(sentence-end-without-period): New variable.

(canonically-space-region): Pay attention to
sentence-end-without-period.
(fill-find-break-point): The first argument CHARSET deleted.
Caller changed.
(fill-region-as-paragraph): Pay attention to charset property
nospace-between-words and text property nospace-between-words.

lisp/textmodes/fill.el

@@ -47,6 +47,12 @@ A value of nil means that any change in indentation starts a new paragraph."
   :type 'boolean
   :group 'fill)
 
+(defcustom sentence-end-without-period nil
+  "*Non-nil means a sentence will end without period.
+For example, Thai text ends with double space but without period."
+  :type 'boolean
+  :group 'fill)
+
 (defvar fill-paragraph-function nil
   "Mode-specific function to fill a paragraph, or nil if there is none.
 If the function returns nil, then `fill-paragraph' does its normal work.")
@@ -136,7 +142,8 @@ number equals or exceeds the local fill-column - right-margin difference."
 (defun canonically-space-region (beg end)
   "Remove extra spaces between words in region.
 Leave one space between words, two at end of sentences or after colons
-\(depending on values of `sentence-end-double-space' and `colon-double-space').
+\(depending on values of `sentence-end-double-space', `colon-double-space',
+and `sentence-end-without-period').
 Remove indentation from each line."
   (interactive "r")
   (save-excursion
@@ -154,7 +161,9 @@ Remove indentation from each line."
 	  (save-excursion
 	    (skip-chars-backward " ]})\"'")
 	    (cond ((and sentence-end-double-space
-			(memq (preceding-char) '(?. ?? ?!)))  2)
+			(or (memq (preceding-char) '(?. ?? ?!))
+			    (and sentence-end-without-period
+				 (= (char-syntax (preceding-char)) ?w)))) 2)
 		  ((and colon-double-space
 			(= (preceding-char) ?:))  2)
 		  ((char-equal (preceding-char) ?\n)  0)
@@ -252,6 +261,8 @@ act as a paragraph-separator."
 The predicate is called with no arguments, with point at the place
 to be tested.  If it returns t, fill commands do not break the line there.")
 
+;; Put `fill-find-break-point-function' property to charsets which
+;; require special functions to find line breaking point.
 (let ((alist '((katakana-jisx0201 . kinsoku)
 	       (chinese-gb2312 . kinsoku)
 	       (japanese-jisx0208 . kinsoku)
@@ -263,18 +274,28 @@ to be tested.  If it returns t, fill commands do not break the line there.")
 			  (cdr (car alist)))
     (setq alist (cdr alist))))
 
-(defun fill-find-break-point (charset limit)
-  "Move point to a proper line wrapping position of the current line.
-
-CHARSET is a non-ascii character set before or after the current position.
+(defun fill-find-break-point (limit)
+  "Move point to a proper line breaking position of the current line.
 Don't move back past the buffer position LIMIT.
 
-If CHARSET has the property `fill-find-break-point-function', this
+This function is called when we are going to break the current line
+after or before a non-ascii character.  If the charset of the
+character has the property `fill-find-break-point-function', this
 function calls the property value as a function with one arg LINEBEG.
-If CHARSET has no such property, do nothing."
-  (let ((func (get-charset-property charset 'fill-find-break-point-function)))
-    (if func
-	(funcall func limit))))
+If the charset has no such property, do nothing."
+  (let* ((ch (following-char))
+	 (charset (char-charset ch))
+	 func)
+    (if (eq charset 'ascii)
+	(setq ch (preceding-char)
+	      charset (char-charset ch)))
+    (if (eq charset 'ascii)
+	nil
+      (if (eq charset 'composition)
+	  (setq charset (char-charset (composite-char-component ch 0)))))
+  (setq func (get-charset-property charset 'fill-find-break-point-function))
+  (if (and func (fboundp func))
+      (funcall func limit))))
 
 (defun fill-region-as-paragraph (from to &optional justify
 				      nosqueeze squeeze-after)
@@ -398,19 +419,29 @@ space does not end a sentence, so don't break a line there."
 	  ;; loses on split abbrevs ("Mr.\nSmith")
 	  (while (re-search-forward "[.?!][])}\"']*$" nil t)
 	    (or (eobp) (insert-and-inherit ?\ )))
-	  (goto-char from)
-	  ;; The character category `|' means that we can break a line
-	  ;; at the character.  Since we don't need a space between
-	  ;; them, delete all newlines between them ...
-	  (while (re-search-forward "\\c|\n\\|\n\\c|" nil t)
-	    (if (bolp)
-		(delete-char -1)
-	      (if (= (char-before (match-beginning 0)) ?\ )
-		  ;; ... except when there is end of sentence.  The
-		  ;; variable `sentence-end-double-space' is handled
-		  ;; properly later.
-		  nil
-		(delete-region (match-beginning 0) (1+ (match-beginning 0))))))
+
+	  (goto-char from)	  
+	  (if enable-multibyte-characters
+	      ;; Delete unnecessay newlines surrounded by words.  The
+	      ;; character category `|' means that we can break a line
+	      ;; at the character.  And, charset property
+	      ;; `nospace-between-words' tells how to concatenate
+	      ;; words.  If the value is non-nil, never put spaces
+	      ;; between words, thus delete a newline between them.
+	      ;; If the value is nil, delete a newline only when a
+	      ;; character preceding a newline has text property
+	      ;; `nospace-between-words'.
+	      (while (search-forward "\n" nil t)
+		(let ((prev (char-before (match-beginning 0)))
+		      (next (following-char)))
+		  (if (and (or (aref (char-category-set next) ?|)
+			       (aref (char-category-set prev) ?|))
+			   (or (get-charset-property (char-charset prev)
+						     'nospace-between-words)
+			       (get-text-property (1- (match-beginning 0))
+						  'nospace-between-words)))
+		      (delete-char -1)))))
+
 	  (goto-char from)
 	  (skip-chars-forward " \t")
 	  ;; Then change all newlines to spaces.
@@ -504,11 +535,9 @@ space does not end a sentence, so don't break a line there."
 		      ;; before a non-ascii character, we may have to
 		      ;; run a special function for the charset of the
 		      ;; character to find the correct break point.
-		      (let ((charset (charset-after (1- (point)))))
-			(if (eq charset 'ascii)
-			    (setq charset (charset-after (point))))
-			(if (not (eq charset 'ascii))
-			    (fill-find-break-point charset linebeg)))))
+		      (if (not (and (eq (charset-after (1- (point))) 'ascii)
+				    (eq (charset-after (point)) 'ascii)))
+			  (fill-find-break-point linebeg))))
 
 		;; If the left margin and fill prefix by themselves
 		;; pass the fill-column, keep at least one word.