Problem Reading Fixed Width Unicode

Question

I'm getting some unexpected results trying to read in a fixed width TXT file that contains unicode characters. When Alteryx reads in columns with unicode characters the field length is no longer "fixed" and seems to get shifted a couple spots.

Expectation:

Actual Result:

I've got the code page set to "Unicode UTF-8".

I've also tried hacking the XML to force the field to be "V_WString" rather than "V_String" (because this isn't an option in the Fixed Width dialog), but this still doesn't resolve.

Sample data and workflow attached. Any solutions would be very much appreciated!

read-in.yxzp

AkimasaKajitani · Answer

I make the workflow.

Firstly, I read the text file as text file that have only one field, then I split to field by RegEx tool.

At reading the text.txt, you have to read it as csv file not flat text. The option is below.
It is important that Delimiters is \0.

Input Data Setting:

RegEx tool Setting:

Result

Please refer the attached workflow.

read-in_AK.yxzp

Garrett · Answer

@Qiu - for this purpose test.txt is the "original file".

And I think I'm all set to use string parsing as last resort workaround. What I'm really looking for is a solution to get the initial Input Data tool configuration to work correctly. Thanks though!

Qiu · Answer

@Garrett 
Can you share your original file rather than the processed one?

Are you accepting other parsing method, as long as get your desired output?