occur patch

occur patch

[Nathaniel Flath]

Hello,
I found it useful to have occur default to the current word instead of
the last regexp, so I modified replace.el to support this.  The patch
is below, if it should be installed into emacs itself - let me know of
any issues.

Thanks,
Nathaniel Flath

diff --git a/lisp/replace.el b/lisp/replace.el
index baea282..12c95cf 100644
--- a/lisp/replace.el
+++ b/lisp/replace.el
@@ -1006,9 +1006,17 @@ which means to discard all text properties."
   :group 'matching
   :version "22.1")

+(defcustom occur-read-method 'history
+  "Method that Occur uses to prompt for a regex."
+  :type '(choice (const :tag "Last regex" 'history)
+		 (const :tag "Current word" 'current-word))
+  :group 'matching)
+
 (defun occur-read-primary-args ()
   (list (read-regexp "List lines matching regexp"
-		     (car regexp-history))
+		     (if (eq occur-read-method 'history)
+			 (car regexp-history)
+		       (current-word)))
 	(when current-prefix-arg
 	  (prefix-numeric-value current-prefix-arg))))

Re: occur patch

[Geoff Gole]

I agree that this is useful, I have something like this in my .emacs.

A minor problem with the patch is that the result of current-word is
used unquoted. Depending on whether people expect more restrictive or
more permissive searching, it may also be desirable to add
beginning/end of symbol:

    (let ((word (current-word)))
      (when word
	(concat "\\_<" (regexp-quote word) "\\_>")))

Re: occur patch

[Stefan Monnier]

> I found it useful to have occur default to the current word instead of
> the last regexp, so I modified replace.el to support this.  The patch
> is below, if it should be installed into emacs itself - let me know of
> any issues.

I think a better solution than a config var is to provide both values.
So a single M-n chooses one and a second M-n chooses the other.


        Stefan

Re: occur patch

[Juri Linkov]

>> I found it useful to have occur default to the current word instead of
>> the last regexp, so I modified replace.el to support this.  The patch
>> is below, if it should be installed into emacs itself - let me know of
>> any issues.
>
> I think a better solution than a config var is to provide both values.
> So a single M-n chooses one and a second M-n chooses the other.

And this is what occur already does now.
Maybe the order of these M-n values should be configurable?