[O] transcluding some org-elements in multiple places

[John Kitchin]

Hi all,

I have an idea for how I could transclude "copies" or links to
org-elements in multiple places and keep them up to date. A prototypical
example of this is I have a set of org-contacts in one place, and I want
to create a new list of people for a committee in a new place made of
"copies" of the contact headlines. But I do not really want to duplicate
the headlines, and if I modify one, I want it reflected in the other
places. I do not want just links to those contacts, because then I can
not do things with org-map-entries, and other org-machinery which needs
the actual headlines/properties present. Another example might be I want
a table in two places, but the contents of them should stay
synchronized, ditto for a code block.

This idea was inspired by https://github.com/gregdetre/emacs-freex.

The idea starts with creating (wait for it...) a new link ;) In a
document where I want to transclude a headline, I would enter something like:

[[transclude:some-file.org::*headline title]]

Then, I would rely on the font-lock system to replace that link with the
headline and its contents (via the :activate-func link property), and to
put an overlay on it with a bunch of useful properties, including
modification hooks that would update the source if I change the the
element in this document, and some visual indication that it is
transcluded (e.g. light gray background/tooltip).

I would create a kill-buffer hook function that would replace that
transcluded content with the original link. A focus-in hook function
would make sure the transcluded content is updated when you enter the
frame. So when the file is not open, there is just a transclude link
indicating what should be put there, and when it is open, the overlay
modification hooks and focus hook should ensure everything stays
synchronized (as long as external processes are not modifying the
contents).

It seems like this could work well for headlines, and named tables, src
blocks, and probably any other element that can be addressed by a
name/ID.

Any thoughts? Do you see any potential issues?




-- 
Professor John Kitchin
Doherty Hall A207F
Department of Chemical Engineering
Carnegie Mellon University
Pittsburgh, PA 15213
412-268-7803
@johnkitchin
http://kitchingroup.cheme.cmu.edu