转换
CAST([字段名] as varchar(4000))
CONVERT(varchar(4000),[字段名])
1.查询数据库内所有的表
--查询数据库内的 用户的 表,u表示用户表(就是你自己创建的表),s表示系统表
select * from sysobjects where xtype='u'
2.查询数据库内所有表内的所有字段
--查询表内的所有的列 name:表字段名称 所属表的id xtype:字段的类型用id表示的
select * from syscolumns
name:字段名称
id:所属于表的系统id
xtype:就是字段的类型ID
可以通过select * from systypes 查询到所属的类型对应的 是int还是varchar!
获取一个表内的所有字段,是系统的函数,object_id('表名') 获取到的是表在系统内的id
select * from syscolumns where id=object_id('部门收文')
and xtype in (select xtype from systypes where name in ('varchar','nvarchar','char','nchar'))--并且字段的类型你个是 nvarchar,varchar char的
3.查询系统内支持所有字段的类型
select * from systypes
4.函数quotename('表名')
当表的名字是sql server中的关键字时,需要在将表明 放在尖括号中如: ['index'] index就是一个关键字。
可以通过函数quotename来解决。
print('select * from '+quotename('sys_log'))
输出:select * from [sys_log]
quotename用法可参考:
http://blog.163.com/zangyunling@126/blog/static/164624505201132110142270/
sql server 游标参考:http://www.cnblogs.com/wudiwushen/archive/2010/03/30/1700925.html
下面存储过程利用游标 以及以上的几个方法 搜索一个数据库内的所有表的固定类型的字段。
alter procedure Full_Search(@string varchar(100)) as begin declare @tbname varchar(100) declare tbroy cursor for select name from sysobjects where xtype='u' and (name like 'T0%' or name like 'T2%') --第一个游标遍历以T0,T2开头的所有表 open tbroy fetch next from tbroy into @tbname --填充数据 while @@fetch_status=0 --假如检索到数据才处理 begin declare @colname varchar(100) declare colroy cursor for select name from syscolumns where id=object_id(@tbname) and xtype in ( select xtype from systypes where name in ('varchar','nvarchar','char','nchar') --数据类型为字符型的字段 ) --第二个游标是第一个游标的嵌套游标,遍历某个表的所有字段 open colroy fetch next from colroy into @colname while @@fetch_status=0 begin declare @sql nvarchar(4000),@j int select @sql='select @i=count(1) from ' +quotename(@tbname) +' where '+ quotename(@colname)+' like '+'''%'+@string+'%''' exec sp_executesql @sql,N'@i int output',@i=@j output --输出满足条件表的记录数 if @j>0 begin declare @v varchar(8000) set @v='select distinct * from ' +quotename(@tbname) +' where '+ quotename(@colname)+' like '+'''%'+@string+'%''' ---输入满足条件表的信息 --set @v='select distinct ' +quotename(@colname)+' from ' +quotename(@tbname) +' where '+ quotename(@colname)+' like '+'''%'+@string+'%''' exec(@v) --print @v end fetch next from colroy into @colname end close colroy --关闭游标 deallocate colroy --释放游标 fetch next from tbroy into @tbname end close tbroy deallocate tbroy end
---------------------用游标 循环 更新Gis记录 declare @id numeric(18, 0) declare @i int set @i=0 declare updateGis cursor for select C0058_ID from T0058_PointObjectGis where c0058_table='C0130_EnerBaseInfo' open updateGis --打开游标 fetch next from updateGis into @id while(@@fetch_status=0) begin update T0058_PointObjectGis set c0058_table='T0130_EnerBaseInfo' where C0058_ID=@id set @i=@i+1 --取下一条记录 fetch next from updateGis into @id end close updateGis deallocate updateGis --删除游标 print('更新记录数------------>'+convert(varchar(50),@i))
游标实例,循环表内数据,并根据参数 查询其他表内数据,拼接SQL 插入另一个表
declare @CityStr nvarchar(400) declare @Type nvarchar(10) declare @sDate datetime --开始时间 declare @eDate datetime --结束时间 declare @CityID nvarchar(10) declare @StationID nvarchar(10) declare @channel_num nvarchar(10) set @CityStr='06' set @Type='O3' set @sDate='2008-03-01' set @eDate='2013-01-20' declare C_SlideAvg cursor for select a.id as CityID,b.id as StationId,c.channel_num from TAB_BASE_GROUP as a,TAB_STATION_BASEINFO as b, tab_channel_baseinfo as c where a.id=b.group_id and a.id in(@CityStr) and b.id=c.station_id and upper(c.channel_name)=upper(@Type) and c.channel_state = 1 open C_SlideAvg --打开游标 fetch next from C_SlideAvg into @CityID,@StationID,@channel_num while(@@fetch_status=0) begin ------------------------------------------------------------------------------------------------------------------------------ declare @days int --间隔的天数 declare @i int set @i=0 set @days=datediff(day,@sDate,@eDate) declare @mDate datetime --开始循环日期相加 while @i<@days begin set @i=@i+1 set @mDate=DATEADD(day,@i,@sDate) --循环的每一天的变量 declare @h int set @h=0 while @h<24--循环24小时 begin declare @h24 datetime set @h=@h+1 --时间每次循环+1 set @h24=DATEADD(Hour,@h,@mDate) declare @oldH datetime set @oldH=DATEADD(Hour,-7,@h24) --当前小时数 减去8小时 declare @sql1 nvarchar(1000) --set @sql1='select avg(val) as SlideAvg from S'+@StationID+'HN where channel_num='''+@channel_num+''' and date_time>='+convert(varchar,@oldH,120)+' and date_time<='''+convert(varchar,@h24,120)+'''' declare @slideAvg real set @sql1='select @slideAvg=avg(val) from S'+@StationID+'HN where channel_num='''+@channel_num+''' and date_time>='''+convert(nvarchar,@oldH,120)+''' and date_time<='''+convert(nvarchar,@h24,120)+'''' --print @sql1 exec sp_executesql @sql1,N'@slideAvg real output',@slideAvg output IF @slideAvg IS NOT NULL AND @slideAvg<>'' begin declare @sql2 nvarchar(1000) set @sql2='insert into Tab_Slideavg values(''06'','''+@StationID+''','''+@channel_num+''','''+convert(varchar,@h24,120)+''','''+convert(varchar,@slideAvg,120)+''')' exec sp_executesql @sql2 end end end ------------------------------------------------------------------------------------------------------------------------------- fetch next from C_SlideAvg into @CityID,@StationID,@channel_num --取下一条记录 end close C_SlideAvg--关闭游标 deallocate C_SlideAvg
![](https://images.cnblogs.com/OutliningIndicators/ContractedBlock.gif)
1 alter procedure p_SlideAvg 2 ( 3 @CityStr nvarchar(400),--城市字符串 4 @StationID nvarchar(10),--站点编号 5 @Type nvarchar(10),--类型 6 @sDate datetime, --开始时间 7 @eDate datetime --结束时间 8 ) 9 --set @CityStr='06' 10 --set @Type='O3' 11 --set @sDate='2008-03-20' 12 --set @eDate='2008-04-20' 13 as 14 declare @CityID nvarchar(10) --城市ID 15 --declare @StationID nvarchar(10)--站点ID 16 declare @channel_num nvarchar(10)--类型的channel_num 17 18 --先删除数据 19 declare @sqlD nvarchar(1000) 20 set @sqlD='delete from TAB_SlideAvg where stationID='''+@StationID+''' and Date_Time>='''+convert(varchar,@sDate,120)+''' and date_time<='''+convert(varchar,@eDate,120)+'''' 21 exec sp_executesql @sqlD 22 23 declare C_SlideAvg cursor for select a.id as CityID,b.id as StationId,c.channel_num from TAB_BASE_GROUP as a,TAB_STATION_BASEINFO as b, 24 tab_channel_baseinfo as c where a.id=b.group_id and b.id=c.station_id 25 and upper(c.channel_name)=upper(@Type) 26 and b.id=@StationID 27 and c.channel_state = 1 28 open C_SlideAvg --打开游标 29 fetch next from C_SlideAvg into @CityID,@StationID,@channel_num 30 while(@@fetch_status=0) 31 begin 32 ------------------------------------------------------------------------------------------------------------------------------ 33 34 declare @days int --间隔的天数 35 declare @i int 36 set @i=0 37 set @days=datediff(day,@sDate,@eDate) 38 declare @mDate datetime 39 --开始循环日期相加 40 while @i<@days 41 begin 42 set @i=@i+1 43 set @mDate=DATEADD(day,@i,@sDate) --循环的每一天的变量 print convert(varchar,@mDate,120) 44 declare @h int 45 set @h=0 46 while @h<24--循环24小时 47 begin 48 declare @h24 datetime 49 --set @h24=DATEADD(Hour,@h,@mDate) 50 set @h24=dateadd(second,59,dateadd(minute,59,dateadd(hour,@h,@mDate)))--时间添加上@h个小时 59分59秒 51 declare @oldH datetime 52 set @oldH=DATEADD(Hour,-7,@h24) --当前小时数 减去7小时 53 declare @sql1 nvarchar(1000) 54 declare @slideAvg real 55 set @sql1='select @slideAvg=avg(val) from S'+@StationID+'HN where channel_num='''+@channel_num+''' and date_time>='''+convert(nvarchar,@oldH,120)+''' and date_time<='''+convert(nvarchar,@h24,120)+'''' 56 exec sp_executesql @sql1,N'@slideAvg real output',@slideAvg output 57 IF @slideAvg IS NOT NULL AND @slideAvg<>'' 58 begin 59 declare @sql2 nvarchar(1000) 60 set @sql2='insert into Tab_Slideavg values('''+@CityID+''','''+@StationID+''','''+@channel_num+''','''+convert(varchar,@h24,120)+''','''+convert(varchar,@slideAvg,120)+''')' 61 exec sp_executesql @sql2 62 end 63 set @h=@h+1 --时间每次循环+1 64 end 65 end 66 67 ------------------------------------------------------------------------------------------------------------------------------- 68 fetch next from C_SlideAvg into @CityID,@StationID,@channel_num --取下一条记录 69 end 70 close C_SlideAvg--关闭游标 71 deallocate C_SlideAvg--销毁游标