I have number of image description tags that use the accented word: "née" . I have tried "-codedcharacterset=utf8" But all I get in the tagged image is "n?e". I am new to the forum and the excellent app.
Background: During lockdown, I am digitising 35mm negs from the pre-digital age (1980s). I am using a Microsoft Access database to catalogue the new digital images. I'm exporting a csv with GPS, date/time, addresses, comments, titles etc and bulk tagging with ExifTool. Everything is working well except accented characters. My virtual Windoze PC and Linux host use the GB English character set as default.
Also, line feed seems to work well in the images when read by the likes of DigiKam. But Google Photos is ignoring them. I know from the distant past that CR/LF is a bit of a minefield. I also know that CR/LF may need to be "escaped" in some way.
I am sure it is an easy fix but do not know where to look.
There are lots of variables when it comes to special characters. What O/S are you using? It should work fine in UTF-8 on Linux. But see FAQ 10 (https://exiftool.org/faq.html#Q10) for more help on this.
Regarding the CR/LF, FAQ 21 (https://exiftool.org/faq.html#Q21) deals with this.
- Phil
Thanks Phil
I'm very grateful.
I apologise for my ineptitude and failure fully to read the FAQs.
I've never been a programmer & I'm long retired from any work activity. It is also a long time since I've dabbled with any kind of coding.
The comment text is created in MS Access running Win10 64 bit in a virtual client but I confess I am copying and pasting "née" from a web search. So I may need to look closer at the char set held by Access. The export from Access is to a csv file on Linux. I am running Ubuntu Linux 20.04 and it reports: LANG=en_GB.UTF-8.
When I display the csv in LibreOffice calc, I see the correct display of "née".
But when I run:
exiftool -P -codedcharacterset=utf8 -csv="/path/file.csv" *.*
I see some messages about illegal characters. I end up with the tags reported as "n?e".
I believe I should be able to sort out the LF CR issue after I study FAQ 21. I did a lot of work on this years ago and wrote some code involving chr(13) + chr(10) to handle a similar issue creating imports to Google Contacts from Access. I recall I had to use \n in the vCards I created. I will study the FAQ and find my old Access code.
If you have any further pointers after reading this I should be grateful.
Best wishes Tom
Quote from: HimAgain on July 03, 2020, 03:24:36 PM
exiftool -P -codedcharacterset=utf8 -csv="/path/file.csv" *.*
I see some messages about illegal characters. I end up with the tags reported as "n?e".
This is an indication that UTF-8 wasn't used on the command line. If you have problems with determining the character set on the command line, try using the
-@ option and reading the arguments from a UTF-8 text file to be sure you have the correct encoding.
- Phil