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

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

Re: Splitting String


From: Pascal Bourguignon
Subject: Re: Splitting String
Date: Tue, 15 Nov 2005 14:09:56 +0100
User-agent: Gnus/5.11 (Gnus v5.11) Emacs/22.0.50 (gnu/linux)

Jeckob@gmx.net writes:

> Hi all,
> having problems with some Code i have written, half of the job is done
> but now i have problems to get on. here is me code until now :
>
> (defun format/align-and-newline ()
>   (interactive)
>   (setq anker (point))
>   (let* ((start (progn (beginning-of-line) (point)))
>        (end   (progn (end-of-line) (point)))
>        (line (buffer-substring start end))
>        (words (split-string line)))
>     (setq testpoint (first words))
>     (if (setq posDoppel (position ?: line) )
>       (if (position ?. testpoint)
>           (insert "\n")
>          (delete-region start end)
>         (insert (format "%-10s" (first words)))
>         (dolist (word (cdr words))
>           (insert (format "%-10s" word)))
>         (if (beginning-of-line)
>             (insert "\n")
>           (setq anker (point)))
>         )
>       )
>     )
>   (goto-char anker)
>   (insert "\n")
>   )
>  (local-set-key (kbd "RET") (function format/align-and-newline))
> ;;end
>
> If i type a line like this:
> abc def ghj klm : nop : qrs : tuv
>
> the code transforms the string to:
> abc       def       ghj       klm       :         nop       :
> qrs       :         tuv
>
> So, the thing im having problems is to get the string like this :
> abc       def       ghj       klm       : nop     : qrs     : tuv
>
> in other words, if there is a colon typed a space before a word, i need
> to have the colon with the word to be in the cell of the list, and not
> the colon alone as a word. So, in the example the list should have 7
> elements and not 10. Maybe somebody can give me a hint... thanks in
> advance 

You just write what you mean!!!


For example, if I mean that: if one word is a single colon, then it
put with a space just before the following word, instead of being put
alone, I'd write:

(defun pre-process-words (words)
  (loop
     with result = '()
     while words
     do (if (string= ":" (first words))
            (progn
              (push (format "%s %s" (first words) (second words)) result)
              (setf words (cddr words)))
            (progn
              (push (first words) result)
              (setf words (cdr words))))
     finally (return (nreverse result))))

(defun format/align-and-newline ()
  (interactive)
  (let* ((anker (point))
         (start (progn (beginning-of-line) (point)))
         (end   (progn (end-of-line) (point)))
         (line  (buffer-substring start end))
         (words (pre-process-words (split-string line))))
    (setq testpoint (first words))
    (if (setq posDoppel (position ?: line) )
        (if (position ?. testpoint)
            (insert "\n")
            (delete-region start end)
            (insert (format "%-10s" (first words)))
            (dolist (word (cdr words))
              (insert (format "%-10s" word)))
            (if (beginning-of-line)
                (insert "\n")
                (setq anker (point)))))
    (goto-char anker)
    (insert "\n")))


But both the following lines:
----------------------------------------
abc def ghj klm  :  nop  :  qrs :  tuv
abc def ghj klm   : nop   : qrs  : tuv
----------------------------------------

will be formated as:
----------------------------------------

abc       def       ghj       klm       : nop     : qrs     : tuv     

abc       def       ghj       klm       : nop     : qrs     : tuv     
----------------------------------------



Since you mean this:

> in other words, if there is a colon typed a space before a word, i need
> to have the colon with the word to be in the cell of the list, and not
> the colon alone as a word. So, in the example the list should have 7
> elements and not 10. Maybe somebody can give me a hint... thanks in
> advance 

then you should parse the colon, SPC, word before splitting the words:

Instead of:    (words (pre-process-words (split-string line)))
you'd write:   (words (parse-line/for-align-and-newline line))

and, for example:

(defun parse-line/for-align-and-newline (line)
  ...
  (cond
     ...
     ((string-match "^\\(: [^ ]+\\) *" line start) ; a colon, SPC and a "word" 
                                      ; made of one or more non-spaces
        (setf word (match-string 1 line)
              start (match-end 0)))
     ...)
  ...)


-- 
__Pascal Bourguignon__                     http://www.informatimago.com/
Litter box not here.
You must have moved it again.
I'll poop in the sink. 


reply via email to

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