Hey Ryan,

I am not sure why you are receiving this error if the data is UTF-8.  I know 
there is a check in the migration tool to make sure that failures due to 
non-UTF-8 characters do not cause the migration tool to fail and stop, but 
instead the tool logs the error and moves on to the next record.

Would it be possible for you to provide me with a copy of your AT database or 
could you export one (or several) of the resources that is failing from your AT 
instance and send it to me? Having an example that is failing incorrectly will 
definitely help me determine the appropriate fix for this.


Laney McGlohon
ArchivesSpace Tech Lead
laneymcglohon Skype

From: <archivesspace_users_group-boun...@lyralists.lyrasis.org> on behalf of 
Ryan Rotter <rrot...@umich.edu>
Reply-To: Archivesspace Users Group 
Date: Tuesday, February 6, 2018 at 5:13 PM
To: Archivesspace Users Group <archivesspace_users_group@lyralists.lyrasis.org>
Subject: [Archivesspace_Users_Group] Encoding errors while importing from 
Archivist Toolkit

While running the ArchivesSpace AT Migration plugin I have users repeatedly 
seeing errors like this:

(Encoding::UndefinedConversionError) "\xB4" from ASCII-8BIT to UTF-8
You're seeing this error because you use JRuby::Rack::ErrorApp::ShowStatus.

This error is what ruby should throw while attempting to reencode invalid ASCII 
to UTF-8 but, I've checked the data and it's not ASCII at all, it's valid 
UTF-8. The strings it hangs up on are all multibyte characters. Why would 
ASpace think the data is ASCII? Is there a way to force the plugin to properly 
represent the encoding to ASpace to prevent this error?

Ryan Rotter
Library IT
University of Michigan
Archivesspace_Users_Group mailing list

Reply via email to