This repository was archived by the owner on Aug 8, 2024. It is now read-only.

Description
What are you trying to achieve?
Currently, sharp/libvips offer options to convert an output file to given colorspace but have no option to "maintain" the colorspace from the original photo. As far as I can tell from reading code/documentation, the challenge is that the code is essentially guessing about the source colorspace -- therefore it's safer to explicitly specify the destination. Additionally, the use case for output as sRGB is probably vastly the most common.
It would be useful to offer an option to maintain (guess) the source colorspace and apply it to the target for those cases where it cannot be specified explicitly and it's desirable to preserve the colorspace.
Libvips has a function, vips_image_guess_interpretation(VipsImage *vips_image) which returns an educated guess about the colorspace of the source image. This can be used during pipeline process during toColo(u)rspace(). The option might be: maintain or guess (guess seems like a more honest description of what's happening behind the scenes.