Width being automatically reset to max (32,767) upon saving
Locked
Width being automatically reset to max (32,767) upon saving
 
|
I am currently running SPSS release 21.0.0.0 and have recently run into an issue that has never occurred before. I have set the width of my string variables to 10 or 20 or 500 (depending on how large I need each to field to be for that variable). I save my data. All looks well. In the past, the next time I open the file, it shows that the widths are exactly how I set them; eg, the 10 is still 10, the 20 is still 20. I'll add more data, save, all is well.
Alas those days are over. Now, all of them are set to 32,767 (ie, the maximum value). I can reset each of the widths and save. But then it goes back to 32,767. My files are now astronomically large and do not function well. This is happening to every single string variable and only string variables. One, dataset is now actually is "dead"; ie, I get an error message saying the data are truncated. And, in fact, the data are missing; they're gone. I have a call into my University computing dept to see if I can get a back up of those data. Meanwhile, this 32,767 width issue is running amuck. I can't seem to get anything to save without it automatically being reset to the max width and turning my data into unmanageable large files. The files are stored on a networked server. I only use SPSS to add new cases (eg, I'm not importing/exporting into excel or Qualtrics or SurveyMonkey). It's just plain hand-entry of data into SPSS. Help? |
Re: Width being automatically reset to max (32,767) upon saving
Administrator
|
"I am currently running SPSS release 21.0.0.0 and have recently run into an issue that has never occurred before. "
<SNIP> I have never seen this or heard of anything like it. Perhaps this is resolved in some patch? You can look on the IBM site just as easily as I can so I will leave the research to you. Maybe Jon has more specifics?
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me. --- "Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis." Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?" |
|
I haven't heard of this problem, but do look for any V21 patches via FixCentral. However, something must have changed, since the problem is new.
One thing to try is to run ALTER TYPE with AMIN to set your strings to the minimum lossless width. If somehow you are getting long strings of nonprinting characters in the variables, the widths won't change, which would be a clue. Otherwise it would at least reset field sizes properly. It won't solve the basic problem, though. On Thursday, May 12, 2016, David Marso <[hidden email]> wrote: "I am currently running SPSS release 21.0.0.0 and have recently run into an -- ===================== 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 |
Re: Width being automatically reset to max (32,767) upon saving
|
|
In reply to this post by David Marso
If this is newly arising in a version that was running okay before, it sounds
=====================
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 me like "something" is corrupted. I don't recall any setting that would have that outcome -- but if it is not a setting, it should be the program. And this seems like a serious problem that we would have heard of here, if it was a terrible fault in one version. The usual set of corrections for program corruption is (a) re-start program, (b) re-boot system, (c) re-install program to fix the program or what is in the registration file. If a virus has corrupted the system, then you may have to re-install the system. -- Rich Ulrich > Date: Thu, 12 May 2016 08:45:06 -0700 > From: [hidden email] > Subject: Re: Width being automatically reset to max (32,767) upon saving > To: [hidden email] > > "I am currently running SPSS release 21.0.0.0 and have recently run into an > issue that has never occurred before. " > <SNIP> > I have never seen this or heard of anything like it. Perhaps this is > resolved in some patch? You can look on the IBM site just as easily as I > can so I will leave the research to you. Maybe Jon has more specifics? > > > |
Re: Width being automatically reset to max (32,767) upon saving
|
|
In reply to this post by hufflepuff
Thanks much to each of your suggestions.
I tried out the "alter type to amin" suggestion. Fortunately, it does not appear that there were any not-visible-to-me entries in the cells. So that part is good. Why it keeps defaulting to 32,767 in the first place remains a mystery. I saw a little on the IBM site about variable lengths increasing by 3 times their original size - having something to do with the local encoding vs unicode setting. So I wondered if that might have something to do with it. Perhaps if I accessed & saved the data using unicode, my colleague accessed & saved using local, then back to me with unicode, etc... But (a) 20*3 n.e. 32,767; (b) it's been working fine for years, doesn't make sense that it would start to get all wonky now; (c) sometimes, it's just me updating and saving. |
Re: Width being automatically reset to max (32,767) upon saving
|
|
If the original file is code page and you
open it in Unicode mode, all string widths are tripled -- but they would
have to be almost 11,000 to begin with to inflate to the maximum of 32,767.
Rick Oliver Senior Information Developer IBM Business Analytics (SPSS) E-mail: [hidden email] From: hufflepuff <[hidden email]> To: [hidden email] Date: 05/12/2016 03:35 PM Subject: Re: Width being automatically reset to max (32,767) upon saving Sent by: "SPSSX(r) Discussion" <[hidden email]> Thanks much to each of your suggestions. I tried out the "alter type to amin" suggestion. Fortunately, it does not appear that there were any not-visible-to-me entries in the cells. So that part is good. Why it keeps defaulting to 32,767 in the first place remains a mystery. I saw a little on the IBM site about variable lengths increasing by 3 times their original size - having something to do with the local encoding vs unicode setting. So I wondered if that might have something to do with it. Perhaps if I accessed & saved the data using unicode, my colleague accessed & saved using local, then back to me with unicode, etc... But (a) 20*3 n.e. 32,767; (b) it's been working fine for years, doesn't make sense that it would start to get all wonky now; (c) sometimes, it's just me updating and saving. -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Width-being-automatically-reset-to-max-32-767-upon-saving-tp5732145p5732149.html Sent from the SPSSX Discussion mailing list archive at Nabble.com. ===================== 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 |
|
|
In reply to this post by hufflepuff
Your problem would have nothing to do with Unicode width expansion. On Thu, May 12, 2016 at 2:35 PM, hufflepuff <[hidden email]> wrote: Thanks much to each of your suggestions. |
«
Return to SPSSX Discussion
|
1 view|%1 views
| Free forum by Nabble | Edit this page |