A day for short answers, I guess.
At 06:01 AM 6/11/2007, Frebus wrote:
>I'm trying to read data form a CSV file into an spss datadocument by a
>script. Up to now I have managed to open the CSV file and to put the
>data in the datadocument by a syntax file which I start from the
>script.
>
> The data need to be appended in an existing datadocument. I use the
> syntax:
>
>Get DATA /Type = TXT
> /FILE = 'C:\Program Files\SPSS\decosoutput.csv'
> /DELCASE = Line
> /DELIMITERS = ";"
> /ARRANGEMENT = DELIMITED
> /FIRSTCASE = 1
> /IMPORTCASE = FIRST 1
> /VARIABLES = test1 F test2 F test3 F test4 F test5 F test6 F
>
>The variables are placed at the right top of the datadocument.
In fact, only the NEW data will be in this working file
('datadocument').
>I want to specify the location where the data need to be put in the
>script/syntax files. How can I do that? Should I make a temporary
>datadocument with the variables at the right top, and then make
>another script to change the data positions?
That's the way. If you want the new data at the end of the file, first
read the new data as you did above. Then use ADD FILES:
ADD FILES
/FILE='OldData'
/FILE=*.
SAVE OUTFILE='BothData'.
Replace 'OldData' and 'BothData' by the names of files you use,
including path.
If you want the new data somewhere in the middle of the new file, it'll
only work if you have your file sorted by certain variables, and want
the new data distributed so the new file is properly sorted.
In that case, after you read the new data,
SORT CASES BY FileKey.
ADD FILES
/FILE='OldData'
/FILE=*
/BY FileKey.
SAVE OUTFILE='BothData'.
'FileKey' is the variable you want the final file sorted by. If it's
sorted by a set of variables, you can specify that set instead of the
single variable 'FileKey'.
-Good luck,
Richard
Locked
