Re: [bug] [babel] babel corrupts undo history

[Samuel Wales <samologist@gmail.com>]

hi bastien,

thanks for revisiting this.

i cannot do thorough bug reports at this time, but there are a few
undo-related issues in org-mode now.  here are a few from memory
[there are others]:

  - occasional buffer or undo-tree corruption from c-c ' [see previous
posts in this thread]
  - disruptive and surprising combining of undo entries when doing
agenda operations where you can't undo properly
  - buffer corruption without c-c '

typical example of last one:

===
* bastien
#+begin_src org
  bastien
  ^bastien
#+end_src
* bastien
===

put point at ^, fill-paragraph, undo.  with undo-tree, at least, the
buffer will be corrupted.

===

often there is no way to fix the buffer corruption.  undo is after all
a way to fix things.  so i'm particularly sensitive to it not working.

if org did not touch undo internal features, then i think undo would
be more trustworthy.  this is what i prefer as a user.

i think not touching undo internals is the safer, more defensive strategy.

===

one thing org sometimes does is try to set buffer-undo-list.  it's
really for speed imo.  i can't think of any reason why org really
needs it.  perhaps i am mistaken and there is a really good reason for
such things, but i suspect it has caused a lot of bugs.

in the case of c-c ', i would prefer having the indentation adding
show up as an undo entry [or whatever would happen if we ripped out
the undo-related setting].  separating the undo of the base and
indirect buffers leads to surprises.  i expect undo to just work in
both places and not to undo things that i did not edit last.

even when there is not a bug per se, when you edit a source block,
there is a gap in the undo record.  like nixon's tape gap during
watergate, it raises questions.  :/

===

to me, undo is a low-level feature that should never be buggy or
surprising.  if it is, then anything that causes those should be
ripped out, even if it means losing a fancy undo-related feature.

for me, the best undo feature of all is reliability.

:]

samuel


On 3/18/14, Bastien <bzg@gnu.org> wrote:
> Hi Samuel,
>
> Samuel Wales <samologist@gmail.com> writes:
>
>> as far as i know, my assessment below is correct, but i cannot
>> confirm.
>
> I revisited this thread.
>
>> i believe that if undo-related code is ripped out of babel, then undo
>> will work correctly in the source buffer and in the edit buffer.
>> i am not clear on what the purpose of changing undo behavior is?
>
> Time flies: maybe it was clear before, but for now undo works fine
> for me, in the source editing buffer and in the original buffer too.
>
> When I do C-c ' from the *Org Src* buffer, then C-/, the buffer is
> restored to its previous state, that is: before the insertion of the
> edited source.
>
> Can you restate what the bug is?
>
> Thanks,
>
> --
>  Bastien
>


-- 
The Kafka Pandemic: http://thekafkapandemic.blogspot.com

The disease DOES progress.  MANY people have died from it.  And
ANYBODY can get it.

Denmark: free Karina Hansen NOW.