Re: [Chicken-users] utf8 and string-ref performance

[Felix]

From: Peter Bex <address@hidden>
Subject: Re: [Chicken-users] utf8 and string-ref performance
Date: Thu, 25 Nov 2010 12:17:24 +0100

> On Thu, Nov 25, 2010 at 06:15:50AM -0500, Felix wrote:
>> >> Sorry, I don't understand? They are not invalidated by GC (in case
>> >> you mean that).
>> > 
>> > So that means the locative will stay around while the string itself
>> > might have been collected?
>> 
>> If it's a weak locative, yes. But dereferencing will signal an
>> error in this case. But I thought this is about being able to
>> point into a string?
> 
> Yeah, but it's stored in a closure and the GC should be prevented from
> cleaning up the original string which may not be referenced anywhere
> anymore I guess.

It's a separate object and a non-weak locative will prevent the data
pointed to from being GC'd. I still don't know whether it is of any
help, though.


cheers,
felix