help-gnu-emacs
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

`print' inside Emacs vs `print' to stdout (with Emacsclient)


From: Thorsten Jolitz
Subject: `print' inside Emacs vs `print' to stdout (with Emacsclient)
Date: Wed, 23 Oct 2013 09:49:08 +0200
User-agent: Gnus/5.130002 (Ma Gnus v0.2) Emacs/24.3 (gnu/linux)

Hi List, 

this is the parse-tree of a minimal Org file

#+begin_src org
* A
B
#+end_src

as found in the *Messages* buffer after printing with 

#+begin_src emacs-lisp
(let ((org-cycle-silently t)
      (print-escape-newlines t)
      (print-circle t))
  (with-temp-buffer
    (insert-file-contents-literally
     "/path/to/tmp.org")
    (org-mode)
    (print
     (org-element-parse-buffer))))
#+end_src

#+begin_src emacs-lisp
#2=(org-data nil #1=(headline (:raw-value "A" :begin 1 :end 7 :pre-blank
0 :hiddenp outline :contents-begin 5 :contents-end 7 :level 1 :priority
nil :tags nil :todo-keyword nil :todo-type nil :post-blank
0 :footnote-section-p nil :archivedp nil :commentedp nil :quotedp
nil :CATEGORY nil :title (#("A" 0 1 (:parent #1#))) :parent #2#) #3=(section
(:begin 5 :end 7 :contents-begin 5 :contents-end 7 :post-blank
0 :parent #1#) #4=(paragraph (:begin 5 :end 7 :contents-begin 5 :contents-end
7 :post-blank 0 :post-affiliated 5 :parent #3#) #("B\n" 0 2 (:parent #4#))))))
#+end_src

Now when I use the exactly same print commands as arguments for an emacsclient
call

#+begin_src sh
  $ emacsclient -s my-server -e "(let ((org-cycle-silently
  t)(print-escape-newlines t)(print-circle t)) (with-temp-buffer
  (insert-file-contents-literally
  \"/path/to/tmp.org\") (org-mode) (print
  (org-element-parse-buffer))))"
#+end_src  

I see this output on the console, i.e. (print-circle t) does not seem in
effect anymore. 
 
#+begin_src emacs-lisp
  (org-data nil (headline (:raw-value "A" :begin 1 :end
  7 :pre-blank 0 :hiddenp outline :contents-begin 5 :contents-end
  7 :level 1 :priority nil :tags nil :todo-keyword nil :todo-type
  nil :post-blank 0 :footnote-section-p nil :archivedp
  nil :commentedp nil :quotedp nil :CATEGORY nil :title (#("A" 0
  1 (:parent #1))) :parent #0) (section (:begin 5 :end
  7 :contents-begin 5 :contents-end 7 :post-blank 0 :parent
  #1) (paragraph (:begin 5 :end 7 :contents-begin 5 :contents-end
  7 :post-blank 0 :post-affiliated 5 :parent #2) #("B\n" 0
  2 (:parent #3))))))
#+end_src

Actually, what I really see is this

#+begin_src sh
(org-data nil
          (headline
           (:raw-value "A" :begin 1 :end 7 :pre-blank 0 :hiddenp outline 
:contents-begin 5 :contents-end 7 :level 1 :priority nil :tags nil 
:todo-keyword nil :todo-type nil :post-blank 0 :footnote-section-p nil 
:archivedp nil :commentedp nil :quotedp nil :CATEGORY nil :title
                       (#("A" 0 1
                          (:parent #1)))
                       :parent #0)
           (section
            (:begin 5 :end 7 :contents-begin 5 :contents-end 7 :post-blank 0 
:parent #1)
            (paragraph
             (:begin 5 :end 7 :contents-begin 5 :contents-end 7 :post-blank 0 
:post-affiliated 5 :parent #2)
             #("B\n" 0 2
               (:parent #3))))))
#+end_src

and it seems, when this output is received as input by another program, that
not only the 'circular list' print-syntax is lost, but new redundant
line-breaks are introduced that are not related to the original Org file. 

But I need those 'circular' labels and references, and would rather not have
those random linebreaks.

The only way to get the 'circular' labels seems converting the whole structure
to a string while or before printing (with `(print (format ...))' or
`prin1-to-string' e.g.), but that causes other problems w.r.t. escaping
special chars and (unwanted) surrounding double quotes.

What I would really need is being able to just `print' a structure to
stdout in an emacsclient call and receive as input of another program on
stdin exactly what would be printed in the Emacs *Messages* buffer when
evaluating the print commands in an emacs-lisp-mode buffer.

Is there any way to achieve this?

-- 
cheers,
Thorsten




reply via email to

[Prev in Thread] Current Thread [Next in Thread]