mouse-based cropping

[ForrestHilton]

OCR desktop provides a gui interface that with a number of options. It generates a list of the rectangles of each word, which could potentially be used to facilitate word highlighting for the screen reader. My goal is to allow a screenshot to be taken similar to gnome-screenshot area selection and still know the position of every word.

some options are:

• Add an option with -f to make OCR desktop search the entire screen for the partial area screenshot that it was passed -- probably, something similar would be necessary to find the position of text copied by the user so it could be highlighted as it is read even though it was accessible to begin with.
• Patch gnome-screenshot to print the position of the screenshot and patch OCR desktop to accept that as an option with -f
• Rewrite this file in python and include it in ocrdesktop: https://gitlab.gnome.org/GNOME/gnome-screenshot/-/blob/master/src/screenshot-area-selection.c -- this would allow us to define an area for repeated screen shots if desired

Does one of these seam like the most practical?

Please bear in mind that I am a high school student with mediocre time management skills. Also considered that we are discussing libraries that I know very little about.

[chrys87]

howdy,

sorry for the late answer. life is sometimes very busy and complex. is this still a thing?