On Tue, Feb 10, 2009 at 8:39 AM, Matthew Toseland
<toad at amphibian.dyndns.org> wrote:
> On Thursday 05 February 2009 00:26, Daniel Cheng wrote:
>> On Thu, Feb 5, 2009 at 3:21 AM, <toad at freenetproject.org> wrote:
>> > Author: toad
>> > Date: 2009-02-04 19:21:34 +0000 (Wed, 04 Feb 2009)
>> > New Revision: 25554
>> >
>> > Modified:
>> >
> branches/db4o/freenet/src/freenet/client/async/SplitFileInserterSegment.java
>> > Log:
>> > Doh! Fix isEmpty
>> >
>> >
>> > Modified:
> branches/db4o/freenet/src/freenet/client/async/SplitFileInserterSegment.java
>> > ===================================================================
>> > ---
> branches/db4o/freenet/src/freenet/client/async/SplitFileInserterSegment.java
> 2009-02-04 19:10:02 UTC (rev 25553)
>> > +++
> branches/db4o/freenet/src/freenet/client/async/SplitFileInserterSegment.java
> 2009-02-04 19:21:34 UTC (rev 25554)
>> > @@ -1355,6 +1355,9 @@
>> > }
>> >
>> > public synchronized boolean isEmpty(ObjectContainer container) {
>> > - return (finished || blocks.isEmpty());
>> > + if(persistent) container.activate(blocks, 2);
>> > + boolean ret = (finished || blocks.isEmpty());
>> > + if(persistent) container.deactivate(blocks, 1);
>>
>> object deactivated as a side effect of isEmpty() !
>> this is quite unexpected and may be very hard to debug later..
>
> Eh? It's an internal object, we do that all the time.
I am saying "it don't work",
I am saying "it is nasty to debug if we forget this some months later"
>>
>> > + return ret;
>> > }
>> > }
>
> _______________________________________________
> Devl mailing list
> Devl at freenetproject.org
> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
>
