sort chinese characters

sort-pyim-hanzi-order.el

(require 'cl-lib)
(defun my-read-file (file)
  (with-temp-buffer
    (insert-file-contents (file-truename file))
    (buffer-string)))

;; (message "content=%s" (my-read-file))
(defun my-format-dictionary ()
  (interactive)
  (let* ((h (make-hash-table :size 1000))
         (text (my-read-file "~/Downloads/dict.txt"))
         (dict (my-read-file "./pyim-basedict.pyim"))
         (newfile (file-truename "./new-base-dict.pyim"))
         key
         val)

    ;; initialized hash table
    (dotimes (i (length text))
      (when (> (setq key (elt text i)) 256)
        ;; only count chinese characters
        (unless (setq val (gethash key h))
          (setq val 0))
        (setq val (1+ val))
        (puthash key val h)))

    (with-temp-buffer
      (dolist (l (split-string dict "\n"))
        (cond
         ((string-match "^\\([a-z]+ \\)\\(.*\\)" l)
          (let* ((pinyin (match-string 1 l))
                 (chars (split-string (match-string 2 l) " "))
                 (newchars (cl-stable-sort chars `(lambda (a b) (> (or (gethash (elt a 0) ,h) 0) (or (gethash (elt b 0) ,h) 0)))))
                 (newl (concat pinyin (mapconcat 'identity newchars " "))))
            (insert (format "%s\n" newl))))
         (t
          (insert (format "%s\n" l)))))
      (write-file newfile))
    (message "%s created." newfile)))