Re: Invalid read syntax (#) in org-element parse tree

[Thorsten Jolitz <tjolitz@gmail.com>]

Nicolas Goaziou <n.goaziou@gmail.com> writes:

>> I want to write an 'unusual' backend that does not need anything else
>> from the exporting framework but the parse-tree as a list. 
>
> Then you don't want the exporting framework at all, only org-element.el.

yes, only the parser. 

> Anyway I'm confused. The parse-tree _is_ a list. To convince yourself,
> evaluate the following in any Org buffer:
>
>   (listp (org-element-parse-buffer))
>
>> So all I need would be a workaround for this read-error issue, i.e.
>> a tip how to get a version of the parse tree that can be used as list
>> in a Lisp program.
>
> Have you tried (setq print-circle t) ?
>
>> I could not find any explanation for the '#1' and '#2' syntax I
>> encountered, so I don't really know what its all about.
>
> It is explained in the info link I gave you.

I read this link, it says:

,--------------------------------------------------------------------
| To represent shared or circular structures within a complex of Lisp
| objects, you can use the reader constructs ‘#n=’ and ‘#n#’.
`--------------------------------------------------------------------

what is not quite the same like 

,-----------
| :parent #1
`-----------

but with your other hints, I now understand the problem. I wanted to see
how the parse tree looks like, so I printed it out (I did not know about
the existance of 'print-circle' then, but it was set to nil). Then I
tried to experiment with the printed representation, but the #1 syntax
gave me an error. 

When I set 'print-circle' to t, the printed result looks like described
in the info page, with elements like 

[...] :parent #66#)))) #67=(headline [...]


-- 
cheers,
Thorsten