RE: view-lossage hard-codes 100 keystrokes

["Drew Adams" <drew.adams@oracle.com>]

    > It's not easily done, since the keys are recorded in a statically
    > allocated C array whose size is a compile-time constant.

    Actually, it is recorded in a Lisp Vector, so it should be
    trivial to increase the size of that vector.

    We could add the size of the lossage buffer as a second
    arg to clear-this-command-keys which already interfaces
    to the recent keys vector.

    > I won't
    > recommend such a change for before the release, but that's just me.

    I don't see an urgent need, but it would be a trivial change.

No, nothing urgent or even very important. If it could be done, I think it
would be helpful. Even increasing 100 to, say 500 or 1000 would be helpful.

For some reason, I find that on the rare occasions that I need to use this,
what I'm looking for is often further back than 100 keystrokes.