django_orm操做

查詢操做和性能優化

1.基本操做

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

增   models.Tb1.objects.create(c1 = 'xx' , c2 = 'oo' )  增長一條數據，能夠接受字典類型數據  * * kwargs   obj  =  models.Tb1(c1 = 'xx' , c2 = 'oo' ) obj.save()     查   models.Tb1.objects.get( id = 123 )          # 獲取單條數據，不存在則報錯（不建議） models.Tb1.objects. all ()                # 獲取所有 models.Tb1.objects. filter (name = 'seven' )  # 獲取指定條件的數據 models.Tb1.objects.exclude(name = 'seven' )  # 獲取指定條件的數據     刪   models.Tb1.objects. filter (name = 'seven' ).delete()  # 刪除指定條件的數據     改 models.Tb1.objects. filter (name = 'seven' ).update(gender = '0' )   # 將指定條件的數據更新，均支持 **kwargs obj  =  models.Tb1.objects.get( id = 1 ) obj.c1  =  '111' obj.save()                                                  # 修改單條數據

2.Foreign key的使用緣由

1 2 3 4

約束 節省硬盤   可是多表查詢會下降速度，大型程序反而不使用外鍵，而是用單表(約束的時候，經過代碼判斷)

extra

1 2 3 4 5

extra( self , select = None , where = None , params = None , tables = None , order_by = None , select_params = None )     Entry.objects.extra(select = { 'new_id' :  "select col from sometable where othercol > %s" }, select_params = ( 1 ,))     Entry.objects.extra(where = [ 'headline=%s' ], params = [ 'Lennon' ])     Entry.objects.extra(where = [ "foo='a' OR bar = 'a'" ,  "baz = 'a'" ])     Entry.objects.extra(select = { 'new_id' :  "select id from tb where id > %s" }, select_params = ( 1 ,), order_by = [ '-nid' ])

F查詢

1 2	from  django.db.models  import  F     models.Tb1.objects.update(num = F( 'num' ) + 1 )

Q查詢

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

方式一：      Q(nid__gt = 10 )      Q(nid = 8 ) | Q(nid__gt = 10 )      Q(Q(nid = 8 ) | Q(nid__gt = 10 )) & Q(caption = 'root' )        方式二：      con  =  Q()      q1  =  Q()      q1.connector  =  'OR'      q1.children.append(( 'id' ,  1 ))      q1.children.append(( 'id' ,  10 ))      q1.children.append(( 'id' ,  9 ))      q2  =  Q()      q2.connector  =  'OR'      q2.children.append(( 'c1' ,  1 ))      q2.children.append(( 'c1' ,  10 ))      q2.children.append(( 'c1' ,  9 ))      con.add(q1,  'AND' )      con.add(q2,  'AND' )        models.Tb1.objects. filter (con)

exclude(self, *args, **kwargs)

1 2	# 條件查詢     # 條件能夠是：參數，字典，Q

　　

select_related(self, *fields)

1 2 3 4

性能相關：表之間進行join連表操做，一次性獲取關聯的數據。      model.tb.objects. all ().select_related()      model.tb.objects. all ().select_related( '外鍵字段' )      model.tb.objects. all ().select_related( '外鍵字段__外鍵字段' )

 

prefetch_related(self, *lookups)

1 2 3 4

性能相關：多表連表操做時速度會慢，使用其執行屢次SQL查詢  在內存中作關聯，而不會再作連表查詢             # 第一次 獲取全部用戶表             # 第二次 獲取用戶類型表where id in (用戶表中的查到的全部用戶ID)             models.UserInfo.objects.prefetch_related( '外鍵字段' )

　　

annotate(self, *args, **kwargs)

1 2 3 4 5 6 7 8 9 10 11 12

# 用於實現聚合group by查詢        from  django.db.models  import  Count, Avg,  Max ,  Min ,  Sum        v  =  models.UserInfo.objects.values( 'u_id' ).annotate(uid = Count( 'u_id' ))      # SELECT u_id, COUNT(ui) AS `uid` FROM UserInfo GROUP BY u_id        v  =  models.UserInfo.objects.values( 'u_id' ).annotate(uid = Count( 'u_id' )). filter (uid__gt = 1 )      # SELECT u_id, COUNT(ui_id) AS `uid` FROM UserInfo GROUP BY u_id having count(u_id) > 1        v  =  models.UserInfo.objects.values( 'u_id' ).annotate(uid = Count( 'u_id' ,distinct = True )). filter (uid__gt = 1 )      # SELECT u_id, COUNT( DISTINCT ui_id) AS `uid` FROM UserInfo GROUP BY u_id having count(u_id) > 1

 

 

extra(self, select=None, where=None, params=None, tables=None, order_by=None, select_params=None)

1 2 3 4 5 6

# 構造額外的查詢條件或者映射，如：子查詢          Entry.objects.extra(select = { 'new_id' :  "select col from sometable where othercol > %s" }, select_params = ( 1 ,))        Entry.objects.extra(where = [ 'headline=%s' ], params = [ 'Lennon' ])        Entry.objects.extra(where = [ "foo='a' OR bar = 'a'" ,  "baz = 'a'" ])        Entry.objects.extra(select = { 'new_id' :  "select id from tb where id > %s" }, select_params = ( 1 ,), order_by = [ '-nid' ])

　　

reverse(self):

1 2 3

# 倒序        models.UserInfo.objects. all ().order_by( '-nid' ).reverse()        # 注：若是存在order_by，reverse則是倒序，若是多個排序則一一倒序

　　下面兩個 取到的是對象，而且注意 取到的對象能夠 獲取其餘字段（這樣會再去查找該字段下降性能
defer(self, *fields):

1 2 3 4

models.UserInfo.objects.defer( 'username' , 'id' )         或         models.UserInfo.objects. filter (...).defer( 'username' , 'id' )         # 映射中排除某列數據

　　

only(self, *fields):

1 2 3 4

# 僅取某個表中的數據        models.UserInfo.objects.only( 'username' , 'id' )        或        models.UserInfo.objects. filter (...).only( 'username' , 'id' )