I have a file of LAtin for quizzes and I need UTF8 but Canvas won't accept it

Question

My Latin is uploaded and has ? and ! and other characters replacing the Latin.

IMpossible to work with

Jeff_F · Answer

Respondus 4 may help in this situation, depending on where the character corruption is occurring. Respondus 4 supports UTF 8 and extended Latin characters when the source file displays them correctly and is saved in a modern Word docx format. In those cases, importing through Respondus and publishing to Canvas can normalize the encoding and preserve the characters.

However, Respondus will not fix characters that are already corrupted in the source file. If the Latin text appears as question marks or symbols in Word or Excel before import, that encoding issue needs to be corrected first. Respondus will only pass through what it receives. Something I've done in the past is to use the Word Find/Replace feature to find repeated characters and replace with either nothing or a dash -. Then to review the dashes to see if any are supposed to be question marks, for example. And if not, find and replace with null (nothing).

So Respondus is worth trying if the characters look correct in the original file, but it is not a guaranteed fix if the problem exists at the file encoding level.

A free trial is available - Respondus 4 - Pricing - Respondus

~ Jeff

RobertBerger · Answer

I saved CSV files in both UTF-8 and plain UTF . I got to the point of seeing the quiz (with ????? embedded) and that was it. I need the UTF8 because it is Latin and there are vowels with macrons and such. For the time being I am forced to load no Latin with any such features. When I got something loaded I will consider how to get around this. I have about 15 files now in DOCX format that seem to avoid the special character issue. We'll see

Jeff_F · Answer

Hello @RobertBerger - that symptom usually points to a character encoding mismatch during the import. When extended Latin characters turn into question marks or odd punctuation after upload, it often means the file was not interpreted as UTF 8.

A quick workaround is to re save the source file explicitly as UTF 8 and then import again. If you can, use a plain text editor that lets you choose encoding and save as UTF 8 without BOM. Also check that the file uses straight quotes and standard commas, since smart quotes can sometimes get converted during export.

If you share what format you are importing (QTI zip, CSV, or another file type) and what tool created the file (Excel, Word, Google Sheets, etc.), people here can point you to the most reliable re save steps for that workflow.