[Paraview] Can a collaborative client share a pipeline without creating a view itself?

[Peter Debuchev]

Dear Utkarsh,
yes, for the time being, I am trying to base my project only on VTK, which
means reinventing lots of ParaView's architecture. At least, I get
*something* running, although progress is slow. But my hope is to switch
back to ParaView once I understand it better and when some bugs are fixed.

The objective is nothing more than what "collaboration" promises: do
collaborative visualization. Without the restriction of a master and slaves
(at least from the user view, although programmatically there may be one
designated master client).

If collaboration is experimental in ParaView, I'd say so in the very
beginning of any article documenting or blogging about those functions.
Still, in an experimental feature, there should be suffiecient interest to
fix bugs, or "unsupported" would be a better term.

One problem I see with the (free, of course) help in this mailing list is
that one never knows whether someone is still considering the problems
adressed in the list after a week or so. Often I see problems half
solved,but the remaining - hard - problem is then silently ignored. I
assume that you communicate internally who is assigned which question,
given that you consider them worthwile at all, so that not two people work
on the same. Just a suggestion, could you send a short "I'll look into
this", and, more importantly "Sorry, we can't solve this problem"? It is
much better to know that one has to go another path than sitting and wait
and maybe ask again and again because nobody dares to say that there is no
solution for a problem that seemed easy in the beginning.

Peter


2016-07-13 15:56 GMT+02:00 Utkarsh Ayachit <utkarsh.ayachit at kitware.com>:

> Probably not, but then I assumed you had moved on. Are you still doing
> this? Can you go back to your original use-case? What is your objective
> with this? Maybe there are alternative approaches than using the
> "collaboration" support, which frankly is rarely used and hence I'd
> consider experimental.
>
> On Tue, Jul 12, 2016 at 1:34 AM, Peter Debuchev <peterdebuchev at gmail.com>
> wrote:
>
>> Dear Utkarsh,
>> is there any chance that this issue is looked at during the upcoming
>> hackathon?
>>
>> Peter
>>
>> 2016-06-28 17:46 GMT+02:00 Peter Debuchev <peterdebuchev at gmail.com>:
>>
>>> Dear Utkarsh,
>>> I tried to reproduce the error as described in my previous email, but
>>> haven't been able to until now, don't know what happend. The attached code
>>> attaches to a pvserver at localhost:11111 and adds a sphere every half
>>> second. Unfortunately it crashes as soon as I connect ParaView to the same
>>> server.
>>> I guess I'll have to give up programming ParaView and work the old way,
>>> writing VTK files and loading them in ParaView. It is way too complicated.
>>>
>>> Peter
>>>
>>> 2016-06-24 7:08 GMT+02:00 Peter Debuchev <peterdebuchev at gmail.com>:
>>>
>>>> Thank you for the patch. There is still something missing. If I modify
>>>> the custom client such that it adds an object every half second or so, then
>>>> first start this clients and then ParaView after a few seconds, ParaView
>>>> misses one of the objects. ParaView finds those objects that are added
>>>> before ParaView starts and those that are added later. But during ParaView
>>>> startup one of the objects is lost.
>>>> I'll see if I can make this reproducible and send a code example in the
>>>> next days.
>>>>
>>>> Peter
>>>>
>>>> 2016-06-22 17:25 GMT+02:00 Utkarsh Ayachit <utkarsh.ayachit at kitware.com
>>>> >:
>>>>
>>>>> > can you give a quick update whether the suggested bug fix is likely
>>>>> to be
>>>>> > solved:
>>>>>
>>>>> Here you go:
>>>>> https://gitlab.kitware.com/paraview/paraview/merge_requests/844
>>>>>
>>>>> Once the dashboards come back clean, this will get merged. If do get
>>>>> around to testing it, I'd suggest getting involved and giving the
>>>>> merge-request a +1 ;).
>>>>>
>>>>> Utkarsh
>>>>>
>>>>
>>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview/attachments/20160714/f5d4334b/attachment.html>