Gimp 3 - python to edit the selection

Gimp 3 - python to edit the selection - Printable Version

+- Gimp-Forum.net (https://www.gimp-forum.net)
+-- Forum: GIMP (https://www.gimp-forum.net/Forum-GIMP)
+--- Forum: Extending the GIMP (https://www.gimp-forum.net/Forum-Extending-the-GIMP)
+---- Forum: Scripting questions (https://www.gimp-forum.net/Forum-Scripting-questions)
+---- Thread: Gimp 3 - python to edit the selection (/Thread-Gimp-3-python-to-edit-the-selection)

Gimp 3 - python to edit the selection - violgamba - 05-29-2025

I'm trying to make a plugin to edit the current image's selection. I'm pretty sure I've worked out how to pull in and edit the selection at this point, but I don't know how to set the modified selection back into the image. I've a few questions:

Is there documentation for how to handle this sort of thing? It was quite the effort to get as far as I've gotten. I am comfortable with C, if that helps.
Can anyone provide an example of how to modify the current image's selection based on a byte array?
Alternately, here is my current logic. If anyone can fill in the blank ("# ???" line) then I'd be very grateful:

Code:

   def do_run(self, procedure, run_mode, image, n_drawables, drawables, config):

        try:

            selection_mask = image.get_selection()

            if selection_mask is None:

                Gimp.message("No selection found. This tool only works with an active selection.")

                return procedure.new_return_values(Gimp.PDBStatusType.EXECUTION_ERROR, None)

            old_buffer = selection_mask.get_buffer()

            rect = old_buffer.get_extent()

            old_buffer_bytes = old_buffer.get(rect, 1.0, "Y float", Gegl.AbyssPolicy.NONE)

            src_pixels = bytearray(old_buffer_bytes)

            dst_pixels = bytearray(old_buffer_bytes)

            for y in range(1, rect.height - 1):

                for x in range(1, rect.width - 1):

                    pixel_index = x + y * rect.width

                    if (src_pixels[pixel_index + 1] != 0 and

                        src_pixels[pixel_index - 1] != 0 and

                        src_pixels[pixel_index + rect.width] != 0 and

                        src_pixels[pixel_index - rect.width] != 0):

                            dst_pixels[pixel_index] = 0

            new_buffer = Gegl.Buffer.new("Y float", rect.x, rect.y, rect.width, rect.height)

            new_buffer.set(rect, "Y float", list(dst_pixels))

            # ???

            return procedure.new_return_values(Gimp.PDBStatusType.SUCCESS, None)

        except Exception as e:

            Gimp.message(f"ERR: {e}")

            return procedure.new_return_values(Gimp.PDBStatusType.EXECUTION_ERROR, None)

RE: Gimp 3 - python to edit the selection - programmer_ceds - 05-29-2025

(05-29-2025, 09:36 AM)violgamba Wrote: I'm trying to make a plugin to edit the current image's selection. I'm pretty sure I've worked out how to pull in and edit the selection at this point, but I don't know how to set the modified selection back into the image. I've a few questions:

Is there documentation for how to handle this sort of thing? It was quite the effort to get as far as I've gotten. I am comfortable with C, if that helps.

Can anyone provide an example of how to modify the current image's selection based on a byte array?

Alternately, here is my current logic. If anyone can fill in the blank ("# ???" line) then I'd be very grateful:

Code:

def do_run(self, procedure, run_mode, image, n_drawables, drawables, config): try: selection_mask = image.get_selection() if selection_mask is None: Gimp.message("No selection found. This tool only works with an active selection.") return procedure.new_return_values(Gimp.PDBStatusType.EXECUTION_ERROR, None) old_buffer = selection_mask.get_buffer() rect = old_buffer.get_extent() old_buffer_bytes = old_buffer.get(rect, 1.0, "Y float", Gegl.AbyssPolicy.NONE) src_pixels = bytearray(old_buffer_bytes) dst_pixels = bytearray(old_buffer_bytes) for y in range(1, rect.height - 1): for x in range(1, rect.width - 1): pixel_index = x + y * rect.width if (src_pixels[pixel_index + 1] != 0 and src_pixels[pixel_index - 1] != 0 and src_pixels[pixel_index + rect.width] != 0 and src_pixels[pixel_index - rect.width] != 0): dst_pixels[pixel_index] = 0 new_buffer = Gegl.Buffer.new("Y float", rect.x, rect.y, rect.width, rect.height) new_buffer.set(rect, "Y float", list(dst_pixels)) # ??? return procedure.new_return_values(Gimp.PDBStatusType.SUCCESS, None) except Exception as e: Gimp.message(f"ERR: {e}") return procedure.new_return_values(Gimp.PDBStatusType.EXECUTION_ERROR, None)

If you type gimp-image-select into the PDB browser you will find a number of procedures for selecting parts of the image e.g. gimp-image-select-rectangle()

RE: Gimp 3 - python to edit the selection - violgamba - 05-31-2025

Quote:If you type gimp-image-select into the PDB browser you will find a number of procedures for selecting parts of the image e.g. gimp-image-select-rectangle()

This is useful in a general sense, but I'm not seeing any option here to set the selection based on a byte array, or anything that a byte array can be transformed into. "gimp-image-select-rectangle" is the nearest match, and it doesn't seem very efficient to turn my byte array into a series of selection rectangles, then call the procedure once for each.

RE: Gimp 3 - python to edit the selection - Ofnuts - 05-31-2025

(05-29-2025, 09:36 AM)violgamba Wrote: I'm trying to make a plugin to edit the current image's selection. I'm pretty sure I've worked out how to pull in and edit the selection at this point, but I don't know how to set the modified selection back into the image. I've a few questions:

Is there documentation for how to handle this sort of thing? It was quite the effort to get as far as I've gotten. I am comfortable with C, if that helps.

Can anyone provide an example of how to modify the current image's selection based on a byte array?

Alternately, here is my current logic. If anyone can fill in the blank ("# ???" line) then I'd be very grateful:

Code:

def do_run(self, procedure, run_mode, image, n_drawables, drawables, config): try: selection_mask = image.get_selection() if selection_mask is None: Gimp.message("No selection found. This tool only works with an active selection.") return procedure.new_return_values(Gimp.PDBStatusType.EXECUTION_ERROR, None) old_buffer = selection_mask.get_buffer() rect = old_buffer.get_extent() old_buffer_bytes = old_buffer.get(rect, 1.0, "Y float", Gegl.AbyssPolicy.NONE) src_pixels = bytearray(old_buffer_bytes) dst_pixels = bytearray(old_buffer_bytes) for y in range(1, rect.height - 1): for x in range(1, rect.width - 1): pixel_index = x + y * rect.width if (src_pixels[pixel_index + 1] != 0 and src_pixels[pixel_index - 1] != 0 and src_pixels[pixel_index + rect.width] != 0 and src_pixels[pixel_index - rect.width] != 0): dst_pixels[pixel_index] = 0 new_buffer = Gegl.Buffer.new("Y float", rect.x, rect.y, rect.width, rect.height) new_buffer.set(rect, "Y float", list(dst_pixels)) # ??? return procedure.new_return_values(Gimp.PDBStatusType.SUCCESS, None) except Exception as e: Gimp.message(f"ERR: {e}") return procedure.new_return_values(Gimp.PDBStatusType.EXECUTION_ERROR, None)

As I understand the doc, you would get a buffer form the drawable (the selection, here) (which is how you get old_buffer) and then write to it. When you close it, the contents are copied to the original drawable. So instead of copying dst_pixels to a new buffer you would just copy them to the original buffer.