Re: [ jEdit-devel ] catching out of memory

[Thomas M. <tho...@we...>]

Am Dienstag, den 24.07.2012, 13:22 -0600 schrieb Dale Anson:

Hello,

> Thomas, do you have a test case to reproduce the OOM Error in search
>  and replace? It seems to me that it should be pretty easy to tell if
> the error is recoverable or not. There might not be anything useful to
> do to recover the memory, but avoiding an application hang and being
> able to alert the user to restart jEdit would be much better than
> having jEdit appear to lock up.

Take and unpack the text file from patch 3527797
http://sourceforge.net/tracker/download.php?group_id=588&atid=300588&file_id=443849&aid=3527797

Search for regular expression:
(.),
and replace all with beanshell snippet:
_0 + " "

that should do the trick.

> 
> Dale
> 
> PS I read the stackoverflow thread. It looks like one guy says don't
> do it and quite a few others say you can, but be careful. What I got
> out of that was a definite "maybe".
> 
> 
> On Tue, Jul 24, 2012 at 12:41 PM, Jarek Czekalski
> <jar...@po...> wrote:
>         W dniu 07/23/2012 10:46 AM, Thomas Meyer pisze:
>         > Am Sonntag, den 22.07.2012, 21:22 +0200 schrieb Jarek
>         Czekalski:
>         >> W dniu 07/16/2012 09:14 PM, SourceForge.net pisze:
>         >>> Patches item #3544730, was opened at 2012-07-16 12:14
>         >>> Submitted By: Thomas Meyer (thomasmey)
>         >>>
>         >>> Initial Comment:
>         >>> OutOfMemory Throwable is an Error. Search&Replace only
>         catches Exceptions. Fix it.
>         >> I don't think we should catch out of memory error. After
>         that "there is
>         >> no way to get a defined behaviour". I agree with the
>         opinion taken from
>         >> stackoverflow.
>         >>
>         >>
>         http://stackoverflow.com/questions/1692230/is-it-possible-to-catch-out-of-memory-exception-in-java
>         >>
>         >> What do you think?
>         > I think it's okay, to catch the OOM Error in the
>         searc&replace (all)
>         > case. It gives a nice user experience, without it jEdit
>         seems tp hang.
>         
>         
>         Did you read the stackoverflow thread? The argument about nice
>         user
>         experience seems a bit amusing compared to serious analyze
>         there. Try to
>         prove they are wrong and that it's absolutely safe to catch
>         OOM in the
>         code the way you do. Only this way you may convince me.
>         
>         > Besides that there are already two places where an OOM Error
>         is catched
>         > in jEdit:
>         >   BufferLoadRequest
>         >   JEditKillRing
>         
>         
>         Maybe they should be reconsidered too. I don't think they are
>         examples
>         of proper oom handling.
>         
>         Jarek
>         
>         
>         ------------------------------------------------------------------------------
>         Live Security Virtual Conference
>         Exclusive live event will cover all the ways today's security
>         and
>         threat landscape has changed and how IT managers can respond.
>         Discussions
>         will include endpoint security, mobile security and the latest
>         in malware
>         threats.
>         http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>         --
>         -----------------------------------------------
>         jEdit Developers' List
>         jEd...@li...
>         https://lists.sourceforge.net/lists/listinfo/jedit-devel
>         
>