotel_id, "parent_group_id" : parent_group_id})        cursor.close();    finally:        db.close();        print "function queryFromHms to close db connection...";def queryFromTransfer():    print "query from transfer beginning..."    db=MySQLdb.connect(host=transfer_connections.get('host'),user=transfer_connections.get('user'),passwd=transfer_connections.get('passwd'),db=transfer_connections.get('db'),port=transfer_connections.get('port'))    try:        for row in totalResult:            cursor = db.cursor()            sql = "select qta_id,hms_id from mapping_hms_qta_price where hms_level=1 and qta_level=1 and hms_id = %s" %(row.get('parent_group_id'))            print sql            cursor.execute(sql)            results = cursor.fetchall()            for subrow in results:                print subrow                row["qta_id"]= subrow[0]            cursor.close();    finally:        db.close();        print "function queryFromTransfer to close db connection...";def outputSupplierAccount():    print "output sql to supplier_account..."    upgradeSql   = "insert into supplier_account  (`supplier_id`, `account`, `create_time`, `is_delete`) values(%(qta_id)s, '%(user_id)s', now(), 0); \n"    callbackSql  = "delete from supplier_account where supplier_id = %(qta_id)s and account = '%(user_id)s'; \n"    upgradeFile  = open("qta_upgrade.sql", "w")    callbackFile = open("qta_callback.sql", "w")    for row in totalResult:        upgradeFile.write(upgradeSql%row)        callbackFile.write(callbackSql%row)    upgradeFile.close()    upgradeFile.close()def outputUserHotelMapping():    print "output sql to eb_auth_user_hotel_mapping..."    upgradeSql   = "insert into eb_auth_user_hotel_mapping (`user_name`, `hotel_id`, `create_time`, `hotel_seq`, `supplier_id`, `group_id`) values('%(user_id)s', '', now(), '', %(qta_id)s, %(parent_group_id)s); \n"    callbackSql  = "delete from eb_auth_user_hotel_mapping where `user_name`='%(user_id)s' and `supplier_id`=%(qta_id)s and `group_id`=%(parent_group_id)s; \n"    upgradeFile  = open("hms_upgrade.sql", "w")    callbackFile = open("hms_callback.sql", "w")       for row in totalResult:         upgradeFile.write(upgradeSql%row)        callbackFile.write(callbackSql%row)    upgradeFile.close()    upgradeFile.close()def outputUserUriMapping():    print "output sql to eb_auth_user_uri_mapping..."    upgradeFile  = open("hms_upgrade.sql", "a")    callbackFile = open("hms_callback.sql", "a")    uris = [1,2,3,5,6,7,8,9,10,21,22,24,34,35,36,37,40,41,42,43,44,46,47,49,50,54,55,56,57,58,59,60,61,62,63,76,77,78,79]    for row in totalResult:        for uri in uris:            upgradeSql   = "insert into eb_auth_user_uri_mapping(`user_name`, `uri_id`, `create_time`) values('%s', %s, now()); \n" %(row['user_id'], uri)            callbackSql  = "delete from eb_auth_user_uri_mapping where user_name='%s' and uri_id=%s; \n" %(row['user_id'], uri)            upgradeFile.write(upgradeSql%row)            callbackFile.write(callbackSql%row)    upgradeFile.close()    upgradeFile.close()def configDbProfile(profile):    print "current DB profile is %s" %(profile)    if profile == "beta":       hms_connections['host'] = ""       hms_connections['user'] = ""       hms_connections['passwd'] = ""       hms_connections['db'] = ""       hms_connections['port'] = 3306       transfer_connections['host'] = ""       transfer_connections['user'] = ""       transfer_connections['passwd'] = ""       transfer_connections['db'] = "data_transfer"       transfer_connections['port'] = 3306    elif profile == "product":       hms_connections['host'] = ""       hms_connections['user'] = ""       hms_connections['passwd'] = ""       hms_connections['db'] = "hms"       hms_connections['port'] = 3307       transfer_connections['host'] = ""       transfer_connections['user'] = ""       transfer_connections['passwd'] = ""       transfer_connections['db'] = ""       transfer_connections['port'] = 3307    elif profile == "productb":       hms_connections['host'] = ""       hms_connections['user'] = ""       hms_connections['passwd'] = ""       hms_connections['db'] = "hms"       hms_connections['port'] = 3307       transfer_connections['host'] = ""       transfer_connections['user'] = ""       transfer_connections['passwd'] = ""       transfer_connections['db'] = "data_transfer"       transfer_connections['port'] = 3308    else:       print "input parameter invalid, choose (beta | product | productb)"       sys.exit(0)if  __name__ == '__main__':   if len(sys.argv) != 2:      print "please input parameter : (beta | product | productb)"      sys.exit(0)   profile = sys.argv[1]     configDbProfile(profile)   queryFromHms();   queryFromTransfer();   outputSupplierAccount();   outputUserHotelMapping();   outputUserUriMapping();