Re: Question on re-search-forward and infinite loop

Re: Question on re-search-forward and infinite loop

[Barry Margolin]

In article <mailman.11588.1259269277.2239.help-gnu-emacs@gnu.org>,
 David Maus <maus.david@gmail.com> wrote:

> Hi,
> 
> By accident I stumpled on a way to force Emacs in a infinite loop with
> `re-search-forward'.
> 
> If I create a shiny new buffer, insert something like this:
> 
> ,----
> | (while (re-search-forward "^[ \t]*"))
> | Something else
> `----
> 
>  -- go back beyond the while-clause and evaluate it, Emacs enters an
>     infinite loop. As far as I was able to find out
>     `re-search-forward' with this regexp moves point to the beginning
>     of the next line above the "S" of "Something else" and stays there
>     forever so it never reaches end-of-buffer.
> 
> As I dunno much (or better: anything) on Emacs' interiors my question
> is: Why does this happen? 

Your regexp matches zero or more whitespace characters at the beginning 
of a line.  If there's an empty line at the end of the buffer, it will 
match that.  But since it's a zero-length match, point is left at that 
position, and the next time around it again matches it.

In general, whenever a regexp can match a zero-length string, I think 
that type of loop will go infinitely.

-- 
Barry Margolin, barmar@alum.mit.edu
Arlington, MA
*** PLEASE post questions in newsgroups, not directly to me ***
*** PLEASE don't copy me on replies, I'll read them in the group ***

Re: Question on re-search-forward and infinite loop

[David Maus]

Hi there,

At Thu, 26 Nov 2009 18:17:16 -0500,
Barry Margolin wrote:
> 
> Your regexp matches zero or more whitespace characters at the beginning 
> of a line.  If there's an empty line at the end of the buffer, it will 
> match that.  But since it's a zero-length match, point is left at that 
> position, and the next time around it again matches it.
> 
> In general, whenever a regexp can match a zero-length string, I think 
> that type of loop will go infinitely.

Thanks, this makes perfect sense to me.

Regards,

  -- David
-- 
OpenPGP... 0x316F4BE4670716FD
Jabber.... dmjena@jabber.org
Email..... maus.david@gmail.com
ICQ....... 241051416

Question on re-search-forward and infinite loop

[David Maus]

Hi,

By accident I stumpled on a way to force Emacs in a infinite loop with
`re-search-forward'.

If I create a shiny new buffer, insert something like this:

,----
| (while (re-search-forward "^[ \t]*"))
| Something else
`----

 -- go back beyond the while-clause and evaluate it, Emacs enters an
    infinite loop. As far as I was able to find out
    `re-search-forward' with this regexp moves point to the beginning
    of the next line above the "S" of "Something else" and stays there
    forever so it never reaches end-of-buffer.

As I dunno much (or better: anything) on Emacs' interiors my question
is: Why does this happen? 

Regards,

  -- David

-- 
OpenPGP... 0x316F4BE4670716FD
Jabber.... dmjena@jabber.org
Email..... maus.david@gmail.com
ICQ....... 241051416

Re: Question on re-search-forward and infinite loop

[Lennart Borgman]

On Thu, Nov 26, 2009 at 10:01 PM, David Maus <maus.david@gmail.com> wrote:
> Hi,
>
> By accident I stumpled on a way to force Emacs in a infinite loop with
> `re-search-forward'.
>
> If I create a shiny new buffer, insert something like this:
>
> ,----
> | (while (re-search-forward "^[ \t]*"))
> | Something else
> `----
>
>  -- go back beyond the while-clause and evaluate it, Emacs enters an
>    infinite loop. As far as I was able to find out
>    `re-search-forward' with this regexp moves point to the beginning
>    of the next line above the "S" of "Something else" and stays there
>    forever so it never reaches end-of-buffer.
>
> As I dunno much (or better: anything) on Emacs' interiors my question
> is: Why does this happen?


If the match on next line is 0 chars long then it will not move forward.