Alteryx Designer Desktop Discussions
Find answers, ask questions, and share expertise about Alteryx Designer Desktop and Intelligence Suite.
Good afternoon,
I just got a new laptop and I did a full install of Alteryx, and now when I run an existing module I am getting an error I have not seen before:
"Want a narrow string, but source has wide character: '?', U+FFFD"
I am running Alteryx 2024.2.1.14
New Comp Specs:
HP ZBook Power G11
32 GB RAM
AMD Ryzen 7 Pro
Windows 11
OS Build: 26100.2033
This module ran fine on old PC and I never got this error. Only change is new install and new PC...
Output tool settings...
Solved! Go to Solution.
Just ruling out some of the obvious candidates. Could you have an illegal character that's being written out to the file?
This could be caused from your input files, for instance, using a different encoding page to what you previous used i.e. you may be using UTF-16 which apparently is a wide character set vs previously using UTF-8 which supposedly uses a narrow character set.
The indicator here is that code page on your output tool saying "ISO latin ...". You could try to change that to UTF-16 or UTF-8 (although I don't think UTF-8 is accepting the wide characters). But then again, you could consider changing the code page for ingestion instead. The error tells me it's the code page choice though on either source or target that's not aligning.
If it says "the source is wide" then the source is likely UTF16 and you'll therefore need UTF16 for the output. Or change the inputs to UTF8 and output to UTF8.
Let me know if that provides any help @ScubaGeekNV
All the best,
BS
Thanks for the suggestion. Both the input and output codes were ISO 8859-1 Latin 1
I can change the codes for the Input tool, but the Output Tool only gives me two Options....
@ScubaGeekNV, awesome.
Yes, it's definitely a code page mismatch.
Being as you've only got a limited set of code pages for the output, you may want to consider actually removing the wide characters from your data. I'd suggest following this community post, it has some good formulas you could lift and shift and try out: https://community.alteryx.com/t5/Alteryx-Designer-Desktop-Discussions/Elegantly-remove-all-ASCII-cha... .. I assume you could upload after this. Path with least resistance.
I don't know if your source data already contains these corrupted characters? You would be able to check if you change the Input code page to UTF8 or UTF16 and check if there's any conversation errors (I can see 50 in your screenshot). If they still remain, it's likely that the source files were created from an Output Tool in a previous workflow that used a more limited code page like ISO 8859-1 Latin I, and it could be good to investigate changing this to UTF-8 and this could already handle.
All the best,
BS
