[ESS-bugs] Object completion in R buffer doesn't finish with options(error=recover)

[Martin Maechler]

>>>>> "RMH" == Richard M Heiberger <rmh at temple.edu>
>>>>>     on Fri, 15 Sep 2006 00:06:09 -0400 (EDT) writes:

    >> This will leave .Last.value in an undesired state for the
    >> user, but who really needs .Last.value when using ESS? --
    >> I've never used it I think in the last 10 years.

    RMH> I like .Last.value and do use it.  I would like ESS to
    RMH> allow .Last.value to take the value of the last item
    RMH> that ESS sent over to R/S-Plus.  My reason is that I
    RMH> use ESS with the S-Plus Windows GUI.  The S-Plus
    RMH> Windows GUI is an independent process, not inferior to
    RMH> ESS, and maintains the .Last.value as if the ESS-sent
    RMH> lines had been typed into the Commands Window.  I want
    RMH> an option that will permit that same behavior to
    RMH> manifest in the Sqpe running inside ESS and in R
    RMH> running inside ESS.

Hmm, am I understanding you correctly and you just want
.Last.value  ``as it is in R'' ??

That would be perfect because that's exactly what I would have
proposed: 
Get rid of all  "ESS messing with .Last.value"

?

    RMH> I looked for that in the ESS code and decided it was
    RMH> very deeply hidden.  So if you are planning to modify
    RMH> .Last.value behavior to prevent the current issue of an
    RMH> infinite loop with options(error=recover) from
    RMH> happening, perhaps you will be able at the same time
    RMH> figure out how to make .Last.value an option.