This is a discussion on Replace Multiple Spaces with One Space? within the SQL Server forums, part of the Microsoft SQL Server category; --> I need to map several columns of data from one database to another where the data contains multiple spaces ...
| |||||||
| FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
| ||||
| I need to map several columns of data from one database to another where the data contains multiple spaces (once occurance of a variable number or spaces) that I need to replace with a single space. What would be the most efficient way to do this? I am using SQL2K. I was thinking a function since I know of no single Transact-SQL command that can accomplish this task. |
| |||
| "Joe Cool" <joecool@home.net> wrote in message news:5pb7t2dkt37mgh0cqnm3f9luuu1rvhglkm@4ax.com... >I need to map several columns of data from one database to another > where the data contains multiple spaces (once occurance of a variable > number or spaces) that I need to replace with a single space. What > would be the most efficient way to do this? I am using SQL2K. I was > thinking a function since I know of no single Transact-SQL command > that can accomplish this task. DECLARE @FieldName varchar(5000) SET @FieldName = ' Reduce any number of' + SPACE(512) + 'spaces up to 512 in a row to a single space ' --You can reduce layers of REPLACE depending how many extra spaces you expect. SELECT REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(RTRIM(LTRI M(@FieldName)), SPACE(16), ' '), SPACE(8), ' '), SPACE(4), ' '), SPACE(2), ' '), SPACE(2), ' ') |
| |||
| On Wed, 14 Feb 2007 20:47:37 -0600, "Russ Rose" <russrose@hotmail.com> wrote: > >"Joe Cool" <joecool@home.net> wrote in message >news:5pb7t2dkt37mgh0cqnm3f9luuu1rvhglkm@4ax.com.. . >>I need to map several columns of data from one database to another >> where the data contains multiple spaces (once occurance of a variable >> number or spaces) that I need to replace with a single space. What >> would be the most efficient way to do this? I am using SQL2K. I was >> thinking a function since I know of no single Transact-SQL command >> that can accomplish this task. > >DECLARE @FieldName varchar(5000) >SET @FieldName = ' Reduce any number of' + SPACE(512) + >'spaces up to 512 in >a row to a single space >' > >--You can reduce layers of REPLACE depending how many extra spaces you >expect. > >SELECT REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(RTRIM(LTRI M(@FieldName)), >SPACE(16), ' '), SPACE(8), ' '), SPACE(4), ' '), SPACE(2), ' '), SPACE(2), ' >') > I believe you misunderstood my question. Here are some sample values: "SMITH JR" (5 spaces) "JONES JR" (7 spaces) Desired result" "SMITH JR" "JONES JR" |
| |||
| Joe Cool wrote: > On Wed, 14 Feb 2007 20:47:37 -0600, "Russ Rose" <russrose@hotmail.com> > wrote: > >> "Joe Cool" <joecool@home.net> wrote in message >> news:5pb7t2dkt37mgh0cqnm3f9luuu1rvhglkm@4ax.com... >>> I need to map several columns of data from one database to another >>> where the data contains multiple spaces (once occurance of a variable >>> number or spaces) that I need to replace with a single space. What >>> would be the most efficient way to do this? I am using SQL2K. I was >>> thinking a function since I know of no single Transact-SQL command >>> that can accomplish this task. >> DECLARE @FieldName varchar(5000) >> SET @FieldName = ' Reduce any number of' + SPACE(512) + >> 'spaces up to 512 in >> a row to a single space >> ' >> >> --You can reduce layers of REPLACE depending how many extra spaces you >> expect. >> >> SELECT REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(RTRIM(LTRI M(@FieldName)), >> SPACE(16), ' '), SPACE(8), ' '), SPACE(4), ' '), SPACE(2), ' '), SPACE(2), ' >> ') >> > > I believe you misunderstood my question. Here are some sample values: > > "SMITH JR" (5 spaces) > "JONES JR" (7 spaces) > > Desired result" > > "SMITH JR" > "JONES JR" Looks to me like his answer produces the desired result. What problem do you see with it? |
| |||
| "Ed Murphy" <emurphy42@socal.rr.com> wrote in message news:45d51798$0$28101$4c368faf@roadrunner.com... > Joe Cool wrote: > >> On Wed, 14 Feb 2007 20:47:37 -0600, "Russ Rose" <russrose@hotmail.com> >> wrote: >> >>> "Joe Cool" <joecool@home.net> wrote in message >>> news:5pb7t2dkt37mgh0cqnm3f9luuu1rvhglkm@4ax.com... >>>> I need to map several columns of data from one database to another >>>> where the data contains multiple spaces (once occurance of a variable >>>> number or spaces) that I need to replace with a single space. What >>>> would be the most efficient way to do this? I am using SQL2K. I was >>>> thinking a function since I know of no single Transact-SQL command >>>> that can accomplish this task. >>> DECLARE @FieldName varchar(5000) >>> SET @FieldName = ' Reduce any number of' + SPACE(512) + >>> 'spaces up to 512 in a >>> row to a single space ' >>> >>> --You can reduce layers of REPLACE depending how many extra spaces you >>> expect. >>> >>> SELECT REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(RTRIM(LTRI M(@FieldName)), >>> SPACE(16), ' '), SPACE(8), ' '), SPACE(4), ' '), SPACE(2), ' '), >>> SPACE(2), ' ') >> >> I believe you misunderstood my question. Here are some sample values: >> >> "SMITH JR" (5 spaces) >> "JONES JR" (7 spaces) >> >> Desired result" >> >> "SMITH JR" >> "JONES JR" > > Looks to me like his answer produces the desired result. What problem > do you see with it? It was a bit of overkill... maybe this will be closer to what you want. CREATE FUNCTION dbo.SingleSpace(@str varchar(8000)) RETURNS varchar(8000) AS --Usage --SELECT dbo.SingleSpace('Function replaces any and all spaces up to 16 in a row with a single space ') BEGIN DECLARE @s varchar(8000) SELECT @s = REPLACE(REPLACE(REPLACE(RTRIM(LTRIM(@str)), SPACE(4), ' '), SPACE(2), ' '), SPACE(2), ' ') RETURN @s END |
| |||
| Russ Rose wrote: > "Ed Murphy" <emurphy42@socal.rr.com> wrote in message > news:45d51798$0$28101$4c368faf@roadrunner.com... >> Joe Cool wrote: >> >>> On Wed, 14 Feb 2007 20:47:37 -0600, "Russ Rose" <russrose@hotmail.com> >>> wrote: >>> >>>> "Joe Cool" <joecool@home.net> wrote in message >>>> news:5pb7t2dkt37mgh0cqnm3f9luuu1rvhglkm@4ax.com... >>>>> I need to map several columns of data from one database to another >>>>> where the data contains multiple spaces (once occurance of a variable >>>>> number or spaces) that I need to replace with a single space. What >>>>> would be the most efficient way to do this? I am using SQL2K. I was >>>>> thinking a function since I know of no single Transact-SQL command >>>>> that can accomplish this task. >>>> DECLARE @FieldName varchar(5000) >>>> SET @FieldName = ' Reduce any number of' + SPACE(512) + >>>> 'spaces up to 512 in a >>>> row to a single space ' >>>> >>>> --You can reduce layers of REPLACE depending how many extra spaces you >>>> expect. >>>> >>>> SELECT REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(RTRIM(LTRI M(@FieldName)), >>>> SPACE(16), ' '), SPACE(8), ' '), SPACE(4), ' '), SPACE(2), ' '), >>>> SPACE(2), ' ') >>> I believe you misunderstood my question. Here are some sample values: >>> >>> "SMITH JR" (5 spaces) >>> "JONES JR" (7 spaces) >>> >>> Desired result" >>> >>> "SMITH JR" >>> "JONES JR" >> Looks to me like his answer produces the desired result. What problem >> do you see with it? > > It was a bit of overkill... maybe this will be closer to what you want. > > > CREATE FUNCTION dbo.SingleSpace(@str varchar(8000)) > > RETURNS varchar(8000) > > AS > --Usage > --SELECT dbo.SingleSpace('Function replaces any and all spaces up > to 16 in a row with a single space ') > > BEGIN > DECLARE @s varchar(8000) > SELECT @s = REPLACE(REPLACE(REPLACE(RTRIM(LTRIM(@str)), SPACE(4), ' '), > SPACE(2), ' '), SPACE(2), ' ') > RETURN @s > END Doesn't work on 11, 14, or 15 spaces. Adding an extra REPLACE(..., SPACE(2), ' ') wrapper will fix it, and extend it to runs up to 26. Adding an inner 8->1 on top of that extends to 167; 16->1 on top of that extends to 2447; 32->1 on top of that extends to 77343, more than enough unless you're messing with TEXT (for which REPLACE doesn't work at all, IIRC) or VARCHAR(MAX). |
| |||
| Do an UPDATE statement with nested REPLACE() functions to change (n) spaces to 1 space: UPDATE Foobar SET mystring = REPLACE (SPACE(2), SPACE(1) ' , ... REPLACE (SPACE(<<fib(n)>>), SPACE(1), mystring) .. ))); The optimal pattern for the substitutions is a Fibbonnaci series with the longest string of spaces in the innermost invocation. You have to pick the right number based on the length of the column. Working out the math is fun, so enjoy. |
| |||
| "Russ Rose" <russrose@hotmail.com> wrote: > How did you guys miss the recursion boat? Create function dbo.udf_CondenseSpaces (@str varchar(8000)) Returns varchar(8000) AS BEGIN Declare @s varchar(8000) Set @s = replace(@str, ' ', ' ') if charindex(' ', @s) > 0 set @s = dbo.udf_CondenseSpaces(@s) return @s END |
| ||||
| also instead of calling a function many times create a function or a proct Sample incomming string declare @somestring varchar(100) set @somestring = 'aaa bbb cccc eee fff f gggg h i aa ' --sample guts of function / proc while charindex(' ',@somestring)>0 begin set @Somestring = replace(@somestring,' ',' ') end print @somestring ---output aaa bbb cccc eee fff f gggg h i aa "Blackburn" <blackburn@2centsediting.mypants.com> wrote in message news:1192727141_459@sp12lax.superfeed.net... > "Russ Rose" <russrose@hotmail.com> wrote: >> > > > How did you guys miss the recursion boat? > > Create function dbo.udf_CondenseSpaces (@str varchar(8000)) > > Returns varchar(8000) > AS > > BEGIN > Declare @s varchar(8000) > > Set @s = replace(@str, ' ', ' ') > > if charindex(' ', @s) > 0 > > set @s = dbo.udf_CondenseSpaces(@s) > > return @s > END |