When reading from a file, we just shove all metadata in the ImageSpec. There are a few cases where the items are extremely large or expensive to read (such as large thumbnails, ICC profiles, or ID manifests), and the app may not need it for the particular use at hand.
This issue touches on whether we should care, and if so, what approaches we might take to allow a more performant image input when the app knows it doesn't need that data. I don't know the right answer, but I threw out some possible approaches and would love feedback from anybody who has thoughts on it. https://github.com/OpenImageIO/oiio/issues/2902 Feel free to just add comments to that ticket to discuss it. -- Larry Gritz l...@larrygritz.com _______________________________________________ Oiio-dev mailing list Oiio-dev@lists.openimageio.org http://lists.openimageio.org/listinfo.cgi/oiio-dev-openimageio.org
