Variable name not imported correctly

[Ov-ille]

I have come across a very strange problem today.

When reading an SPSS File, (at least) one of the variables is assigned a different name, than it has in the SPSS file.
I've never noticed that behaviour before, only when struggeling with the known issue #119 with reading long string variables (they are being split into multiple variables there).
Weirdly the variable in this case is a numeric variable with just 0/1 values, so I doubt it is related to the same problem.

A solution I have found for this issue as well as #119 is to open the file in SPSS make any change to it and save it again. Then everything is imported correctly.

Reproducing the issue

• Download the following SPSS file: https://www.dropbox.com/s/p652d20d9gqigm0/test.SAV?dl=0
• execute the code:

import pyreadstat as sav
import re
# load data
data, meta = sav.read_sav(test.sav)
# access variable in question --> this will give an error!
print(data["BRANDAA_SUN_1"])
# instead the variable was renamed while importing into python:
print(data["BRANDAA"])

• open SPSS file, make any change to it and save it again (or download the following file, where I have did a minor change (added a variable label to one variable: https://www.dropbox.com/s/32du1xh0523yzb2/test2.SAV?dl=0)
• execute the same code and this time there is no error when accessing column "BRANDAA_SUN_1"

Setup Information:
pyreadstat 1.1.4 was installed with pip (I also tried it with 1.1.2)
a virtual environment created with venv
Python3.8 (plain)
Windows10, 64bit

[ofajardo]

I think the file has been created using the IBM spss dll files instead of the full application, but it should be possible to read it correctly since pspp does it correctly. I have submitted a ticket to Readstat, we have to wait for them to fix it.

[ofajardo]

another user reports a similar issue with another file, this time apparently created with SPSS #250, file added to the bug report in Readstat.