On Fri, May 16, 2008 at 2:13 PM, m. allan noah <kitno455 at gmail.com> wrote: > On Fri, May 16, 2008 at 12:21 PM, m. allan noah <kitno455 at gmail.com> wrote: > >> 5. One new SANE_TYPE value: HW_BUTTON > > actually, as i am trying to implement this- i notice that the > constraint is not really enough info to distinguish between types, as > the RANGE and WORD_LIST are applicable to INT and FIXED types. we > really need to keep the existing SANE_TYPE value on buttons. so, i am > going to suggest a change to use a capability bit instead: > SANE_CAP_HW_BUTTON > > It becomes hard to interpret all the possible combinations of > SANE_CAP, but we may be able to provide some guidance in the standard.
And, after yet more digging, it appears that SANE_CAP_HARD_SELECT is already providing just what we need, with the fujitsu and plustek backends using it for buttons, and the hp3900 backend using it for some read-only values that would be better handled with just SANE_CAP_SOFT_SELECT. No other backends use it. We can clarify the textual description in the standard to indicate that front-ends should poll for changes in value. allan -- "The truth is an offense, but not a sin"
