@meatbag I'm on linux and the best I have found working for me is #ocrmypdfhttps://github.com/ocrmypdf/OCRmyPDF It uses #tesseract under the hood and for static text it's okay. For tables and other material that is difficult to parse it's not usefull. When PDF has a text then the tools I am using for reading these include #firefox and #evince
I know it is not going to resonate much with people in English speaking countries, but this has been a major issue for people who use alphabets with special characters (my native Czech included).
For years #PDF viewers on #Linux such as #Evince couldn't save characters outside the basic charsets in PDF forms. It turned out to be a complex problem and multi-year effort, but it's finally done and you can use unicode in PDF forms. First available in #Fedora 37 where we've already backported it.
📱 On my way to #Debconf22 and it was the first time I could show the QR code of a train ticket on #phosh's lockscreen to the conductor (had to resort to unlock ➡️ nautilus ➡️ evince so far).
For that I stitched together a lockscren plugin that leverages #evince's libevince. Hope the other tickets will works just as well tomorrow.
