On Sat, May 18, 2013 at 5:18 PM, Jeffrey Ratcliffe <jeffrey.ratcliffe at gmail.com> wrote: > GUIs often either use the SANE API directly, or control a scanner by > wrapping scanimage. > > Generally, there is no advantage in using scanimage, as it is slower, > but in certain situations, for instance remote access, it is not > possible to use the SANE API directly. Also, debugging a backend > problem is much easier with scanimage.
I am curious about both of these assertions- can you elaborate the advantages of scanimage here? > The major visual difference for the user in these two interface > methods, is that, apart from groups, scanimage does not expose the > option titles. As the option names are not translated, a GUI therefore > cannot use the sane-backend translations and has to use its own. > > This could all be avoided if scanimage were to expose the option > titles. The translations from sane-backends could be used, and the two > interface methods would look the same. > > I therefore propose an additional option for scanimage that exposes > the option title in the --help output. > > Is a patch for scanimage along these lines likely to be accepted? Yes- we would accept such a patch, as it would be useful for command-line users as well. allan -- "The truth is an offense, but not a sin"
