Re: Hard to switch from vi

[Jeremy Hankins <nowan@nowan.org>]

[-- Attachment #1: Type: text/plain, Size: 2141 bytes --]

Lennart Borgman <lennart.borgman.073@student.lu.se> writes:
> Jeremy Hankins wrote:

>> I personally consider viper to be more of an implementation of vi in
>> elisp; it generally works quite well with other emacs modes without
>> interfering with their use.  There are a few exceptions, though -- I'd
>> love to get viper to work better with folding.el, for example.
>
> What version of Emacs do you use? What are the problems with folding.el?

It's the debian unstable emacs-snapshot package 20061003-1, so 22.  I'm
using folding.el v3.35, which is current, I believe.  I've hacked
folding.el a bit -- mainly to set up an option to not narrow when
folding-shift-in is called.  I've attached a patch against folding.el,
if you're interested, and I'm using this advice for viper-search:

;; This advises viper-search so that it behaves well with folding-mode.
(defadvice viper-search (around jjh/folding-viper-search activate)
  (let ((executing-kbd-macro t))
    ad-do-it)
  (let* ((start (match-beginning 0))
         (end (match-end 0))
         (match (buffer-substring start end)))
    (when folding-mode
      (folding-shift-in))
    (search-forward match end t nil)
    (goto-char start))
  (or executing-kbd-macro
      (memq viper-intermediate-command
            '(viper-repeat viper-digit-argument viper-command-argument))
      (viper-flash-search-pattern))
  ad-return-value)

It's a bit ugly, because it sets executing-kbd-macro so that it can
override the way viper-flash-search-pattern is called.  Unfortunately
this disables the message about the search wrapping the buffer, but
without it folding marks are flashed instead of the search match.  If
anyone has a better suggestion I'd be interested -- I'm pretty new to
elisp.  Ideal would be a hook in viper-search, I guess.

What I'd really like, though, is for ex-style line addressing to work
properly, but I don't really see how that's possible without significant
modifications to viper.  From what I can see it looks like viper handles
generating addresses for lines in a variety of places, and they'd all
have to be tracked down to make it behave with folding-mode.


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: Patch against folding.el --]
[-- Type: text/x-patch, Size: 2699 bytes --]

--- folding.el-orig	2006-09-29 16:58:26.000000000 -0400
+++ folding.el	2006-10-04 08:55:59.000000000 -0400
@@ -2004,7 +2004,9 @@
 				     (folding-point-folded-p goal))
 				 (folding-shift-in t))
 			     (goto-char goal)))
-		    (folding-narrow-to-region (point-min) (point-max) t)))
+		    (folding-narrow-to-region
+		     (and folding-narrow-by-default (point-min))
+		     (point-max) t)))
 	      (if (or folding-stack (folding-point-folded-p (point)))
 		  (folding-open-buffer))))))))
 
@@ -2114,6 +2116,10 @@
 
 ;;; ........................................................ &v-Config ...
 
+(defvar folding-narrow-by-default t
+  "If t (default) things like isearch will enter folds.  If nil the
+folds will be opened, but not entered.")
+
 ;; Q: should this inherit mouse-yank-at-point's value? maybe not.
 (defvar folding-mouse-yank-at-point t
   "If non-nil, mouse activities are done at point instead of 'mouse cursor'.
@@ -3604,22 +3610,26 @@
 many folds as necessary are entered to make the surrounding text
 visible. This is useful after some commands eg., search commands."
   (interactive)
-  (let ((goal (point)))
-    (if (folding-skip-ellipsis-backward)
-	(while (prog2 (beginning-of-line)
-		      (folding-shift-in t)
-		      (goto-char goal)))
-      (let ((data (folding-show-current-entry noerror t)))
-	(and data
-	     (progn
-	       (setq folding-stack
-		     (if folding-stack
-			 (cons (cons (point-min-marker) (point-max-marker))
-			       folding-stack)
-		       '(folded)))
-	       (folding-set-mode-line)
-	       (folding-narrow-to-region (car data) (nth 1 data))
-	       (nth 2 data)))))))
+  (labels
+      ((open-fold nil
+		  (let ((data (folding-show-current-entry noerror t)))
+		    (and data
+			 (progn
+			   (when folding-narrow-by-default
+			     (setq folding-stack
+				   (if folding-stack
+				       (cons (cons (point-min-marker) (point-max-marker))
+					     folding-stack)
+				     '(folded)))
+			     (folding-set-mode-line))
+			   (folding-narrow-to-region (car data) (nth 1 data)))))))
+    (let ((goal (point)))
+      (while (folding-skip-ellipsis-backward)
+	(beginning-of-line)
+	(open-fold)
+	(goto-char goal))
+      (when (not folding-narrow-by-default)
+	(widen)))))
 
 ;;}}}
 ;;{{{ folding-shift-out
@@ -3764,9 +3774,11 @@
        (re-search-forward "[\n\C-m]" nil 0 (1- line)))
   (let ((goal (point)))
     (while (prog2 (beginning-of-line)
-		  (folding-shift-in t)
-		  (goto-char goal))))
-  (folding-narrow-to-region (point-min) (point-max) t))
+	       (folding-shift-in t)
+	     (goto-char goal))))
+  (folding-narrow-to-region
+   (and folding-narrow-by-default (point-min))
+   (point-max) t))
 
 ;;}}}
 

[-- Attachment #3: Type: text/plain, Size: 106 bytes --]


-- 
Jeremy Hankins <nowan@nowan.org>
PGP fingerprint: 748F 4D16 538E 75D6 8333  9E10 D212 B5ED 37D0 0A03

[-- Attachment #4: Type: text/plain, Size: 152 bytes --]

_______________________________________________
help-gnu-emacs mailing list
help-gnu-emacs@gnu.org
http://lists.gnu.org/mailman/listinfo/help-gnu-emacs