[ESS] Does R fix() or edit() work for you these days?]

[Martin Maechler]

>>>>> "RMH" == Richard M Heiberger <rmh at temple.edu>
>>>>>     on Mon, 10 Sep 2007 14:15:44 -0400 (EDT) writes:

    RMH> I think Ottorino is looking for page(). 

yes, thank you Rich.

edit() or fix() is about *changing* objects, and as Kasper has
nicely explained,  I'd expect ESS users to rather change the
source (file) of objects than the objects themselves.

In addition to Rich's and Kasper's good points let me mention
the one other situation: You want to modify an object of which
the code is not easily available to you:

There I'd rather use  C-c C-d   [ess-dump-object-into-edit-buffer]
which uses R-mode (and a file !) for the object you edit.


    RMH> page() is similar to edit() in its behavior, since it
    RMH> opens up an emacs buffer containing the text image of
    RMH> the data.frame.  It differs in an important way.  It
    RMH> doesn't freeze the R process (edit requires C-x # to
    RMH> save the object back in the running environment).
    RMH> page() has an additional advantage, it persists for the
    RMH> duration of the emacs session.  When you use edit(),
    RMH> the buffer vanishes as soon as it is saved back.
    RMH> page() satisfies the "source is real" criterion because
    RMH> it opens a buffer that is not associated with a file.
    RMH> When the emacs session if closed, the buffer vanishes.

    RMH> Rich