indeed! Maybe I misunderstood the OP.

You can find more on the API here:

http://orgmode.org/worg/dev/org-element-api.html

There are also a bunch of functions to do other things, e.g.

(org-todo) (org-cut-subtree) (org-entry-put) and many others

that let you change properties of headings, cut them, etc...

These are documented in the org code. The best way to learn how to use them
is to reverse engineer how you would make the change you want manually,
e.g. which keys do you press, then use C-h k on those keys to see what
commands get run.

John

-----------------------------------
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


On Mon, Sep 12, 2016 at 2:21 PM, Thorsten Jolitz <tjol...@gmail.com> wrote:

> Joost Kremers <joostkrem...@fastmail.fm> writes:
>
> Hi,
>
> > I was wondering if there is some sort of (semi)official API for handling
> > org files programmatically. That's to say, is there a documented way for
> > non-org Emacs packages to manipulate (the contents of) org files?
> >
> > Specifically, I'm wondering about creating and deleting entries (by
> > entries I mean headers with their contents), changing TODO state, moving
> > entries, that sort of thing.
>
> by "non-org Emacs packages" you mean Emacs packages written in Elisp,
> but not part of Org-mode?
>
> The org-mode parser converts an Org document into a nested list and
> provides many convenience functions to work on this parse tree. So
> org-element.el (and maybe ox.el too) is the core library for converting
> an Org text document into an Elisp data structure and working with that,
> have a look at these two functions:
>
> ,----
> | 3965:(defun org-element-parse-buffer (&optional granularity visible-only)
> | 4043:(defun org-element-map
> `----
>
> If you feel you don't need the whole parse tree, but rather want to act
> locally on the Org element at point, you might want to look at
> org-dp.el with just two core functions (create and rewire an Org
> element) and a mapping functions (plus quite a few utilities in
> org-dp.el and org-dp-lib.el):
>
> ,----
> | 523:(cl-defun org-dp-create
> |        (elem-type &optional contents insert-p affiliated &rest args)
> | 642:(cl-defun org-dp-rewire
> |        (elem-type &optional contents replace affiliated element &rest
> args)
> | 766:(defun org-dp-map
> |        (fun-with-args rgxp &optional match-pos backward-search-p beg end
> |         silent-p)
> `----
>
> Note that I recently added 4 "Tempo" Templates (a bit like Yasnippets)
> to org-dp that make it easy to insert a 'create' or 'rewire' call with
> just those parameters the interpreter uses for the element to be created
> or rewired:
>
> ,----
> | M-x tempo-template-org-dp-create
> | M-x tempo-template-org-dp-create-with-comments
> | M-x tempo-template-org-dp-rewire
> | M-x tempo-template-org-dp-rewire-lambda
> `----
>
> e.g. calling the third one with elem type "headline" enters this template
>
> #+BEGIN_SRC emacs-lisp
>   (org-dp-rewire 'headline
>   "\n" ;cont
>   t ;ins
>   nil ;aff
>   nil ;elem
>   :level 1 ;1..8
>   :priority nil ;65|66|67
>   :todo-keyword TODO
>   :title ""
>   :tags '( )
>   :commentedp nil
>   :pre-blank 0
>   :footnote-section-p nil
>   )
> #+END_SRC
>
> while for elem type "example block" its only:
>
> #+BEGIN_SRC emacs-lisp
>   (org-dp-rewire 'example-block nil t ;cont ins
>   nil ;aff
>   nil ;elem
>   :switches ""
>   :preserve-indent ""
>   :value ""
>   )
> #+END_SRC
>
> since example-blocks have no content (:value is their content) and only
> 3 parameters that are actually interpreted.
>
> Using this system, creating or rewiring an Org Element from Elisp
> requires only to define the values of the interpreted parameters, all
> the low level stuff (actually creating and inserting the new/modified
> element in text form) is left to the interpreters (from org-element.el).
>
> You just declare what you want and don't worry anymore how it is done
> (=> dp stands for declarative programming, in this context at least ;-)
>
> --
> cheers,
> Thorsten
>
>
>

Reply via email to