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

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

Re: How to remove verbosity from the data passing mechanism using alist


From: Fren Zeee
Subject: Re: How to remove verbosity from the data passing mechanism using alist or plist ?
Date: Mon, 6 Dec 2010 11:45:17 -0800

On Mon, Dec 6, 2010 at 6:13 AM, Thien-Thi Nguyen <ttn@gnuvola.org> wrote:
> [cc changed to help-gnu-emacs -- this does not belong on emacs-devel]
>
> () Fren Zeee <frenzeee@gmail.com>
> () Sun, 5 Dec 2010 12:15:08 -0800
>
>   [code]
>
> Here is ‘find-my-marker-GOLD’, munged a bit:
>
> (defun find-my-marker-GOLD ()
>  "Starting from anywhere in the file, find my marker GOLD its value
>   and location."
>  (interactive)
>  (save-excursion
>    (save-match-data
>      ;; Starting from the end of the accessible region, find GOLD.
>      (goto-char (point-max))
>      ;; a/ This value of this expression is discarded.
>      ;;    It turns out to be the same as "location" below,
>      ;;    so if you save it, you can avoid a call to ‘point’.
>      ;; b/ The optional args default to ‘nil’, and can be dropped.
>      (search-backward-regexp "GOLD=\\([0-9]+\\)\n" nil nil nil)
>
>      ;; In the following, i have deleted the eol comments, which
>      ;; obscure more than enlighten (on my small computer screen).
>
>      ;; You might consider using `((k0 . ,v0)
>      ;;                            (k1 . ,v1))
>      ;; for succinctness (note backquote and comma placement).
>      ;; An intermediate solution is to use ‘acons’.
>      (list
>       (list
>        :GOLD-value
>        ;; The ‘replace-regexp-in-string’ is not necessary.
>        ;; The initial regexp match (above) already sets the match
>        ;; data; you can use ‘(match-string 1)’ to retrieve it.
>        (string-to-number
>         (replace-regexp-in-string "GOLD=\\([0-9]+\\)\n" "\\1"  (match-string 
> 0))))
>       (list
>        :GOLD-location
>        (point))))))
>
>   [Q] Are there any defects in this method of passing struct and what
>   improvements are possible ?
>
> All defects are misalignments of intention and implementation.
> If you don't know your intention clearly, it's easy for a defect
> to creep in.  Improvements, likewise, depend on intention and pov.
>
>       Specifically, are there ways to reduce verbosity without using cl
>   or staying purely in elisp ?
>
> You can use shorter variable names.  You can make your program
> less piecewise-constructive and more table-oriented.
>
>   [Q] Is there a way to avoid lengthy calling statement like
>         (car (assoc-default :GOLD-value    GOLD )
>       inside let,
>
>       since the first argument of let is an alist of the form
>            ((sym1 val1) (sym2 val2))
>

OK lets take the suggestion below and show me how it works with my case ?

> You need the ‘car’ because you do ‘(list k v)’.
> If you use ‘(cons k v)’, then you do not need the ‘car’.

What I have is

(list (list k1 v1) (list k2 v2))

Now, kindly show me where I put the car to get both of these uniformly
communicated outside of the function to another function ? Later in
some cases I may have (k3 v3) pair !!!

>
>   [Q] Is there a way to using plists for return from find-my-marker-GOLD
>       and utilize in the user function test-GOLD
>
> Yes, but probably you will want to avoid plists.
>



reply via email to

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