<html><body><div style="color:#000; background-color:#fff; font-family:times new roman, new york, times, serif;font-size:12pt"><div>I am new to Cairo and have fairly limited knowlege of graphics (Windows GDI,GDI+ years ago). I am now on Ubuntu 12.04 and need to render images in png and jpg from GTK in C/C++. I know how to do this when specifying .png and .jpg file names: cairo_surface_t * imgpng = cairo_image_surface_create_from_png("sample_7.png"); cairo_set_source_surface(cr, imgpng, 0,40); cairo_paint(cr); GError * err = NULL; GdkPixbuf * imgpxb = gdk_pixbuf_new_from_file("parrots.jpg",&err); gdk_cairo_set_source_pixbuf(cr, imgpxb, 0, 150); cairo_paint (cr); My program will receive image data from another computer via standard protocols. There might be hundreds of small to small-medium images coming, so it would more efficient to batch the file content of these into a huge buffer and pass that across (from the other computer) along with the image format, buffer offset and length of each chunk. Of course I could copy a chunk into a tmp file and use the png surface and gdkpixbuf filename api calls above, but it seems such a shame having to put the data back into a file for cairo/gdk to read and convert it to the format needed. Is there a better way? Ken P.S. only png and jpg files will be used </div></div></body></html>