How to properly parse a buffer into a list ?

How to properly parse a buffer into a list ?

[Philippe M. Coatmeur]

Hi ; I have this perl-script that prints email elements separated by
the string : "|_|".

When I start-process-shell-command it I get a buffer containing my
elements, and I then parse this buffer using this function :

(defun buffer-to-list (buf)
  "Make & return a list (of lists) LINES from lines in a buffer BUF"
  (with-current-buffer buf
    (save-excursion
      (goto-char (point-min))
      (let ((lines '()))
        (while (not (eobp))
          (push
	   (split-string
	    (buffer-substring (point) (point-at-eol)) "\|_\|")
	   lines)
          (beginning-of-line 2))
	lines))))

Problem is, this function moves in terms of lines (it puts evey single
line in a list cell), but my elements can easily span over several
lines, and I'd still wouldlike one such element to be a single list
cell..? I'm trying to simplify it to ignore the notion of line, and
just search for the separator string, like this :

  (with-current-buffer buf
    (goto-char (point-min))
    (let ((cells '()))
      (while (not (eobp))
	(push
	 (buffer-substring (point) (re-search-forward "|_|"))
	 cells))
      cells))

But it always returns "string not found" (when I explicitly run
(re-search-forward "|_|") in said buffer it founds it) so apparently
I'm not positioning the point where it should be, so my question
really is : What is the mecanics of search-looping through elements
between a separator string ?

Philippe

Re: How to properly parse a buffer into a list ?

[Pascal J. Bourguignon]

Philippe M. Coatmeur <philippe.coatmeur@gmail.com> writes:

> Hi ; I have this perl-script that prints email elements separated by
> the string : "|_|".

If you could modify it so that it prints a sexp, it would be easier to
read it in emacs.

    (("abc" "@" "example" "com")
     ("def" "@" "example" "com"))

you could then just use:

    (with-current-buffer buf
       (goto-char (point-min))
       (read buf))


> Problem is, this function moves in terms of lines (it puts evey single
> line in a list cell), but my elements can easily span over several
> lines, and I'd still wouldlike one such element to be a single list
> cell..? 

How do you separate email addresses then?


-- 
__Pascal Bourguignon__                     http://www.informatimago.com/
A bad day in () is better than a good day in {}.

Re: How to properly parse a buffer into a list ?

[Ralf Fassel]

* Philippe M. Coatmeur <philippe.coatmeur@gmail.com>
| But it always returns "string not found" (when I explicitly run
| (re-search-forward "|_|") in said buffer it founds it)

I'd guess the try after the last match returns the error.
Try using the 'noerror' flag of 're-search-forward' (or plain
'search-forward', since you're not using regexps anyway-).

R'

Re: How to properly parse a buffer into a list ?

[Philippe M. Coatmeur]

At Mon, 11 Jun 2012 15:17:54 +0200,
Pascal J. Bourguignon wrote:
> 
> Philippe M. Coatmeur <philippe.coatmeur@gmail.com> writes:
> 
> > Hi ; I have this perl-script that prints email elements separated by
> > the string : "|_|".
> 
> If you could modify it so that it prints a sexp, it would be easier to
> read it in emacs.
> 
>     (("abc" "@" "example" "com")
>      ("def" "@" "example" "com"))
> 
> you could then just use:
> 
>     (with-current-buffer buf
>        (goto-char (point-min))
>        (read buf))

Simple, brilliant, bravo Pascal :) (Damnit, if it was'nt for the
second "i" in "brilliant", this sentence would be readable in both
english and french :( )

> 
> 
> > Problem is, this function moves in terms of lines (it puts evey single
> > line in a list cell), but my elements can easily span over several
> > lines, and I'd still wouldlike one such element to be a single list
> > cell..? 
> 
> How do you separate email addresses then?

Well I overlooked this problem, but your solution solved it even
before we were introduced ;)

> 
> 
> -- 
> __Pascal Bourguignon__                     http://www.informatimago.com/
> A bad day in () is better than a good day in {}.

Re: How to properly parse a buffer into a list ?

[Philippe M. Coatmeur]

At Mon, 11 Jun 2012 15:17:54 +0200,
Pascal J. Bourguignon wrote:
> 
> Philippe M. Coatmeur <philippe.coatmeur@gmail.com> writes:
> 
> > Hi ; I have this perl-script that prints email elements separated by
> > the string : "|_|".
> 
> If you could modify it so that it prints a sexp, it would be easier to
> read it in emacs.
> 
>     (("abc" "@" "example" "com")
>      ("def" "@" "example" "com"))
> 
> you could then just use:
> 
>     (with-current-buffer buf
>        (goto-char (point-min))
>        (read buf))
> 

Simple & brilliant, thx Pascal ; Way to go from 13 to 5 lines ;)

> 
> > Problem is, this function moves in terms of lines (it puts evey single
> > line in a list cell), but my elements can easily span over several
> > lines, and I'd still wouldlike one such element to be a single list
> > cell..? 
> 
> How do you separate email addresses then?

I overlokeed this problem, but your solution solved it before we were
even introduced ;)

> 
> 
> -- 
> __Pascal Bourguignon__                     http://www.informatimago.com/
> A bad day in () is better than a good day in {}.