Diacritics in UTF-8 Glossary File Not Working about deepl-python HOT 3 OPEN

Hi @DigitalProf Mike, thanks for creating this issue.

I reproduced the problem that the first term contains the byte order mark (BOM); there seems to be an error on our side. I've reported this issue to the team.

I also looked into Excel. Exporting using "CSV UTF-8 (Comma delimited)" gives the correct encoding, but includes the BOM. Unfortunately I could not find an easy way to omit the BOM.

As a workaround (until we can resolve the BOM issue on our side), could you try entering a dummy first entry in the CSV? For example "entry-to-be-ignored,entry-to-be-ignored". Your remaining glossary entries should be unaffected and work correctly. Please make sure the entries appear correctly in Excel -- when I open your link above, many of the entries already include wrong characters (I guess because Excel assumed the wrong encoding as the file does not contain a BOM).

from deepl-python.

Thanks for the action on this issue, Daniel! I apologize for how the link to the Excel file works. The access to a OneDrive file via the browser does not give one a chance to state that the file is in fact in UTF-8. When the file is opened in Excel, the software asks for confirmation that the file is indeed in UTF-8. I should have tested the link myself. Sorry about that! :-)

As to the Byte Order Mark (BOM), Excel does in fact place that into the file by default. I have checked, but do not see how to do otherwise for exporting from Excel. I have, however, tested this aspect of the problem by opening the file in Notepad++ and changing the encoding scheme to remove the BOM. I have tested that, but it does not change how DeepL handles the file.

Last night, I sent along the Python code I used to upload the glossary. The code is from your site, but in copying the code into my message last evening, I believe that I now see the problem. I have checked this out, but I am thinking that I simply need to open the file in Python with UTF-8 encoding by adding this:

, encoding="utf-8"

If this is in fact the issue, I suggest that the sample code be changed on GitHub. It appears about half-way down the page on the GitHub site in the section, “Creating a glossary.”

Cheers,

Mike

Python Code for Creating a Glossary

from deepl-python.

from deepl-python.