Re: strange symbols in write outfile
Posted by
Jon K Peck on
Aug 22, 2013; 2:11pm
URL: http://spssx-discussion.165.s1.nabble.com/strange-symbols-in-write-outfile-tp5721679p5721682.html
The first three bytes in the output are
the Byte Order Mark (BOM) that indicate that this is a Unicode text file.
You don't see them in Notepad, because
Notepad correctly interprets them as a non-data part of the file and interprets
the contents as Unicode characters. If you save the file from Notepad
choosing Save As with the encoding set to Ascii, they will be removed.
If you want the file in the old-style
code page encoding, you can add ENCODING="Locale" to the syntax
or run Statistics in code page mode.
In the example below, the actual character
codes would be the same in both code page and Unicode modes.
Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: 720-342-5621
From:
"Maguin, Eugene"
<[hidden email]>
To:
[hidden email],
Date:
08/22/2013 08:04 AM
Subject:
[SPSSX-L] strange
symbols in write outfile
Sent by:
"SPSSX(r)
Discussion" <[hidden email]>
I’m writing an output data file using
write outfile and the first line has some strange symbols in it prior to
the data record.
 30 4 3 1 1 4.000000 4.000000 3.750000
2.000000 3.400000 1 2 3 1
I may be wrong but I think these magically
appeared with 21. They do something, I’m sure, (and I think that something
may have been mentioned on the list before but I don’t recall); I’m curious
what it is. Can they be suppressed (because their presence screws up the
data read in Mplus, which could be a limitation in that program). And,
the surprising thing, the damn things can’t be seen or removed in notepad.
Gene Maguin