[chuck-users] some bugs on windows ChucK

Spencer Salazar spencer at ccrma.stanford.edu
Thu Sep 13 17:38:07 EDT 2012


Hey Graham,

Regarding your second issue, what often occurs in test6.ck is that the
OSC message is sent and received before the listenEvent() shred is
able to execute. Thus, it is processed by the OscRecv, but since an
OscEvent matching that address ("/twoStrings,ss") has not yet been
created, it silently ignores the message.

You can resolve this by putting a me.yield(); after sporking
listenerEvent(). In that case, the message should be received every
time. But in my experience the scenario in which that is necessary is
rare.

spencer


On Thu, Sep 13, 2012 at 4:17 PM, Graham Coleman <ravelite at gmail.com> wrote:
> Hello Ge,
>
> 1. Seems to work! Nicely done and thanks for the fix.
>
> 2. On this machine test6.ck never seems to receive the message, but r.ck and
> s.ck seem to work when run separately on the command line (from different
> VMs). Do you think there would be a problem in particular from sending and
> receiving OSC from the same VM? Is there a problem with the usage in
> test6.ck?
>
> But it doesn't crash.
>
> also rock on!
>
> Graham
>
> On Thu, Sep 13, 2012 at 3:00 PM, Ge Wang <ge at ccrma.stanford.edu> wrote:
>>
>> Hi Graham!
>>
>> Thanks for your contribution and insight -- we have hopefully addressed
>> both of the issues you outlined in the latest release.
>> Let us know if something still doesn't seem right?
>>
>> Thanks and rock on!
>>
>> Ge!
>>
>>
>> On Wed, 12 Sep 2012, Ge Wang wrote:
>>
>>> Hi Graham!
>>>
>>> Greetings old friend!  Thanks for the excellent tests and yummy
>>> breadcrumbs! There is definitely something funky with OSC since 1.3, on
>>> 64-bit and 32-bit command chuck on OSX (1.3.1.0), I don't get the same crash
>>> as you did on windows 7, but getting memory errors on cleanup.
>>>
>>> Looking into it now!
>>>
>>> Ge!
>>>
>>> On Tue, 11 Sep 2012, Graham Coleman wrote:
>>>
>>>> Greetings,
>>>>
>>>> First, congratulations to chuck team and all responsible for the recent
>>>> releases.
>>>>
>>>> I found a few bugs, these were tested in the latest ChucK and
>>>> miniAudicle
>>>> builds (win32) on 64-bit Windows 7 machines.
>>>>
>>>> 1. Machine.add doesn't seem to support drive specifications in paths.
>>>> (test1.ck)
>>>>
>>>> <<< Machine.add( "C:/Users/Public/test2.ck" ) >>>; //doesn't work
>>>>
>>>> <<< Machine.add( "/Users/Public/test2.ck" ) >>>; //works
>>>>
>>>> 2. A crashing bug when dealing with string non-literals. (test6.ck)
>>>>
>>>> I was unable to reproduce it outside of an OSC context, I don't know if
>>>> it
>>>> affects non-Windows systems.
>>>>
>>>> Basically, I was capturing string arguments from received OSC messages,
>>>> and
>>>> I found that one usage worked fine:
>>>>
>>>> <<< tsEvent.getString() >>>; //this should work
>>>>
>>>> but that trying to store it into a string object was crashing:
>>>>
>>>> tsEvent.getString() => string str2; //this should crash
>>>>         <<< str2 >>>;
>>>>
>>>> regards,
>>>>
>>>> Graham
>>>>
>>
>> _______________________________________________
>> chuck-users mailing list
>> chuck-users at lists.cs.princeton.edu
>> https://lists.cs.princeton.edu/mailman/listinfo/chuck-users
>>
>
>
> _______________________________________________
> chuck-users mailing list
> chuck-users at lists.cs.princeton.edu
> https://lists.cs.princeton.edu/mailman/listinfo/chuck-users
>


More information about the chuck-users mailing list