[Opensim-dev] Major issue with core code.

Michael Cerquoni nebadon2025 at gmail.com
Sun Mar 27 20:28:45 UTC 2011


Well I took your suggestion and added in the scanning loop, and its still
not working correctly, its reporting the wrong prim # atleast compared to
what the viewer is seeing.  here is a screenshot

http://www.onikenkon.com/screenshots/link_order_issue_03.png

So it seems that this method will also not work..

On Sun, Mar 27, 2011 at 8:02 AM, Wade Schuette <wade.schuette at gmail.com>wrote:

> Here's an easy to understand snippet of code that does the job, regardless
> how inelegant.
> ====
>
> // The function is to find out what prim numbers correspond to those names
> right now (in case it has changed).
> // Author :  Wade Schuette;   Date: March 27, 2011;
> // Anyone is free to copy modify transer or whatever this code (just take
> off my name if you do!)
>
> // Note: I'm sure there are vastly more elegant ways to keep all the
> specific names out of the routines,
> //  and put them into lists at the start, but my goal was just to
> illustrate a very simple way that works
> //  and can be read by a new scripter.
>
> // global variables set as a side-effect of the getPrimNums() call;
> // My example  linked set is a body with various parts.  This code is run
> from the root;
>
> integer heart;
> integer liver;
> integer stomach;
>
> // subroutine to find which prim corresponds to known names, by brute force
> exhaustive search.
> getPrimNums()
> {
>         integer i;
>         integer numPrims = llGetNumberOfPrims();
>         llSay(0, "This link set has " + (string) numPrims + " prims.");
>         for (i=1; i <= numPrims; i++)
>         {
>             llSay(0,  "Prim " + (string) i + " is " + llGetLinkName(i) );
>             if  (llGetLinkName(i) == "heart") { heart = i;}
>             else  if  (llGetLinkName(i) == "liver") { liver = i;}
>             else  if  (llGetLinkName(i) == "stomach") { stomach = i;}
>         }
> }
>
>
> default
> {
>     state_entry()
>     {
>
>         getPrimNums();
>
>         // OK, those are assigned and the integers can be used in your
> code.
>         llSay(0, "Please confirm from the list above...");
>         llSay(0, "Heart is prim number " + (string) heart);
>         llSay(0, "Stomach is prim number " + (string) stomach);
>         llSay(0, "Liver is prim number " + (string)  liver);
>
>     }
> }
>
> ====
>
>
> On Sun, Mar 27, 2011 at 9:16 AM, Marck <marck00 at nexgo.de> wrote:
>
>> The prim numbering used to change even when an object crossed a region
>> border. This was fixed some time ago.
>>
>> For an example how to circumvent this issue in a script, see
>> http://forums.osgrid.org/viewtopic.php?p=10732#p10732. A more generalized
>> solution of this can be found in thread "Single Script Color/Linkset
>> Changer"; in both cases, the relevant function is named "getLinkNums". The
>> basic idea is to put a label either in the linked prim's name or description
>> field and then retrieve the proper prim number by searching for the
>> label(s).
>>
>> Michael Cerquoni wrote:
>>
>>> Any chance you have an example of how you do this that you can share? I
>>> would like to see how your doing this.
>>>
>>> On Sun, Mar 27, 2011 at 1:41 AM, Wade Schuette <wade.schuette at gmail.com
>>> <mailto:wade.schuette at gmail.com>> wrote:
>>>
>>>    Just a note - While I agree entirely that the prim order should stay
>>>    constant for an object the user hasn't changed, as an application
>>>    developer I always start by polling  all linked prims to determine
>>>    which one is which number (today), and then work off that list.
>>>    Otherwise,  adding a single new prim to the object, or unlinking it
>>>    in order to resize or retexture a prim that refuses to change when
>>>    linked, will totally break the user script, and is maddening.
>>>    With polling, I can unlink or relink or add pieces whenever I feel
>>>    like it and the code doesn't break.   The one extra step only has to
>>>    be written and debugged once and used as a utility subroutine after
>>>    that.   What WOULD kill it is if the root prim changed, of course.
>>>
>>>    And,  I developed that habit in Second Life, although it's totally
>>>    crucial in OpenSim since we have to keep unlinking complex objects
>>>    in order to change a stubborn linked prim that refuses to change
>>>    while it is linked.  Why is THAT, by the way?  Can THAT be fixed?
>>>
>>>    Wade
>>>
>>>
>>>
>>>    On Sun, Mar 27, 2011 at 4:01 AM, Michael Cerquoni
>>>    <nebadon2025 at gmail.com <mailto:nebadon2025 at gmail.com>> wrote:
>>>
>>>        I spoke to Melanie, she says she has a fix for this! apparently
>>>        this has been an issue for a while, I had no idea, thanks
>>>        Melanie :D
>>>
>>>        On Sat, Mar 26, 2011 at 11:44 PM, Michael Cerquoni
>>>        <nebadon2025 at gmail.com <mailto:nebadon2025 at gmail.com>> wrote:
>>>
>>>            I have just stumbled across a major problem with the core
>>>            code, it seems that every time we rez an object its linkset
>>>            prim order is changing.  This makes it impossible to script
>>>            things that call upon a certain # in the link-set.  I have
>>>            filed a mantis:
>>>
>>>            http://opensimulator.org/mantis/view.php?id=5421
>>>
>>>            see screenshots of issue here:
>>>
>>>            http://www.onikenkon.com/screenshots/link_order_issue_01.png
>>>            http://www.onikenkon.com/screenshots/link_order_issue_01.png
>>>
>>>            This is a major issue that should be resolved before 0.7.1
>>>            is tagged i believe.
>>>
>>>            --
>>>            Michael Emory Cerquoni - Nebadon Izumi @ http://osgrid.org
>>>
>> _______________________________________________
>> Opensim-dev mailing list
>> Opensim-dev at lists.berlios.de
>> https://lists.berlios.de/mailman/listinfo/opensim-dev
>>
>
>
>
> --
> R. Wade Schuette, CDP, MBA, MPH
> 6751 Scio Church Rd.
> Ann Arbor, MI 48103-9270
> cell: 1 (734) 635-0508
> fax:  1 (734) 864-0318
> wade.schuette at gmail.com
>
> _______________________________________________
> Opensim-dev mailing list
> Opensim-dev at lists.berlios.de
> https://lists.berlios.de/mailman/listinfo/opensim-dev
>
>


-- 
Michael Emory Cerquoni - Nebadon Izumi @ http://osgrid.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://opensimulator.org/pipermail/opensim-dev/attachments/20110327/3d827424/attachment-0001.html>


More information about the Opensim-dev mailing list