"Warning: Some character(s) could not be encoded in Latin" -- but no data loss

[calberga]

In writing keywords, I have suddenly started getting the quoted error.  But the data contains no special characters, and is correctly written.  Further, identical data is written with no warning message. and example of the data involved is:

("Alberga, Katherine Ansell" 11534 8847 38993 33538)

I note that many (but not all) of the effected files have been "losslessly" rotated in IrfanView.

[Phil Harvey]

If you send me a sample file and the exact command you are using I should be able to figure this out.  My email is philharvey66 at gmail.com

- Phil

[calberga]

I will a little later this afternoon, thanks.

[Phil Harvey]

Thanks.

The warning is because the ProfileDescriptionML contains special characters that can't be encoded in Latin.  This occurs because you are using the -L option with an implied -tagsFromFile, which extracts all information in Latin encoding.  It doesn't matter that you're not using ProfileDescriptionML because ExifTool extracts everything when you use -tagsFromFile.

- Phil

[calberga]

That seems to be the source of the warning in all of the images that triggered it.  All taken other people.  Looks as if there area bunch of languages, including Japanese.

Thanks for letting me know that I can ignore that one!

I would be tempted to switch to something other than Latin-1, but it contains anything that I can type with US-International keyboard, and the simple minded text editor I use doesn't understand anything else.

[Phil Harvey]

If you drop the -L when copying Comment to FileName, you will avoid this warning.

The -L has no effect in this case anyway -- the Comment tag doesn't have a known encoding so ExifTool can't recode it anyway.  But even if Comment was recoded, you would have problems with non-Latin characters because ExifTool would convert them to "?", which is illegal in a file name anyway.  So -L is doubly useless here.

- Phil