RE: PSPP file fails when I try to import strings

[jhwhite]

Alan,

 

I meant ‘variable names.” The only time I get a clean import with the variable names – following the traditional  PSPP format with no spaces, etc., and use  Gnemeric files. I did try to adjust the width during the import process per your suggestion – that is precisely how I first imported it. Still, it will not make the change and get the same error message.

 

 

Take care,

John

______________________________

 

Email: JHWhite@TechWriteInc.com

Website: www.techwriteinc.com

 

From: Pspp-users <pspp-users-bounces+jhwhite=techwriteinc.com@gnu.org> On Behalf Of Alan Mead
Sent: Monday, April 5, 2021 11:50 AM
To: pspp-users@gnu.org
Subject: Re: PSPP file fails when I try to import strings

 

John,

I assume from the screenshot below (and the reference to MinGW) that you are using Windows? What version of PSPP are you using? The crash sounds like a bug.

About the "variable labels," I'm confused. This term is a name SPSS has for short labels for the variables (like in your screenshot, each variable had a label of "None"). It sounds like that's what you mean, but how do you have "variable labels" in a spreadsheet? Or is the original file an SPSS SAV file? If so, PSPP will read that directly. If not, I don't know what you mean by "variable labels". Also, same thing for value labels, which can be more important.

Regarding string width, I don't have much experience importing spreadsheets into PSPP but I dislike doing it in SPSS because of the issue that is causing trouble for you: I don't know of a way to specify the length of string variables; SPSS just guesses (and in some versions, not well). This is particularly a problem when you have to read data periodically and combine them (the different string lengths prevent merging... Why that is, I don't know, BTW... So frustrating).

So, as an alternative, I would chose import data and use the wizard to configure the import then paste the syntax so you can manually adjust the variable formats. In the version of PSPP I have, you can also adjust the variable formats in the next-to-last step (see below) (in light testing, without any crashes).

The version I'm running (GNU pspp 1.5.3) is an experimental installation package of a recent PSPP created by John Darrington for testing, and is likely to be more current than what most people are running (possibly with new bugs but definitely with bug fixes). PSPP is currently working on a new way to make Windows packages. I can send anyone who trusts me the same install package John shared, or maybe John could share it again.

And finally, probably you know this, but your problem is the "Width" column, which shows the number of characters for string variables (and looks wrong). The "Columns" column shows how the string values will be shown and that's only cosmetic.

-Alan

On 4/5/2021 10:13 AM, jhwhite@techwriteinc.com wrote:

Hi,

 

I have been working with PSPP.  I have a file in Excel, and unfortunately, PSPP does not import Excel – at least from what I can see. When I use the CSV version, I lose all of my variable labels.  I did some research and found Gnemeric files should import OK. I found this in one of my PSPP reference books, “GNU PSPP Reference Manual.” I imported my Excel files into Gnemeric.  All of the fields are as they should be. I can successfully import the file into PSPP, and all of my variable labels are showing.  When imported into PSPP, all strings go to a default width of 8, where I need 45 for the width.  When I change the width during the import process, I get an error message.  I am not sure if this system shows email messages in HTML, but below is an image of the problem.   The message is “Assertion Failed! “_expression_: caseproto_get_width (c->proto, case idx)==var_get_width (v).” Any suggestions?

 

 

 

Take care,

John

______________________________

 

Email: JHWhite@TechWriteInc.com

Website: www.techwriteinc.com

 

-- 

Alan D. Mead, Ph.D.

President, Talent Algorithms Inc.

science + technology = better workers

http://www.alanmead.org

The irony of this ... is that the Internet is

both almost-infinitely expandable, while at the

same time constrained within its own pre-defined

box. And if that makes no sense to you, just

reflect on the existence of Facebook. We have

the vastness of the internet and yet billions

of people decided to spend most of them time

within a horribly designed, fake-news emporium

of a website that sucks every possible piece of

personal information out of you so it can sell it

to others. And they see nothing wrong with that.

-- Kieren McCarthy, commenting on why we are not 

                    all using IPv6