I don't understand. Is the problem that edits made in the indirect
buffer are not spliced into the undo history of the Org-mode buffer?
Samuel Wales <samolog...@gmail.com> writes:
> c-c ' c-c ' on this. then undo or undo-tree-undo.
>
> what happens is data corruption. to me, undo is a low-level
> operation that should always work, even with a syntactically
> invalid block.
>
> a relevant variable is org-src-preserve-indentation. it is
> possible to change it without changing a block.
>
> ===
>
> #+BEGIN_SRC sh :noweb yes :results verbatim output
> (
> cat <<EOF
> test
> EOF
> ) 2>&1
> :
> #+END_SRC
>
> ===
>
> Emacs 24.2, recent Org, recent undo-tree-mode.
>
> Thanks.
>
> Samuel
>
> P.S. You might be wondering why I use a subshell and a
> redirection followed by a null command. This is because I
> prefer to bypass Babel's error mechanism, which I find
> confusing. Doing exactly as above is the only way I know to
> get Babel to work the way I prefer (although {} might or
> might not work also). I always do this for every block. IMO it would
> be great for newcomers if there were options in Babel sh blocks for
> exactly this.
--
Eric Schulte
https://cs.unm.edu/~eschulte
PGP: 0x614CA05D
