I don't have a suggested solution. I frankly don't think this is an RFE
that I think is a particularly useful addition.
Rick
On Sun, Nov 18, 2012 at 6:55 AM, Rony G. Flatscher <
[email protected]> wrote:
>
> On 18.11.2012 12:51, Rick McGuire wrote:
>
> No, it would not. This assumes that the memory for the instance structure
> has not been reused by the interpreter after the termination. Since it is
> just an object instance, this would not be the case. Forcing these objects
> to remain referenced would create an undesirable memory leak.
>
> Thanks!
>
> What do you suggest/advise as an approach, if you think I could do it
> (time permitting for the needed research and testing)?
>
> ---rony
>
>
>
>
> On Sun, Nov 18, 2012 at 6:47 AM, Rony G. Flatscher <
> [email protected]> wrote:
>
>> In order to remove a crash in ooRexx when using ooRexx from OpenOffice
>> (AOO) via BSF4ooRexx, I am contemplating to implement RFE
>> <http://sourceforge.net/p/oorexx/feature-requests/356/><http://sourceforge.net/p/oorexx/feature-requests/356/>
>> .
>>
>> However, short of having the needed overview about how the pieces fit
>> together, a RFH (request for hints/help) of which parts of the interpreter
>> need to be taken care of.
>>
>> My initial (unsubstantiated) thoughts go along the lines:
>>
>> - it is sufficient to become able to learn whether a Rexx interpreter
>> instance is in the process or has terminated, i.e. whether a
>> context->Terminate() has been issued for that Rexx interpreter instance,
>>
>> - This assumes that if the Rexx interpreter needs to interrupt and
>> end a Rexx program, e.g. because of a syntax condition that does not
>> get
>> trapped, the Rexx interpreter will use Terminate() on that Rexx
>> interpreter
>> instance. If the Rexx interpreter ends a Rexx program in error
>> differently,
>> then it would be important to learn somehow that no Rexx interpreter
>> instances are usable anymore.
>>
>> - The idea would be to add boolean field named isTerminated to
>> the context structure, which initially is set to false.
>> context->Terminate() then would change that field to true.
>>
>> - This way, native code will become able to learn whether the Rexx
>> interpreter instance is still usable by merely inspecting that field
>> in the
>> context structure.
>>
>> - In the AOO use case the native code in BSF4ooRexx would then
>> become able to learn whether a callback into the Rexx interpreter
>> instance
>> would be safe or not, and react accordingly (e.g. raising an
>> appropriate
>> Java exception indicating the correct cause), such that no crash of
>> ooRexx
>> needs to occur anymore.
>>
>> Would that yield a feasible, sensible solution?
>>
>> ---rony
>>
>>
>
>
> ------------------------------------------------------------------------------
> Monitor your physical, virtual and cloud infrastructure from a single
> web console. Get in-depth insight into apps, servers, databases, vmware,
> SAP, cloud infrastructure, etc. Download 30-day Free Trial.
> Pricing starts from $795 for 25 servers or applications!
> http://p.sf.net/sfu/zoho_dev2dev_nov
> _______________________________________________
> Oorexx-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/oorexx-devel
>
>
------------------------------------------------------------------------------
Monitor your physical, virtual and cloud infrastructure from a single
web console. Get in-depth insight into apps, servers, databases, vmware,
SAP, cloud infrastructure, etc. Download 30-day Free Trial.
Pricing starts from $795 for 25 servers or applications!
http://p.sf.net/sfu/zoho_dev2dev_nov
_______________________________________________
Oorexx-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/oorexx-devel
