What's the easiest way to go from pixbuf to Pixmap?

[foranuj]

Would something like this be the answer?

def pixbuf_to_pixmap(pixbuf):
    width = pixbuf.get_width()
    height = pixbuf.get_height()
    pixels = pixbuf.get_pixels()

    return fitz.Pixmap(fitz.csRGB, width, height, bytearray(pixels), False)

[JorjMcKie]

The overloaded constructor you are referring to is correct.
You can use any object type of bytes, bytearray of io.BytesIO. In so far, a prior conversion to bytearray is not necessary if pixbuf.get_pixels() already is bytes.
This area must however be structured like a Pixmap.samples.

[foranuj]

You think this or a cleaner version of this code should be part of the library to allow easy conversion between pix buf and pixmap?
Currently there doesn't seem to be a function available to use.

[JorjMcKie]

what is pixbuf actually?

[JorjMcKie]

If just being given a bytes-like memory area, it could be the samples of 3 or 4 different pixmaps. You need all the other parameters to interpret it correctly. So there for me there seems to be no easier / more elegant constructor, if we don't want to talk about sheer mannerisms.
But I may be wrong.

[foranuj]

pixbuf -> GdkPixbuf