SPSSX Discussion

Normalize a String1

Classic

List

Threaded

7 messages Options

Libardo López Guzmán

Normalize a String1

Hi dear Listers.
I would appreciate your help to obtain a Sintax SPSS v15 to normalize a
String (millions cases). Sorry for my mistake in the past post.
The problem is that i have spaces (1,2,3....), tabulators (1,2....) in
between as separators and it must be replaced with one space.
Let me show some examples. The string may have up to 12 parts.
Sara Smith Dockter
Lee Hunter Casidy
George Harvey Mora Aito
Marck Mack

After normalization,

Sara Smith Dockter
Lee Hunter Casidy
George Harvey Mora Aito
Marck Mack

Thanks for your help,

Libardo

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

Albert-Jan Roskam

Re: Normalize a String1

Hi,

A not-so-elegant solution is below. It assumes that the string variable with the names is the first variable in the file.
It also assumes that a directory d:/temp is present. It would be nicer to skip the step with the auxiliary txt file.

Cheers!!
Albert-Jan

* sample data.
data list free / oldvalue (a60).
begin data
'Sara Smith Dockter '
' Lee Hunter Casidy'
'George Harvey Mora Aito '
' Marck Mack '
end data.

*actual code.
compute casenum = $casenum.
exe.
dataset name source.

begin program.
import spss, re

f = open('d:/temp/workfile.txt', 'w')
dataCursor = spss.Cursor([0])
f.write("casenum\tnewvalue\r\n")
for i in range(spss.GetCaseCount()):
oldval = dataCursor.fetchone()[0]
newval = re.sub(r"\s+", ' ', oldval)
newval = newval.strip()
writestr = str(i + 1) + "\t" + newval + "\r\n"
f.write(writestr)
dataCursor.close()
f.close()
end program.

GET DATA /TYPE = TXT /FILE = 'D:\temp\workfile.txt'
/DELCASE = LINE /DELIMITERS = "\t"
/ARRANGEMENT = DELIMITED
/FIRSTCASE = 2 /IMPORTCASE = ALL
/VARIABLES = casenum F1.0 newvalue A25 .

match files / file = * / file = source / by = casenum.
exe.
dataset close all.

----- Original Message ----
From: Libardo Lopez <[hidden email]>
To: [hidden email]
Sent: Saturday, January 24, 2009 2:09:47 PM
Subject: Normalize a String1

Hi dear Listers.
I would appreciate your help to obtain a Sintax SPSS v15 to normalize a
String (millions cases). Sorry for my mistake in the past post.
The problem is that i have spaces (1,2,3....), tabulators (1,2....) in
between as separators and it must be replaced with one space.
Let me show some examples. The string may have up to 12 parts.
Sara Smith Dockter
Lee Hunter Casidy
George Harvey Mora Aito
Marck Mack

After normalization,

Sara Smith Dockter
Lee Hunter Casidy
George Harvey Mora Aito
Marck Mack

Thanks for your help,

Libardo

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

Oliver, Richard

Re: Normalize a String1

The multiple space problem can be solved fairly easily without resorting to Python:

loop if index(rtrim(stringvar), " ")>0.
compute stringvar=replace(stringvar, " ", " ").
end loop.

I thought there was a way to specify tab characters in command syntax, but it isn't what I thought it was; so I don't have a simple syntax solution for the tab problem.

________________________________

From: SPSSX(r) Discussion on behalf of Albert-jan Roskam
Sent: Sat 1/24/2009 12:39 PM
To: [hidden email]
Subject: Re: Normalize a String1

Hi,

A not-so-elegant solution is below. It assumes that the string variable with the names is the first variable in the file.
It also assumes that a directory d:/temp is present. It would be nicer to skip the step with the auxiliary txt file.

Cheers!!
Albert-Jan

* sample data.
data list free / oldvalue (a60).
begin data
'Sara Smith Dockter '
' Lee Hunter Casidy'
'George Harvey Mora Aito '
' Marck Mack '
end data.

*actual code.
compute casenum = $casenum.
exe.
dataset name source.

begin program.
import spss, re

f = open('d:/temp/workfile.txt', 'w')
dataCursor = spss.Cursor([0])
f.write("casenum\tnewvalue\r\n")
for i in range(spss.GetCaseCount()):
oldval = dataCursor.fetchone()[0]
newval = re.sub(r"\s+", ' ', oldval)
newval = newval.strip()
writestr = str(i + 1) + "\t" + newval + "\r\n"
f.write(writestr)
dataCursor.close()
f.close()
end program.

GET DATA /TYPE = TXT /FILE = 'D:\temp\workfile.txt'
/DELCASE = LINE /DELIMITERS = "\t"
/ARRANGEMENT = DELIMITED
/FIRSTCASE = 2 /IMPORTCASE = ALL
/VARIABLES = casenum F1.0 newvalue A25 .

match files / file = * / file = source / by = casenum.
exe.
dataset close all.

----- Original Message ----
From: Libardo Lopez <[hidden email]>
To: [hidden email]
Sent: Saturday, January 24, 2009 2:09:47 PM
Subject: Normalize a String1

Hi dear Listers.
I would appreciate your help to obtain a Sintax SPSS v15 to normalize a
String (millions cases). Sorry for my mistake in the past post.
The problem is that i have spaces (1,2,3....), tabulators (1,2....) in
between as separators and it must be replaced with one space.
Let me show some examples. The string may have up to 12 parts.
Sara Smith Dockter
Lee Hunter Casidy
George Harvey Mora Aito
Marck Mack

After normalization,

Sara Smith Dockter
Lee Hunter Casidy
George Harvey Mora Aito
Marck Mack

Thanks for your help,

Libardo

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

====================To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

Maguin, Eugene

Re: Normalize a String1

Richard,

Are you saying that after running the syntax below, the string

'multiple space problem'

Will go to

'multiple space problem '?

Thus, the replace command will pull all characters to the right of the ' '
to the left by one place.

>>>The multiple space problem can be solved fairly easily without resorting
to Python:

loop if index(rtrim(stringvar), " ")>0.
compute stringvar=replace(stringvar, " ", " ").
end loop.

Gene Maguin

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

Peck, Jon

Re: Normalize a String1

In reply to this post by Oliver, Richard

To get rid of tabs, run something like
compute strvar = replace(strvar, string(09, pib1),' ').

Do this before replacing blanks in case there are mixed strings and blanks.

(Of course, there is a nicer way to do all this in Python with regular expressions, but I won't go there today.)

HTH,
Jon Peck

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Oliver, Richard
Sent: Saturday, January 24, 2009 12:55 PM
To: [hidden email]
Subject: Re: [SPSSX-L] Normalize a String1

The multiple space problem can be solved fairly easily without resorting to Python:

loop if index(rtrim(stringvar), " ")>0.
compute stringvar=replace(stringvar, " ", " ").
end loop.

I thought there was a way to specify tab characters in command syntax, but it isn't what I thought it was; so I don't have a simple syntax solution for the tab problem.

________________________________

From: SPSSX(r) Discussion on behalf of Albert-jan Roskam
Sent: Sat 1/24/2009 12:39 PM
To: [hidden email]
Subject: Re: Normalize a String1

Hi,

A not-so-elegant solution is below. It assumes that the string variable with the names is the first variable in the file.
It also assumes that a directory d:/temp is present. It would be nicer to skip the step with the auxiliary txt file.

Cheers!!
Albert-Jan

* sample data.
data list free / oldvalue (a60).
begin data
'Sara Smith Dockter '
' Lee Hunter Casidy'
'George Harvey Mora Aito '
' Marck Mack '
end data.

*actual code.
compute casenum = $casenum.
exe.
dataset name source.

begin program.
import spss, re

f = open('d:/temp/workfile.txt', 'w')
dataCursor = spss.Cursor([0])
f.write("casenum\tnewvalue\r\n")
for i in range(spss.GetCaseCount()):
oldval = dataCursor.fetchone()[0]
newval = re.sub(r"\s+", ' ', oldval)
newval = newval.strip()
writestr = str(i + 1) + "\t" + newval + "\r\n"
f.write(writestr)
dataCursor.close()
f.close()
end program.

GET DATA /TYPE = TXT /FILE = 'D:\temp\workfile.txt'
/DELCASE = LINE /DELIMITERS = "\t"
/ARRANGEMENT = DELIMITED
/FIRSTCASE = 2 /IMPORTCASE = ALL
/VARIABLES = casenum F1.0 newvalue A25 .

match files / file = * / file = source / by = casenum.
exe.
dataset close all.

----- Original Message ----
From: Libardo Lopez <[hidden email]>
To: [hidden email]
Sent: Saturday, January 24, 2009 2:09:47 PM
Subject: Normalize a String1

Hi dear Listers.
I would appreciate your help to obtain a Sintax SPSS v15 to normalize a
String (millions cases). Sorry for my mistake in the past post.
The problem is that i have spaces (1,2,3....), tabulators (1,2....) in
between as separators and it must be replaced with one space.
Let me show some examples. The string may have up to 12 parts.
Sara Smith Dockter
Lee Hunter Casidy
George Harvey Mora Aito
Marck Mack

After normalization,

Sara Smith Dockter
Lee Hunter Casidy
George Harvey Mora Aito
Marck Mack

Thanks for your help,

Libardo

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=======
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

Oliver, Richard

Re: Normalize a String1

In reply to this post by Maguin, Eugene

It will keep going through the string value, replacing all instances of two spaces with one space until there are no more instances of two consecutive spaces.

________________________________

From: SPSSX(r) Discussion on behalf of Gene Maguin
Sent: Sat 1/24/2009 2:16 PM
To: [hidden email]
Subject: Re: Normalize a String1

Richard,

Are you saying that after running the syntax below, the string

'multiple space problem'

Will go to

'multiple space problem '?

Thus, the replace command will pull all characters to the right of the ' '
to the left by one place.

>>>The multiple space problem can be solved fairly easily without resorting
to Python:

loop if index(rtrim(stringvar), " ")>0.
compute stringvar=replace(stringvar, " ", " ").
end loop.

Gene Maguin

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

====================To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

Libardo López Guzmán

Re: Normalize a String1

Thanks so much to all of you.

With your help, the syntax to solve my needs is:

Compute oldvalue = ltrim(oldvalue).

*To get rid of tabs.

compute oldvalue = replace(oldvalue, string(09, pib1),' ').

loop if index(rtrim(oldvalue), " ")>0.
compute oldvalue=replace(oldvalue, " ", " ").
end loop.
execute.

Tanks again,

Libardo

On Sat, Jan 24, 2009 at 5:53 PM, Oliver, Richard <[hidden email]> wrote:

> It will keep going through the string value, replacing all instances of two
> spaces with one space until there are no more instances of two consecutive
> spaces.
>
> ________________________________
>
> From: SPSSX(r) Discussion on behalf of Gene Maguin
> Sent: Sat 1/24/2009 2:16 PM
> To: [hidden email]
> Subject: Re: Normalize a String1
>
>
>
> Richard,
>
> Are you saying that after running the syntax below, the string
>
> 'multiple space problem'
>
> Will go to
>
> 'multiple space problem '?
>
> Thus, the replace command will pull all characters to the right of the ' '
> to the left by one place.
>
>
> >>>The multiple space problem can be solved fairly easily without resorting
> to Python:
>
> loop if index(rtrim(stringvar), " ")>0.
> compute stringvar=replace(stringvar, " ", " ").
> end loop.
>
> Gene Maguin
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD
>
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD
>

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD